News

www.eenewseurope.com February 2018

electronics europe News

V2X Communications

LTE vs DSRC

european
business press
 CONTENTS FEBRUARY 2018

Dear readers, www.een

ewseurop
e.com

News
As researchers want to give more autonomy to artificial intelligence, which in turn
could give more autonomy to self-driving cars, while startups want to give more February
2018

autonomy to wireless devices (through wireless charging), autonomy takes electron

ics euro
pe News
different meanings yet it sounds like a new buzz word, implying calculated
intelligence for resilience. You’ll find many examples of that in our February
edition of eeNews Europe. V2X Com
mun ications
LTE vs D
On the electronics and memory front, the race for miniaturization doesn’t stop SRC
with Moore’s law, moving to monolayer 2D materials and device concepts is
what scientists believe will boost overall device densities far beyond the usual
process node shrinking scenario. This is exemplified with the “atomristor”, an
atomically-thin memristor

Month after month, year after year, discover how technology matures and
moves from the lab to the industrialization stage.
Enjoy this edition of eeNews Europe and don’t hesitate to get in touch to
tell us your stories, from starting up a company to developing your first pro-
totypes, achieving your first product design-in, sharing your insight on this
fast-paced industry or to contribute your expertise on some of the topics europe
business an
press

we’ll be covering along the year.

EW18-Cove
r.indd 1

02.01.18
14:18

Julien Happich
Editor in Chief

4 - 49: OPINION 32 - 37: WIRELESS NETWORKING

Uncommon Market: Haptic knob Why would the US consider
leverages magnetorheological fluids nationalising 5G?
for crisp force-feedback Axios is reporting that the Trump
administration is considering
Last word: Accelerating your quality nationalising the 5G network in
system’s actionability in response to perceived threats
manufacturing from China and the advances it
has made in 5G and AI. To back up its claims, Axios
6 - 27: NEWS & TECHNOLOGY obtained a powerpoint presentation and memo pro-
duced by a senior National Security Council official.
The upside to Apple’s halving
of iPhone X production 38 - 42: PROGRAMMING & DEBUGGING
Apple Inc. will slow production of its TOOLS
flagship iPhone – the $999 iPhone Model-level debugging
X – by 50 percent in the first three A “model-level” debugger
months of 2018 because of a lack can provide a side-by-side
of demand, according to a Nikkei report. And if true that has view of a simulation model
both negative and positive implications. of a control system with the
source code produced by an
28 - 31: 3D INTEGRATION & IC PACKAGING automatic code generator.

Atomically-thin memristors could yield 46: READER OFFER

ultra-high density memory This month, Lime
Applying 2D materials to the Microsystems is giving away
memristor concept, researchers have five Grove starter kits for
demonstrated the scalability of non- LimeSDR Mini, worth 200
volatile resistance switching down to Euros each, for eeNews
atomically-thin devices. Once appropri- Europe’s readers to win.
ately stacked, the “atomristors”, as they
describe the devices, could yield memristor densities in the
range of 1015/mm3, translating to a theoretical areal density 48: DISTRIBUTION CORNER
of 6.4Tbit/in2 for single-bit single-level memory storage.

www.eenewseurope.com eeNews Europe February 2018 News 3

 UNCOMMON MARKET HAPTICS

Haptic knob leverages magnetorheological

fluids for crisp force-feedback
By Julien Happich

D
eveloped by List, a research institute of CEA Tech or on the contrary you could create inertial effects with the knob
focused on smart digital systems, the MATISS (Mag- still turning as you release it” Hafez continued.
neto-Textural Inertial Spinning System) is a novel type Next, the idea is to develop a library of haptic effects through
of rotary knob that integrates a highly programmable passive that proprietary haptic interface, whose parameters customer
brake together with a small electric motor to allow for many could modify to customize these effects or program different
types of haptic effects including force-feedback, elastic force stop and click patterns, even perceivable textures.
and coil rebound. How about joysticks then?
In particular, the passive brake consists of a magnetorheo- “The real novelty was that we were able to tightly encapsu-
logical fluid flowing between intertwined bell-shaped walls, late the magnetorheological fluid between multiple imbricated
whose viscosity can be tuned arbitrarily by
controlling its phase transition from liquid to solid
(and vice versa) by a magnetic field (from a built-
in coil). While the liquid phase offers no resistance
to the knob’s rotation, the solid phase ensures a
solid stop.
The prototype originated as follow up work
from prof. Carlos Rossa’s thesis “A hybrid actua-
tion system for haptic interfaces”, which already
hinted at the device’s construction with fluid gaps
arranged between dome-shaped enclosures.
Interviewed by eeNews Europe, Moustapha
Hafez, head of the Sensory and Ambient Interfac-
es Laboratory at CEA List gave us more technical
details about the device.
“Because we can control the passive brake
electronically, we can create virtual hard stops,
to simulate a collision or provide progressive
resistance, and anything in between. Due to the
very nature of the magnetorheological fluid we
use, there is hardly any friction in the off-state,
something in the range of 2mN/m. But if we ap-
ply 1.5A to the coil, the magnetic field turns the
fluid solid and this locks the knob with a force of
2N/m. That’s three orders of magnitude between the “on” and bell-shaped walls, an innovative passive brake architecture that
“off” force couples” Hafez explained. yields very strong braking forces. The rotating knob is a one-
But how fast is the knob’s response time? We asked. axis implementation, but we are also developing 2-axis joysticks
“The phase transition occurs in milliseconds, and in this constructed with one module per degree of freedom, mechani-
implementation, the coil’s inductance and the time it takes to cally enclosed one in another”, Hafez said. “Control sensitivity
create the magnetic field is the limiting factor, between 5 and can be tuned through the signal amplification chain. So with
10ms. That still means we can create very fast and closely- the same basic modules, we can make a high accuracy joystick
spaced clicking sensations, at frequencies in the 100Hz range. that a surgeon would control with three fingers to operate a sur-
The frequency spectrum we can perceive as humans ranges gical system, or we could create heavy machinery and gaming
from a few Hertz to hundreds of Hertz, with a peak of sensitivity joysticks with real hard stops and a more reactive force”.
at 250Hz. Hence we have plenty of room to emulate vibrations “We are currently developing such joysticks for a partner we
and even textures just by controlling the passive brake” the cannot disclose at the moment, but this may be another dem-
researcher told eeNews Europe. onstration at CES 2019”, the scientist concluded.
“In our CES demo, opening a safe, we coupled the haptic In addition to gaming and the applications demonstrated at
effects with audio rendering of the clicks. By fusing the two and CES, applications for MATTIS technology range from transpor-
stimulating two sensory modalities, we really augmented the tation to construction to manufacturing. Some potential applica-
overall haptic rendering”. tions include driving or operating assistance for cars, buses,
“What’s more, the knob also integrates a motor, which, com- trucks, agricultural vehicles, construction machinery, planes,
bined with the passive brake, can be used as a reverse force helicopters and submarines. Such haptic joysticks could be
input to create an elastic torque, or a spring effect imparting a used to better control drones, allowing the operator to feel the
notched sensation to the clicks. The motor can also be driven to effects of wind and proximity to obstacles, but also for a variety
create static vibrations, or to reset the knob to its initial position of remotely controlled robots.

4 News February 2018 @eeNewsEurope www.eenewseurope.com

 NEWS & TECHNOLOGY MARKET ANALYSIS

The upside to Apple’s halving

of iPhone X production
By Peter Clarke

A
pple Inc. will slow production of its flagship iPhone – the X in the holiday season in Europe, the US and China and that
$999 iPhone X – by 50 percent in the first three months customers are opting to keep their mobile phones for longer.
of 2018 because of a lack of demand, according to a This implies that the additional features being brought into
Nikkei report. And if true that has both negative and positive the latest smartphones, such as to-the-edge displays and
implications. facial recognition, are not considered worth the premium price
It could be an indicator of saturation in the global smart- Apple was trying to establish. Of course, having milked the
phone market. This has already occurred in the personal and early-adopting technophiles, it may be enough for Apple to trim
tablet computer markets and perhaps the smartphone’s time the price back to a more reasonable level to re-ignite demand.
is coming in 2018. It could be an indicator that a slowdown in Indeed that is Apple’s standard practice. It does appear to
Moore’s Law is being reflected in features that are not suffi- have over-estimated its initial manufacturing, but better to have
ciently attractive to command a premium price. Or it could be iPhones warehoused than to be unable to meet demand and
that Apple just got its pricing wrong. But there are also other see customers migrate to Samsung.
possibilities that speak to issues of consumerism versus eco- But some observers are seeing the iPhone X’s relative flop as
logical sustainability. an indicator of global smartphone saturation and that the kudos
First quarter production of the iPhone X could now be about that attached to a particular brand and model is going away…or
20 million units the Nikkei report said, quoting a note written by at least going away from the smartphone to move on to some
JP Morgan analysts. Given the significance of Apple as a buyer other product as a market driver.
of semiconductors – it takes about 10 percent of global output That was certainly the tone of the senior executives of Tai-
– this could have a big impact on slowing what was booming wan Semiconductor Manufacturing Co. Ltd. who recently spoke
growth in the chip market in 2017. about a series of market drivers including the PC, the mobile
The current thinking is that Apple will hit sales and profits ex- phone and the smartphone and the idea that cryptocurrency
pectations in its upcoming 1QFY18 financial results announce- and IoT could be next.
ment on Thursday February 1, but that the iPhone X slowdown Another view is that the two-year cycle of consumer elec-
could impact the forecast for 2QFY18. tronics replacement was largely driven by the two-year cycle of
For some people any reduction in the recent feverish mood Moore’s Law, made manifest in products by hardware, software
of the chip market is a bad thing. Clearly those in the fabbed and service providers because they knew what was good for
and fabless chip companies like plenty of demand their shareholders – if not for global ecology and sustainability.
and the chance to push Perhaps as Moore’s Law falters and computation
prices upwards. transitions from deterministic architectures to a mix
And of course the of von Neumann and neuromorphic computing, the
iPhone X situa- benefits of the two-year upgrade are less apparent.
tion is particularly There may also be a growing awareness amongst
relevant to Apple consumers that replacing a phone every two years
suppliers such – or even more frequently – drives enormous profits
as STMicroelec- for the consumer electronics giants while making
tronics, Dialog little difference to their own lives, except the culti-
and AMS. But for vation of near-focus eye-strain and social-media
buyers of chips driven anxiety.
a slowing of the It is to be hoped that in the coming era of voice-
chip market will command, computer vision
likely come as a and sensor
relief. The prices and neural
of those hard-to- networks
order memory and these
logic ICs and lead tech-
times could be nologies
about to reduce as can be
demand comes in used to
to closer balance create
with supply. genuine
Nikkei reported com-
that there are mon
indicators of lower wealth.
than expected
sales of the iPhone

6 News February 2018 @eeNewsEurope www.eenewseurope.com

 POWER SEMICONDUCTORS

Dynex walks the line with foundry deal

By Nick Flaherty

I
t is accepted wisdom in the electronics business that you the deal seems to be to offer up the extra capacity to foundry
don’t compete with your customer. That has been particu- customers. Then the foundry parts will have a different design
larly true in the semiconductor business and foundries, and to the mainstream Dynex parts that CRRC uses, so they are not
sparked the growth of the ‘pure play’ foundries such as TSMC actually competing.
and Global Foundries that now dominate the industry. Their This leaves Dynex stuck in the middle. It can try to expand
success has come from a focus on the process technology and sales of its standard products, thereby selling to competitors of
offering that to a wide range of customers. What they don’t do CRRC, or offer up its own process technology to its own com-
is make their own products. petitors via the foundry. That’s a tough line to walk. If prospects
That makes the announcement of Dynex in offering a foundry improve and CRRC takes more product there’s less capacity
service as worrying. The company already makes a range of for foundry customers. When the downturn comes (and Dynex
IGBT power devices, and now has a deal to supply a ‘leading management say that’s the case now, in contrast to the rest
semiconductor. of the industry) there’s less product business and less foundry
That means that there is spare space in the fab, and that is business, a double hit. This approach has been tried many
not a good sign at the peak of the semiconductor cycle. Again times before with limited success.
the established
wisdom is to fill
up the fab with
your own prod-
ucts.
So does this
mean Dynex is
struggling to
be successful?
Although it hasn’t
been hugely prof-
itable in the last
couple of years
and revenues
have fallen (from
$46m to $40m,
although the
exchange rate
hasn’t helped), it
has grown at the
site from 351 to
368 people in the
first nine months
of 2017.

Now the power business tends to do things its own way. The
fabs are smaller than their CMOS cousins and are fully paid for
and running specialised processes that are optimised to par-
ticular types of devices. You can’t run standard CMOS in these
fabs, and you can only run a limited number of different styles
of power device without some hefty, and expensive, process
development.
All this limits the options for foundry customers to differenti-
ate their devices from those of Dynex.
Dynex is a Canadian company with a fab in Lincoln, UK, that
carries 59 years of British heritage through Marconi MEDL and
GEC Plessey Semiconductors. Since 2008 Dynex has been part
of a Chinese conglomerate.
That conglomerate, Shanghai-based CRRC Times Electric, is
the key.
It buys a substantial part of the output of the Lincoln fab and
so as well as the owner is the major customer. If CRRC isn’t
able to take more product, which is a surprise given the boom
in infrastructure modernisation across its home market, then
So the new boss at the foundry division, Mark Kempton,
Business Unit Director for Semiconductor Devices, has his work
cut out walking that line, but is upbeat of course.
“The formation of the Dynex Foundry Services business is
a logical expansion of the company’s commercial product and
service offerings,” he said. “The receipt of this first external or-
der for high voltage IGBT die builds on decades of high-quality
semiconductor fabrication experience at Dynex. The excellent
chip performance, confirmed by the customer’s own testing,
serves to highlight the quality of our design and manufacturing.
We expect this business to grow into a substantial part of our
operations over the coming years as we expand our customer
base.”
There’s no doubt about the quality and design expertise, but
the history of the hybrid product/foundry business model does
not bode well.

www.eenewseurope.com eeNews Europe February 2018 News 7

 NEWS & TECHNOLOGY CIRCULAR ECONOMY

Getting value from e-cycling

By Stephen Harding

B
etween 2009 and 2012, the European Union (EU) funded powder then moves on via a conveying system to another part
an aptly-titled project called the Recovery of Electronic of the plant for handling.
Waste through Advanced Recycling and Demonstration When this powder is being screened, the powder may ac-
(REWARD). Although the project has conclud-
ed, electronics recycling and recovery is an
ongoing task in many recycling plants.
Achieving the circular economy, or a so-
ciety in which nothing is wasted or disposed
of until its potential value has completely
depleted, seems to be one of the EU’s key
objectives in the twenty-first century. The EU
has introduced a number of schemes, initia-
tives and directives to promote the idea that
anything reusable be reused or recycled.

Of course, this isn’t an entirely new

concept. Businesses have long understood
the benefits of drawing the most value from
assets as possible. The only difference is that
now it’s become a more important issue for
wider society, with the competitive advantage
provided by asset utilisation replaced with
economic and environmental responsibility.
While it is often promoted from an environ-
mental perspective, there is a strong eco-
nomic benefit in recycling products such as
electronics.
In a typical electronic application, there will be hundreds of resources with little
As EU Commission environmental spokes-
reusable value and many components made of high value materials.
person Enrico Brivio says of electronics recy-
cling, “the more valuable our waste becomes due to pressures cumulate and block the mesh, a phenomenon known as screen
on our resources, the more we will need the technologies and blinding.
innovation to pump it back into the economy instead of burying This is not an uncommon occurrence and, from Gough Engi-
it or burning it”. neering’s experience in material processing, can be prevented
by using screens that include an ultrasonic vibratory motor.
Effective materials recovery is critical in ensuring that the
electronics recycling, or e-cycling, process is as effective as it However, it’s also vital that no valuable electronics or metals
can be. In a typical electronic application, there will be hundreds are mixed in with the foam ahead of the grinding process. To
of resources with little reusable value and many components ensure this, recycling plant managers must use a screening and
made of high value materials. Recycling plant managers will separating system that can effectively sort the shreds of poly-
naturally want to recover as many valuable materials as pos- urethane from the steel fragments or gold-containing circuitry.
sible with little waste mixed in.
Each recycling processing plant will require something differ-
This is only possible with effective screening and separation ent from its screening and separation systems, whether it’s due
equipment for recovery operations. In most applications where to the flammability of processed items or granularity. As such,
screening is necessary, there is a risk that screen blinding will it is vital that recycling plant managers consult with a materi-
impact throughput rates. als management specialist to determine the best equipment to
maximise the value of their operations.
For example, one stage in the recycling process of a stan-
dard refrigerator involves the unit itself being shredded down to The EU anticipates there to be more than 12 million tonnes of
separate insulation foam, electronic matter, metals, plastics and electronic waste across Europe each year by 2020. By investing
gases. From there, each component goes its separate way. in the right equipment to sort through waste and recover valu-
The foam, for instance, is ground into a powder to release able materials effectively and efficiently, recycling plant manag-
the gases contained within it and the resulting polyurethane ers can be rewarded with lower output and input costs, less
processing downtime and a greater contribution to Europe’s
Stephen Harding is Managing Director of recycling materials circular economy.
management specialist Gough Engineering -
www.goughengineering.com

8 News February 2018 @eeNewsEurope www.eenewseurope.com


Perovskite solar cell for tinted windows
By Nick Flaherty
R
esearchers at the University of California, Berkeley, have
created a cesium-doped perovskite solar cell that is
transparent at room temperature but turns dark at high
temperatures. This could lead to windows that automatically tint
on a sunny day to block the heat while also
generating electricity.
Peidong Yang, a professor of chemistry
and faculty scientist at Berkeley Lab, and his
colleagues have tweaked the chemical struc-
ture of perovskite so that the material turns
from transparent to opaque when heated and
also converts sunlight into electricity.
The thermochromic solar cell uses an in-
organic halide perovskite with added cesium,
lead, iodine and bromine has a relatively low efficiency of 7%
and the transition from transparent window to opaque solar cell
requires heating the window to the boiling point of water, but
Pay-as-you-go cloud-based verification
By Julien Happich
C
anadian startup Metrics Technologies
came out of stealth on the 1st of Febru-
ary 2018, presenting itself as a provider
of pay-by-minute Software as a Service (SaaS)
cloud-based verification solutions. With its first
product, the Metrics Cloud Simulator & Verifica-
tion Manager, the company wants to enable
semiconductor and systems companies to in-
stantly accommodate peak simulation demand, optimize budgets,
and accelerate verification closure.
The tool now commercially released is the fruit of nearly three
calendar years and 30+ person years of development, it has
already been successfully deployed at multiple companies, the
company says.
Metrics is delivering the benefits of a true cloud solution to SoC
design and verification, including on-demand deployment, high
reliability, and team collaboration. The company is also removing
the traditional high overhead for configuring and maintaining the
hardware and software needed for a robust verification environ-
ment. At the same time, it is addressing the historical obstacles
to broad cloud adoption, such as security concerns and workflow
integration.
“We founded Metrics to bring the compelling advantages of a
secure, collaborative, cloud-based workflow to SoC design and
verification teams,” said Doug Letcher, President and CEO of
Metrics and former VP of Engineering at verification services firm
XtremeEDA.
“Our solutions are built to plug-and-play into existing workflows
– to enable companies to achieve optimal utilization of software
licenses, compute hardware, and engineering staff time.”
Because the Metrics Cloud Simulator & Verification Manager
is a true cloud solution, it has the rapid elasticity to instantly scale
commensurate with variable demand. The simulation resources
are automatically adjusted up or down, dynamically by the minute,
without having to purchase additional hardware, licenses, or man-
age disk space. “Rapidly increasing demand for functional simula-
tion, and particularly the need to avoid the delays that occur during
www.eenewseurope.com
ENERGY HARVESTING
the team is already working on versions that work at lower tem-
peratures and with higher conversion efficiency. One advantage
is that the new material retains its conversion efficiency after
many cycles between transparent and a reddish tint.
“This class of inorganic halide perovskite
has amazing phase transition chemistry,”
Yang said. “It can essentially change from one
crystal structure to another when we slightly
change the temperature or introduce a little
water vapour.”
“The solar cell shows fully reversible per-
formance and excellent device stability over
repeated phase transition cycles without any
color fade or performance degradation,” said
Minliang Lai, a graduate student in Yang’s group. “With a device
like this, a building or car can harvest solar energy through the
smart photovoltaic window.”
peak demand, is driving companies to adopt
cloud-based verification,” explains Letcher.
Metric Cloud Simulator & Verification Man-
ager includes a complete, SystemVerilog IEEE
1800-2012 compliant simulator. The simulator
includes support for constraint solving, SVA
checking, functional coverage, and all standard
test bench and RTL constructs. UVM is thus
fully supported. Because it is a true cloud simulator, every autho-
rized team member can have immediate unlimited access even
during peak demand.
Customers are provided with a web-based interface, enabling a
365/24 access for productive collaboration around the world. The
cloud verification manager supports continuous integration to keep
code functional and teams productive. Its regression triage flow
helps debug test failures, with the ability to view and filter test his-
tory, including logs, waves, and code revision data. The tool sup-
ports a data-driven functional coverage closure via measurement &
optimization and provides visibility into a project at engineering and
management levels.
Plug-and-play in existing workflows, the cloud solution can
be deployed into existing design and verification workflows via
a securely hosted Git repository. Verification results can also be
downloaded for use with other verification tools. Companies can
specify appropriate role-based permission levels for different users
to launch simulation jobs and view pertinent verification results.
The data is fully encrypted during transmission and can be fully
encrypted at rest. Metrics also uses a high-reliability, redundant
architecture to protect all customer data.
The company uses the Google Cloud Platform (GCP) and is a
Google Strategic Technology Partner. Each Metrics customer gets
a dedicated cluster of machines within GCP, which is one of the
most secure and reliable compute environments available today,
Metrics says.
The pricing model is entirely pay-by-minute, at $.04/minute/
simulation job.
eeNews Europe February 2018 News 9
 NEWS & TECHNOLOGY COVER STORY

V2X Communications - LTE vs DSRC

Mark Patrick, Mouser Electronics

W
ith all the hype surrounding self-driving vehicles based
on artificial intelligence (AI), image recognition and so-
phisticated sensors, it is easy to forget about another
important interrelated technology that also promises to help
revolutionise driving. It is, in fact, destined to be a critical factor
in making autonomous vehicles a reality. Known by the acronym
V2X, it covers both vehicle-to-vehicle (V2V) and vehicle-to-infra-
structure (V2I) communications.
V2X refers broadly to a set of standards and technologies
that will allow vehicles to interact with public roads and other
road users - not necessarily by having to rely on use of cutting-
edge electronics hardware, but potentially via tried and tested
networking protocols and technologies. Despite its relative
simplicity, V2X makes a multitude of interesting applications
possible, including higher degrees of driving assistance, more
efficient road usage and collision avoidance.
For many years, the leaders of the V2X pack have been the
US-developed dedicated short-range communications (DSRC)
and the European cooperative intelligent transport systems
(C-ITS). The National Highway Traffic Safety Administration
(NHTSA) is currently finalising a proposal to make V2X technol-
ogy a mandatory feature of US vehicles, by perhaps as early
as the 2020 timeframe, with DSRC expected to be the solution
upon which it is based. However, while V2X has great potential,
adoption and finalisation of dedicated regional V2X standards
has been surprisingly sluggish, taking well over a decade
already. It has been so slow in fact, that the mobile communica-
tions industry has recently seized the opportunity to aggres-
sively push its own V2X standard, cellular-V2X (C-V2X).
In January 2017, when the NHTSA issued its proposal to
make V2X a legally-required component of all US vehicle de-
signs moving forward, it named DSRC as its preferred choice,
but went to considerable lengths to say that it would also look
favourably upon other technologies that could match or exceed
the capabilities of DSRC, while remaining backwards-compati-
ble with DSRC’s critical safety features. The long V2X marathon
has, as a result, suddenly turned into a sprint, and the finishing
line is now approaching rapidly. Has the former front-runner
delayed too long and will it end up losing out?
The Established Technologies: C-ITS
and DSRC
C-ITS and DSRC both use variants of the familiar IEEE 802.11
Wi-Fi standards as the physical and medium access layers of
their protocol stack. For DSRC this protocol layer is defined
by the IEEE 802.11p standard, while C-ITS uses the similar
ETSI-defined ITS-G5 standard, which is basically a version of
802.11p adapted to suit European regulations. While there are
significant differences in the higher-level functioning of C-ITS
and DSRC, both can run on broadly-compatible hardware and
reside in almost the same frequency range. Both these V2X
systems can communicate effectively between fast-moving ve-
hicles over a distance of at least 300m and at data rates ranging
from approximately 3-27Mbps.
In 1999, the US FCC allocated the 5.850-5.925GHz band for
high-priority road safety and traffic management applications,
and the European Commission has dedicated 5.875-5.905GHz
for similar purposes. IEEE 802.11p and ITS-G5 are designed to
work in these respective ranges. To make them better optimised
for the cluttered and fast-moving automotive environment, the
standards feature modi-
fications to typical IEEE
802.11 Wi-Fi frequency
usage which reduce the
impact of Doppler shifts
and multipath fading.

One of the key dif-

ferences between these
automotive protocols
and other members of
the 802.11 family is that
the initial handshake and
association period can be
reduced to a bare mini-
mum. The protocol can
send data almost immedi-
ately, and can also defer
authentication, encryption
and full identification to
higher level protocols.
Therefore, vehicles and
infrastructure can begin
exchanging essential
data on speed and posi-

10 News February 2018 @eeNewsEurope www.eenewseurope.com


tion within tens of milliseconds
of detecting each other. This is
obviously ideal for applications
such as collision avoidance,
and a stark difference from the
seconds-long handshaking nego-
tiation that are common in other
Wi-Fi variants.
While C-ITS and DSRC are
very similar at the MAC/PHY
access level, above that, they
diverge significantly at the net-
working and transport level. Both
support TCP/UDP over IPv6,
but for time-critical features they
each have their own special-
ised low-overhead messaging
standard. DSRC uses the wave
short message protocol (WSMP),
which is defined as part of IEEE 1609. However, C-ITS employs
a more ambitious multi-hop routing system, the basic transport
protocol (BTP) over the GeoNetworking service, as defined by
the ETSI EN 302 636 series of standards. GeoNetworking is a
geographically aware routing technique, which establishes an
ad-hoc network that is efficiently arranged according to the
physical locations of nodes.
C-V2X: The Mobile Industry’s Contender
C-V2X is the new name for the V2X standard that was initially
launched as LTE-V2X by respected international mobile com-
munications standards body the 3rd Generation Partnership
Project (3GPP), in its 3GPP Release 14. The change of name
reflects the fact that C-V2X now promises an upgrade path from
LTE to the much anticipated 5G mobile standard.
For device-to-device mode direct communications, C-V2X
uses the same 5.9GHz spectrum as DSRC and C-ITS. However,
C-V2X devices can also operate in traditional mobile bands to
communicate with the mobile network in so-called vehicle-to-
network (V2N) mode. C-V2X device-to-device communications
do not require a SIM card, a mobile subscription or any mobile
network infrastructure.
C-V2X is a much younger technology than either DSRC or
CITS, and is still under development, but early tests have shown
that at the data rates required for V2X applications C-V2X
reaches a range comparable to DSRC’s 300m at urban driving
speeds and achieves 20-30% better range at faster speeds.
C-V2X vs the Incumbents
Supporters of DSRC and C-ITS will argue that they are far more
suitable for V2X applications than C-V2X because they have
been carefully constructed from first principles for one clearly-
defined purpose, rather than being bolted on to an existing
standard. For example, DSRC and C-ITS are designed to offer
low networking overheads and low latency, they claim. Latency
is a key issue for V2X. Functions like collision avoidance will
require round trip latencies below 50ms. The NHTSA believes
that “At this time, DSRC is the only mature communication
option that meets the latency requirements to support vehicle
communication based crash avoidance.” - although it has
clearly kept the door open for other technology platforms. In
addition, proponents point out that DSRC and C-ITS have been
www.eenewseurope.com eeNews Europe
COVER STORY
in development for almost twenty years and this means they are
stable and assured. Large-scale tests are underway in various
locations. Meanwhile, the 5G Automotive Association (5GAA),
main industry body promoting C-V2X, promises that it “can be
commercially ready by 2018.”
Opponents of DSRC and C-ITS will say that the very fact that
these technologies have been under development for close to
two decades without reaching the mass market in any meaning-
ful form clearly suggests that they may never be market-ready.
Also that they are grounded in old technology that lacks a viable
upgrade path. It may also be pointed out that although LTE and
5G technology might not have originally been designed with
V2X in mind, C-V2X can leverage the economies of scale that
the huge global mobile telecommunications market present in
order to reduce hardware, software and development costs.
Can Old and New Standards Co-Exist?
The 5GAA states that C-V2X technology does not have to be a
complete replacement for C-ITS and DSRC. C-V2X is designed
to be able to share the available spectrum with 802.11p-based
V2X systems by switching to vacant channels within the 5.9GHz
band. In addition, it can support the upper protocol layers of
C-ITS and DSRC - effectively replacing only the IEEE 802.11
MAC and PHY radio layer with 3GPP-specified equivalents,
while allowing the specialised higher-level DSRC and C-ITS V2X
messaging standards to run as normal. With these capabilities,
C-V2X appears to meet the NHTSA’s requirements for manda-
tory V2X devices.
Perhaps the question of which technology is better is not
actually as critical as it seems. Automobile OEMs could choose
the underlying technology they prefer, or even hedge their bets
and provision for supporting both - IC vendor Qualcomm has,
for instance, recently announced its 9150 C-V2X chipset which
does just this.
Come and meet us at the embedded world 2018, stand 111
Hall 3A. Discover our newest products and get a chance to
win one of the hundreds Dev Kits at the booth.
Mouser Electronics
Authorised Distributor
www.mouser.com
February 2018 News 11
 NEWS & TECHNOLOGY
Morris Chang gives TSMC’s outlook
for 2018, says goodbye
By Peter Clarke
A
s is customary at the beginning of the year chairman
Morris Chang has given TSMC’s view of the growth
prospects for the semiconductor and foundry markets in
the year to come.
However, while speaking at the company’s
conference to discuss its fourth quarter 2017
results, Chang said it would be the last time
he would be present at the quarterly financial
conference before he retires on June 5.
Chang told financial analysts that the global
semiconductor market would grow by between
6 and 8 percent in 2018, while the market
excluding memory would grow more slowly at
between 5 and 7 percent. Chang’s numbers are
at odds with those of bullish forecaster Future
Horizons but in line with other market analysts.
Chang expects foundries to continue to take
manufacturing market share from IDMs, as has
been the long-term trend since the formation
of TSMC and the foundry business model. The
total foundry market will grow by 9 to 10 per- TSMC’s Chairman Morris Chang. was based on financial speculation it was a
cent in 2018 but Chang expects TSMC to take
market share from other foundries and grow at 10 to 15 percent.
In the first half of 2018 TSMC sales will be up slightly more
than 15 percent year-on-year while in 2H18 sales growth will
moderate to slightly less than 10 percent year-on-year, Chang
said.
Chang said the growth markets that will drive TSMC in 2018
were high performance computing (HPC), IoT and automotive.
Of these the strongest is HPC with strong de-
mand for artificial intelligence processors in all
electronic devices and demand for cryptocur-
rency mining chips.
At the end of the conference Chang made
a few extra remarks as he said this would be
his last appearance before the analysts prior to
his resignation. “I really have spent many years
with you,” he said. “I have been here almost
30 years. I enjoyed it. I had a good time. I hope
you had a good time too. I will miss you and
thank you very very much.”
TSMC reported net income of NT$99.29
billion (about US$3.36 billion) on revenue of
NT$277.57 billion (about US$9.40 billion) in
4Q17. The revenue was up 5.9 percent year-on-
year. TSMC’s Co-CEO Mark Liu.
For 1Q18 TSMC said it expected revenue of
between US$8.40 and US$8.50 billion, an increase of 12.8 per-
cent year-on-year at the mid-point. The management team also
disclosed it has set a capital expenditure budget of between
US$10.5 billion and US$11 billion. This is similar to 2017 and is
likely to be kept “flattish” for the next few years.
With TSMC expecting a compound annual growth rate over a
similar time of 5 to 10 percent this will result in capex falling as
a percentage of revenue from 30 to 35 percent as it is today to
20 to 30 percent in the future, the company said.
12 News February 2018 @eeNewsEurope
FOUNDRY ANALYSIS
However, a major talking point of the conference was the
surging demand that TSMC is seeing for cryptocurrency mining
processors.
Co-CEO Mark Liu quickly addressed this in his comments
with reference to cryptocurrency mining IC
demand, saying: “In the past TSMC’s open
innovation ecosystem has helped incubate
numerous growth drivers for the semiconductor
industry. In the 1990s it was PC chipsets, in the
early 2000s graphics processors, in the mid-
to-late 2000s it was chips for mobile phones
and starting in 2010 smartphone chips. Those
waves continuously sprout in our ecosystem.”
Liu continued: “Lately we’ve observed
demand for cryptocurrency mining has surged.
Cryptocurrency mining requires high per-
formance, low-power computing. TSMC’s
advanced technology suits it well.”
Liu added that TSMC has sized the 2018
market for CCM carefully but that because it
difficult market to assess far into the future.
“However we are quite certain that deep learning and block-
chain, which are at the core of cryptocurrency mining, will lead
to new waves of semiconductor innovation and demand for
years to come.”
Later in the call Liu revealed that of its four major growth
areas mobile phones, while responsible for nearly 50 percent of
sales, will be flat in 2018 suggesting market saturation is being
reached. The growth areas are high perfor-
mance computing (HPC), IoT and automotive,
he said.
Liu said Smartphones would continue to
include new features in 2018 including 3D au-
thentication and 3D sensing with 7nm process
being used for premium phones and 12nm for
mainstream phones. Both are on-course to
ramp in 2018.
However, HPC business will enjoy the most
growth from sales of GPUs for machine learn-
ing and cryptocurrency mining ASICs, Liu said.
These will use processes from 16nm down to
7nm. The key applications for IoT are smart
voice processing assistants, application pro-
cessors for wearables and wireless microcon-
trollers for smart homes. 22nm processes will
be key to those applications.
Morris Chang, chairman of TSMC speaking, also weighed in
cryptocurrency mining in response to a question from an ana-
lyst. He said: “The urge to mine cryptocurrency is very strong
but the price of cryptocurrency is very volatile. Demand right
now, last year, has been very strong. We have sized the demand
carefully. I believe we satisfy customers without being too opti-
mistic,” he said.
TSMC declined to give a figure for how much of TSMC’s
revenue cryptocurrency mining is currently responsible.
www.eenewseurope.com
 CYBERSECURITY

When AI enters the maze of cybersecurity

By Julien Happich

T
aking place in Lille, approximately one hour North of Paris
by train, the 10th edition of the International Cybersecurity
Forum (Forum International de la Cybersécurité or FIC in
French) gathered approximately 8,600 visitors and the 13000m2
of exhibition floor were pretty packed, reflecting a 20% atten-
dance increase over last year.
Broadly, the focus was on how to secure the cyberspace,
protecting the digital services and infrastructures that nations
and citizens have come to depend upon, implementing the right
tools to protect data, “a strategic asset” as put it Juri Luik, Es-
tonia’s Minister of Defence who came to testify how his country
had been the first in Europe to suffer a full-out cyberattack from
its Russian neighbour.
“Unlike in the past, the defence sector depends on the civil-
ian sector that is driving the technology, innovation is led by
the private sector and the state has to adjust to that. All states
shall invest in cybersecurity as national borders do not exist
in the cyberspace where everything is merged into one battle- Florence Parly’s keynote, France’s Minister of the Armed
field”, Luik said during a plenary session titled “Innovation and Forces at FIC.
Cybersecurity”.
“It is not just a technical matter for tech junkies, but it “All our servers are situated in France and our encrypted
impacts politics too” he added, mentioning last year’s Euro- data is hosted by French cloud provider OVH in compliance
pean-level cyber defence exercises focusing on situational with European laws” highlighted Nicolas Diaz, Communication
awareness, crisis response mechanisms and strategic commu- Consultant at Yes We Hack during an interview with eeNews
nication. Europe. With its Bounty Factory and its 3700+ registered ethi-
Florence Parly, France’s Minister of the Armed Forces joined cal hackers, the company can offer continuous testing while
Luik on stage to highlight the necessary cooperation between federating and ranking the white-hats, also raising their profiles
all the European states on cybersecurity. “Everything is con- for would-be recruiters in cybersecurity.
nected, the IoT collects data, analyse it and with this hypercon-
nectivity, all our society and e-lifestyle is threatened by invisible
untraceable enemies” Parly said. “The digital space is structur-
ing the battlefield, at home and abroad, and if we don’t innovate
and invest in research for cybersecurity, others will do it in our
place. But I won’t let this happen”, she added before discussing
the need for reorganizing Europe’s numerical resilience, to bet-
ter anticipate threats, protect our networks and fight back when
necessary.
As cyber-sovereignty as become a major stake, the French
Minister of the Armed Forces unveiled plans for the French
army to invest 1.6 billion Euros for cyber-defence between 2019
and 2015, increasing the ranks of its cyber-soldiers from 3,000
today to 4,000 by 2025. She also announced the creation of a
European kernel of cyber-defence to share incoming threats in
real time.
A plethora of encryption and network traffic monitoring solu-
tions were on display, but among the 350 exhibitors were also
companies offering cyber security testing environments and Yes We Hack’s ethical hacking services.
so-called ethical hackers ready to safely hack into your system
and disclose your network’s vulnerabilities. AI in cybersecurity
In the European context, security consulting company Yes One hot trend at the conference was the use of Artificial
We Hack claims to offer the first Bug Bounty platform in Europe. Intelligence (AI) to counter cyber-attacks or stop malware in its
One that allows companies to crowd-source their security tests, track. In a talk titled “Connected Intelligence”, Trend Micro’s
relying on so-called “white hats” or ethical hackers who’ve technical director Renaud Bidou gave us his vision for 2020,
agreed to follow responsible vulnerability disclosure procedures, arguing that learning from experience, more bugs and more
abiding by the rules, principles and legislation of the European patches would follow, opening room for new worms exploiting
economic area. Yes We Hack’s Bounty Factory helps companies unpatched vulnerabilities. “What will never change is that there
create a Bug Bounty Program (BBP) which rewards individual will always be bugs, patches, and users will click” (on links to
hackers for finding bugs in code (software, web sites, network malware or compromised documents), he said.
protocols etc.). “Hackers will be hackers and someone will eventually enter

www.eenewseurope.com eeNews Europe February 2018 News 13

 NEWS & TECHNOLOGY
Trend Micro’s slide on the use of AI for threat detection.
your system, this will never change. What’s new is that the
surface attack will be less controllable as it expands, from
contained servers to workstations, laptops, smartphones,
connected watches, homes, smart factories and the pervasive
cloud” Bidou noted.
“But the good thing is that we can now use AI to secure the
connected world and detect on-the-fly suspicious unknown
threats. Tomorrow, smart security won’t be good enough, we
need intelligent security.”
Describing a successful phishing scenario where a hacker
could impersonate a colleague with all his credentials, say the
CFO, asking by direct email to the CEO an unusual money
transfer, Bidou said that even such an attack could be auto-
matically thwarted by an AI algorithm, modelling all human
behaviours in the company as well as the traffic patterns on
the network to identify any deviation from standard behaviours.
“An AI-driven security system could email back to the CFO
and check if he really made that request, it could also warn the
CEO” Bidou put as an example.
“In 2020, data will overload our traditional analysis capa-
bilities, but AI takes its value from data and it will have plenty
of it from IoT. So let’s take advantage of this context to make
security more reliable”.
Asked if AI would also benefit hackers in the arms race,
Bidou noted that AI was already being used by hackers to
scavenge data on social networks and identify the profiles that
would be the most susceptible to phishing. “That way, they
augment their chances to successful phishing. Some adverts
on the web are even compromised and monitored in order to
identify the most gullible profiles”, the technical director told
eeNews Europe.
On the exhibition floor, a
spokesperson for PhishMe
somehow agreed that with
increasing encryption and
security checks at endpoints,
phishing based on social engi-
neering remains the surest way
to get administration credentials
and compromise a network.
On its website, the company
reports that more than 90% of
breaches can be attributed to
successful phishing campaigns.
Hence, it proposes to train
employees and partners so they
become part of the cyberse-
curity solution. PhishMe does
that by organizing innocuous
phishing campaigns (taking the One of FIC’s master classes by Securitude Cyber Solutions
unapologetic clicking end-user Co-Founder Yul Bahat.
14 News February 2018 @eeNewsEurope
CYBERSECURITY
to awareness campaigns about phishing).
Now, could hackers step up their game with AI to analyse
and understand how colleagues communicate and what lan-
guage patterns they use, so their phishing campaigns would be
more successful? The possibility exists, though it may take a
long time before such efforts pay-off, as humans are unique in
understanding the specific social traits and forms of expression
of their collaborators, argued the PhishMe representative.
Going back to the use of AI for cybersecurity, American
company Vectra leverages artificial intelligence to automate
security operations, identifying threats by detecting abnormal
data traffic across the network, starting at endpoints. Whether it
is to exfiltrate data or encrypt it with ransomware, hackers need
first to compromise an endpoint and scan for the protective
measures in place.
Vectra’s Cognito tool analyses the behaviour of every piece
of equipment and of all users across the network and identi-
fies any deviation from standard. The tool automates the hunt
for cyber attackers and can instantly prioritize the highest-risk
threats, triggering immediate response such as endpoint isola-
tion if need be. “80% of the algorithms they use are pre-trained
for certain patterns but 20% of them learn on the go in the
network so detection can be customised to the environment
were Cognito operates”, explained David Clarys from distribu-
tion partner Exclusive Networks. The AI algorithms and the
data analysis are centralized in a dedicated Vectra appliance.
Last year, the company reported a triple-digit revenue growth,
quarter on quarter.
Likewise, Cambridge startup Darktrace leverages unsuper-
vised machine learning to detect data anomalies and threats
across all network types, from physical, virtualized, and cloud,
through to IoT and industrial control systems. The company
likes to describe its security solution, the Enterprise Immune
System, as a set of AI algorithms that does not require previ-
ous experience of a threat or pattern of activity in order to
understand that it is potentially threatening. The Enterprise Im-
mune System works automatically, without prior knowledge or
signatures, detecting and fighting back against subtle, stealthy
attacks inside the network, in real time. On its website, the
company claims its unique, unsupervised machine learning has
already identified over 63,500 previously unknown threats in
over 5,000 networks, including zero-days, insider threats and
subtle, stealthy attacks.
Detection is dead!
“Detection is dead!” Exclaimed
in one of the FIC master classes
Yul Bahat, Co-Founder of novel
cybersecurity consulting firm
Securitude Cyber Solutions. “An-
tivirus are no longer an adequate
response” Bahat argued, observ-
ing that attacks are no longer fron-
tal but diffuse and that the core
protection mechanism offered by
antiviruses has consistently failed,
tricked by unknown signatures,
unknown behaviours, time delays
and so on. The last unfamous ex-
amples being the WannaCry and
Petya worms last year.
According to Bahat, compara-
tive studies of antiviruses showing
100% efficacy are all lies, tests are
www.eenewseurope.com

done in a lab environment, with
all the features turned “on”. But
false positives are too high when
all the features are “on”, no one
use them all he argues.
“I have a problem with the en-
tire idea of detection, looking at a
code and deciding if it is good or
not. It doesn’t matter how good
detection is, it is an impossible
task. With the evasive techniques
they use, hackers have become
experts in hide & seek. We just
lie to ourselves, pretending that
if we can’t see the attackers then Minerva-labs’ hostile environment “freezes” evasive malware would be attackers” told us Peter
they are not there”.
into inaction.
Bahat took again the Wanna-
Cry worm as an example of environment-aware attacks, noting
that first the worm had been cautiously isolated in a sandbox by
most antiviruses, but then because it featured a time delay and
stayed put for some days, it had been released again.
As a response to hackers, Bahat proposes innovative de-
fence scenarios, based mostly on re-enforced network isolation
and hacker deception.
“We want to keep malicious stuff as contained and far away
as possible, we need to assume that everything is bad. You
must shut down all direct access to your assets”, Bahat said,
proposing what he calls “content disarmament & reconstruc-
tion” as one solution. “You take out everything that is impor-
tant, take it apart and reconstruct it entirely with known safe
code”.
For web browsing isolation, Bahat suggests that all the
information be processed via the proxy who only sends an
interactive HTML5 video stream in real time instead of the real
pages. The users doesn’t even know the difference, but that
keeps the end-points isolated. Then once the session is over,
all the content together with the virtual browser is destroyed.
Deception is the all-time favourite of Bahat. “First we
confuse the attacker, we fake the environment with simula-
tion, it all starts with an endpoint. Since evasive malware uses
environment awareness to detect what antiviruses are there,
if it has been sandboxed or not, what analysis tools are there,
why not put debuggers, sandboxes, virtual machines on all
endpoints? Why not simulate a thousand guard dogs?
The idea is that if all the antiviruses look like they are present
on the machine, with sandboxes in operation, the malware will
choose to do nothing and freeze. “We are not trying to detect
the malware, we stop it”, clarifies Bahat.
“Now the goal of malware is to move to other endpoints, so
we want to give hackers a thousand fake servers, so they have
no way to know where to go. All of it is simulation, we just make
it look like they are real, and if they choose one server, they’ll
trigger an alert. We want to be able to track their moves, setting
up believable traps or honey-pots”, Bahat continued.
“But we need believable decoys. We can deploy fake end-
points, fake PLC machines in an industrial environment or fake
files on servers”. According to Bahat, deception was the biggest
trend among last year’s innovations. Next, once the attack has
been identified, the malware can be removed.
Securitude Cyber Solutions offers its services to international
companies, figuring out what their needs are and suggesting
the best fitting solutions on the market. On its stand, partner
Minerva-lab provides the anti-malware evasion platform for end-
points to deceive threats into their dormant state. Other booth
www.eenewseurope.com
CYBERSECURITY
partner Cymmetria advertises its
decoy-creation platform dubbed
MazeRunner, allowing companies
to place a plethora of deception
elements in the hackers’ critical
path while creating a realistic en-
vironment to detect and quaran-
tine attackers, gathering forensic
data as they move innocuously in
this fake environment.
“Since we create dummy end-
points, servers and webservers,
all with fake credentials, the only
people to use these credentials
Glock, Managing Director Europe
for Cymmetria. So it becomes
easy for the fake environment to log all of the attacker’s moves,
tools and tactics within the deception network, giving an insight
on the new hacking strategies while containing them.
Antivirus companies have already jumped on the deception
bandwagon, as Symantec illustrated during a brilliant technical
demonstration were pretty much all of these new tricks were in
place.
A Cyber Range demonstration from Cyber Test Systems.
The European Union’s looming Network and Information
Security (NIS) Directive will make it mandatory in 2018 for all
operators of essential services (networked utilities providers and
public services) as well as digital service providers to implement
“state of the art” cybersecurity technologies to manage the
security risks of their networks and systems. This means that
security incident response teams will need to be trained and be
able to prove it. This is how CyberTest Systems claims its Cyber
Range becomes an essential tool as a crisis management train-
ing ground.
CyberTest Systems’ CEO Arnaud Kopp, describes the Cyber
Range as an infrastructure deployed to mimic a company’s real
infrastructure. Like a test lab, where the security teams can
be put to test with cyber-attacks confined to the Cyber Range
rather than delivered into the wild and putting other users at
risk. As well as providing safe grounds for training, the company
can invite other parties such as Yes We Hack to contribute with
multiple attack scenarios. A basic Cyber Range costs around
100,000 Euros revealed the CEO, adding that cost increased
with the complexity of the infrastructure to replicate.
eeNews Europe February 2018 News 15
 NEWS & TECHNOLOGY
Intel shows important milestone
for quantum research at CES 2018
By Jean-Pierre Joosting
D
uring his keynote address at CES 2018, Intel CEO Brian
Krzanich announced the successful design, fabrication
and delivery of a 49-qubit superconducting quantum
test chip
Only two months after delivery of a
17-qubit superconducting test chip, Intel
has unveiled “Tangle Lake,” a 49-qubit
superconducting quantum test chip. The
chip is named after a chain of lakes in
Alaska – an analogy to the extreme cold
temperatures and the entangled state
that quantum bits (or “qubits”) require to
function.
Tangle Lake is an important milestone
in developing a complete quantum
computing system – from architecture
to algorithms to control electronics. The
49-qubit test chip will allow research-
ers to better assess and improve error
correction techniques and simulate
computational problems.
According to Krzanich, the nascent
field of quantum computing will solve problems that today
might take our best supercomputers months or years to resolve,
Dupont prints carbon and silver inks
onto stretchable film for heated garments
By Julien Happich
I
n partnership with technical fabrics Taiwanese provider
Formosa Taffeta Company, DuPont Advanced Materials (Du-
Pont) has developed a powered
smart clothing technology for on-
body heating, dubbed Intexar Heat.
The new fabric is a thin lightweight
and durable heating solution for out-
door clothing and is designed to be
easily integrated into garments.
Intexar Heat consists of a thin
layer of carbon resistors, intercon-
nected by an underlying layer of silver
electrodes printed on a stretchable
thermoplastic polyurethane (TPU)
laminate. The silver electrodes feed
currents throughout the resistor array
to radiate a controlled heat within
garments. Typically, the active layer
is sandwiched between a plain or
customized outer protective layer that
shields the ink from exposure and the
fabric making up the garment.
“Intexar Heat is a revolutionary stretchable ink and film that
16 News February 2018 @eeNewsEurope
QUANTUM COMPUTING
such as drug development, financial modelling and climate
forecasting.
“In the quest to deliver a commercially viable quantum
computing system, it’s anyone’s game,” said Mike Mayberry,
corporate vice president and manag-
ing director of Intel Labs. “We expect
it will be five to seven years before the
industry gets to tackling engineering-
scale problems, and it will likely require
1 million or more qubits to achieve com-
mercial relevance.”
The need to scale to greater numbers
of working qubits is why Intel, in addition
to investing in superconducting qubits,
is also researching another type called
spin qubits in silicon. Spin qubits could
have a scaling advantage because they
are much smaller than superconducting
qubits. Spin qubits resemble a single
electron transistor, which is similar in
many ways to conventional transistors
and potentially able to be manufactured
with comparable processes. In fact, Intel has already invented a
spin qubit fabrication flow on its 300mm process technology.
when powered, creates a comfortable warmth,” according to
Michael Burrows, global business manager, DuPont Advanced
Materials.
“Our team has worked hard to
develop a heater that feels like
fabric, doesn’t rely on cables,
thick wires or big batteries, and
can stand up to very cold environ-
ments.”
Formosa Taffeta Co will be the
first textile manufacturer to take
advantage of Intexar Heat technol-
ogy as part of its Permawarm line,
providing clothing brands with a
complete garment heater system
including the Intexar heater layer,
connectors and control software.
Intexar materials also can enable
biometric monitoring in smart cloth-
ing. Pulse rate, respiratory rate,
muscle activity and form awareness
are all measurable using sensors and conductive pathways built
from Intexar, notes DuPont.
www.eenewseurope.com
 ELECTRONIC DESIGN AUTOMATION

Startup launches analog IP

from digital tools
By Peter Clarke

M
ovellus Inc. (San Jose, Calif.) has launched itself on the market with a focus
on the use of digital design and verification tools to implement analog circuit
functionality.
The advantage of the approach is that because the functions are designed digitally
time-to-market can be shorter and IP can be more easily ported to alternative process
nodes or layer counts.
The company has launched a series of phase-locked loop (PLL) delay-locked loop
(DLL) and low-voltage drop out (LDO) regulator generators.
It is not clear how the circuits compare with similar analog implementations in terms
of performance, power consumption or area but the company claims that since its Launching
formation in April 2014 it has produced working silicon with multiple customers. These
include semiconductor and systems companies working in artificial intelligence, net- February
2018

Increased
Reliability
for Industrial Applications
New 1.27mm pitch
working, and FPGAs. Archer Kontrol connectors
The PLL generator is available in TSMC 28nm HPC and Globalfoundries 14nm LPP in horizontal and vertical
process nodes. Meanwhile several FinFET process nodes are being prepared. The DLLs layouts with 12-80 pin
are designed to be robust against environmental and process variations and occupy
10,000 gates independent of the process technology while meeting a wide range of combinations.
specifications, Movellus said. The IP generators combine proprietary analog circuit Designed with surface mount solder
architectures (RTL) with add-on software products that expand digital synthesis, static
tabs for additional board retention
timing analysis, and place and route tools to create digital implementations of analog
functions, Movellus said. strength, it can withstand lateral
Fabless chip company Mythic Inc. (Austin, Texas) has experience of Movellus’ tools and twisting forces in high vibration
and IP. Founded in 2012 as Isocline Engineer Corp., Mythic is adopting a “processing- environments.
in-memory” approach to neural network implementation.
“Mythic delivers AI that performs hybrid digital/analog calculations inside flash ar- Temperature range of
rays, resulting in orders of magnitude increase in efficiency,” said David Fick, CTO of -55°C to +125°C
Mythic. “The Movellus PLL generator has unique IP and software technology that ex- Assists with blind mating
pands existing digital tools to enable them to automatically generate PLLs, eliminating
manual analog design methods.” Fick continued: “Their generator allowed us to focus Fully shrouded connector
on designing our SoC without worrying about any schedule slips due to late manual an- system
alog customizations needed to meet our goals. For example, when we needed to push Tested to perform up
a metal stack change, Movellus was able to provide a correct IP in a matter of hours.”
to 500 operations
“By expanding existing digital design tools to create functionality previously only
achievable in analog design, Movellus dramatically improves efficiency and reduces
development and verification time by months,” said Muhammad Faisal, co-founder and
CEO of Movellus, in a statement.
“I invested in Movellus after watching Muhammad and his team cross the critical
hurdle of proving their technology in silicon with customers,” said Jim Hogan, managing
partner at Vista Ventures.
Prior to founding Movellus Faisal held positions at PMC-Sierra and Intel. The www.harwin.com/
management team also includes Jeffrey Fredenburg, co-founder and vice president of
engineering, who leads the development of Movellus analog generator technology. The
kontrol
company’s chief operating officer is Saeid Ghafouri who joined Movellus having held
executive positions at CoWare, Magma Design Automation, interHDL, Synopsys, Silicon
Architect, and Cadence.

www.eenewseurope.com eeNews Europe February

Harwin 2018
Archer Kontrol News
EETimes Europe Feb 18.indd 1 17 10:47
12/01/2018
 NEWS & TECHNOLOGY
Drones learn autonomic navigation from cars and bicycles
By Christoph Hammerschmidt
T
he DroNet algorithm, developed by researchers at the
University of Zurich (UZH) enables drones to fly com-
pletely independently along the streets of a city and in
buildings. The algorithm had to learn traffic rules and driving
examples from cyclists and car drivers.
For navigation, commercial drones use GPS, which works
well at high altitude. But what happens when the drones fly
independently between buildings or in the dense road network,
where cyclists and pedestrians can sud-
denly cross their way? Until now, com-
mercial drones have not been able to
react quickly to such unforeseen events.
Researchers from the University of
Zurich and the National Centre of Com-
petence in Research (NCCR) Robotics
have now developed the DroNet algo-
rithm, which can safely guide drones
through the streets of a city. This was
set up as a fast residual network with
eight levels and generates two outputs
for each input image: one for navigation
to fly around obstacles and one for the probability of collision to
detect and react to dangerous situations. DroNet detects static
and dynamic obstacles and reduces speed to prevent colli-
sions. With the algorithm, the researchers see themselves one
step closer to the goal of integrating independently navigating
drones into our everyday life.
The Swiss researchers’ drone uses a normal camera such
as that of a smartphone, and a powerful algorithm for artificial
intelligence to evaluate the observed situations. This algorithm
Artificial Intelligence enters automotive development
By Christoph Hammerschmidt
T
he German Research Center for Artificial Intelligence
(DFKI), the world’s largest non-profit research centre
for Artificial Intelligence (AI), and engineering services
company IAV GmbH announced that they open a joint R&D
laboratory dedicated to the application of AI in the development
processes of the automotive industry.
In the new test environment at DFKI in Kaiserslautern (Ger-
many), special analysis methods of artificial intelligence for
use in test procedures in automotive development are being
researched and developed, with focus on
machine learning technologies such as
deep learning and time series analysis. IAV
plans to apply the potential of AI to devel-
opment procedures and methodologies in
powertrain development. This includes the
use of AI in engine control systems such as
ECUs as well as sounding out the potential
of AI to increase the efficiency and robust-
ness of the development process.
The application potential of intelligent
data analysis methods for monitoring and optimizing test data,
ECUs and test benches in the automotive industry is regarded
as extraordinary high. For example, a modern engine control
unit has more than 50,000 setting parameters that are decisive
18 News February 2018 @eeNewsEurope
AUTONOMOUS NAVIGATION
consists of a so-called Deep Neural Network. The software
learns to solve complex tasks using numerous training exam-
ples. “It shows the drone how it solves certain tasks and difficult
situations,” explains Davide Scaramuzza, professor for robotics
and perception at the UZH.
One of the biggest challenges for Deep Learning is to collect
several thousand of such training examples. In order to gather
enough data, the researchers collected travels of cars and bi-
cycles that navigated in urban environ-
ments and complied with traffic regula-
tions. By imitating, the drone learned
to follow rules such as “How to follow
the road without getting into oncoming
traffic” or “How to stop when obstacles,
such as pedestrians, construction sites
or other vehicles block my way”. The
researchers were also able to show that
their drone was not only able to navi-
gate through roads, but also found its
way into completely different environ-
ments for which it was never trained -
for example in multi-storey car parks or office corridors.
The study shows the potential of drone deployments for
surveillance tasks or parcel deliveries in busy environments as
well as for rescue operations in the event of urban catastrophes.
However, the research team warns of exaggerated expecta-
tions of what light, low-cost drones can do. “Many technologi-
cal problems still have to be solved before the most ambitious
applications can become reality,” explains PhD student Antonio
Loquercio from the UZH.
for fuel consumption, wear and tear and the overall perfor-
mance of the engine. Deep learning technologies, or more pre-
cisely the use of neural networks in the control unit, enable the
latter to “learn” independently how to optimally adjust the input
variables. The use of such networks in the time series analysis
of engine test data also opens up new approaches to predictive
engine health monitoring, thus improving the prediction of wear
and maintenance cases. Such methods are to be researched
and developed in the new research laboratory.
At the same time, FLaP, as the new lab
is called, will also work on new visualiza-
tion options for the diverse measurement
data from the neural networks. The aim is to
create a toolbox of AI tools that can be used
intuitively and appropriately by automotive
engineers.
“In addition to the promising possibilities
for use on existing hardware in produc-
tion vehicles, FLaP also investigates novel
applications for self-learning neural net-
works,” says Prof. Dr. Andreas Dengel, Head of the Smart Data
& Knowledge Services Research Unit at DFKI. “The aim is both
to gain fundamental knowledge in the long term and to develop
concrete solutions to current problems in the short term.”
www.eenewseurope.com

Electric mobility threatens stability of electricity supply, study says
By Christoph Hammerschmidt
M
any see electric driving as a technology of the future.
But it’s not that simple: an analysis by consulting firm
Oliver Wyman warns of widespread power cuts if elec-
tromobility quickly finds too many friends.
The study predicts that the low-voltage grid will be over-
loaded if electric vehicles exceed 30 percent. According to the
current regulatory framework, this would be the case in five to
ten years’ time. Comprehensive
problems are to be expected
from around 2032 onwards.
Although the study explicitly
refers to the German electricity
supply network and the regula-
tory framework conditions that
apply here, the problem is likely
to arise in other industrialized
countries in a similar form, albeit
with different parameters.
The authors of the study as-
sume that from 2035 onwards,
more than one in three cars will
be driven by battery-powered
vehicles. These cars charge their
energy via public and private
charging stations, which are
powered by the low voltage grid (400V to 1000V AC in most
European countries). The low-voltage grid is not designed for
this number of electric vehicles.
A calculation example shows that in particular suburban and
rural areas are susceptible to power outages caused by charg-
ing electric vehicles: with a local network size of 120 house-
holds, 36 electric cars are already sufficient to overload the
network at certain points. Without preventive measures, power
outages across the board are to be expected from 2032 on-
wards. In order to avoid this, operators would have to invest up
to 11 billion euros in the expansion of their networks under the
current framework conditions and with an electrification of 50
BMW signs up low temperature battery tech
By Nick Flaherty
C
ar maker BMW has signed an intellectual property (IP)
agreement with EC Power in the US on low temperature
lithium ion battery technology.
Most lithium ion cells struggle to oper-
ate below freezing which can be a problem
for electric and hybrid vehicles, requiring
additional heaters to boost the tempera-
ture of the battery packs. EC Power’s pat-
ented All-Climate-Battery (ACB) technol-
ogy operates in sub-zero temperatures by
creating a self-heating function that warms
up batteries from a low ambient tempera-
ture to the optimal operating condition in
a matter of seconds, either prior to or during driving. This rapid
self-heating function, which uses a custom material in the cell
and so requires no external power source, also enables low-
temperature fast charging in electric vehicles.
www.eenewseurope.com
ELECTROMOBILITY
percent of the automobiles. Thomas Fritz, author of the analy-
sis, explains: “There is already an urgent need for action on the
part of network operators, since network expansion requires
considerable lead times and high investments.”
As an economically more attractive alternative to the slow
and expensive expansion of the grid, the authors of the study
propose to make charging processes more flexible. The charg-
ing processes of electric cars
are usually so short that they are
connected to the mains for the
longest time at night without be-
ing actively charged. Most charg-
ing processes therefore have a
time flexibility. You do not neces-
sarily have to start at the moment
the car is plugged in. Instead, the
charging process can also begin
later at night without an electric
car user having to forego his fully
charged vehicle the next day.
“Due to the increased flexibility,
network utilization is spread over
a longer period of time so that
there is no network congestion.
This minimizes the risk of a wide-
spread power failure. Above all, an intelligent software solution
is required for implementation” comments Jörg Stäglich, head
of the energy team at Oliver Wyman.
Even if only 30 percent of electric vehicles take part in the
flexible charging process, the peak load drops significantly.
The higher the percentage of electric cars with flexible charging
processes, the lower the likelihood of network congestion. If the
network operators succeed in winning at least 92.5 percent of
the owners of e-vehicles for flexible charging, there is no need
to expand the network, even if the e-vehicle quota is 100 per-
cent. According to energy expert Fritz, flexible charging is thus
a “real alternative to conventional grid expansion”.
The self-heating mechanism creates an electrochemical
interface provides internal warm-up to 0ºC within 20 seconds
at -20ºC and within 30 seconds at -30ºC, consuming only 3.8
per cent and 5.5 per cent of cell capac-
ity, respectively. The ACB cell yields a
discharge/regeneration power of 1.06 and
1.4kW/kg at a 50 per cent state of charge
and at -30ºC provides up to 12 times the
power of state-of-the-art lithium-ion cells.
The company has used computer-aided
discovery for the development of innova-
tive battery and energy storage systems
through its AutoLion software and has a
battery factory in central Pennsylvania. The factory currently
produces thousands of prismatic and pouch cell in 10 to 20 Ah
capacity to demonstrate viability for mass production and in-
vehicle testing.
eeNews Europe February 2018 News 19
 NEWS & TECHNOLOGY
Graphene girders boost silicon anodes in lithium batteries
By Nick Flaherty
R
esearchers at the UK’s University of
Warwick have found a way to double the
life of lithium ion batteries using graphene
girders. The team at the Warwick Manufacturing
Group (WMG) found that replacing graphite in
the anodes of lithium-ion batteries using silicon
with graphene girders also increased the energy
capacity.
Researchers and manufacturers have long
sought a way to replace graphite with silicon, as it is easily
available and provides ten times the energy density, but it has
a number of problems too. The volume expansion of the silicon
particles impede further charge-discharge efficiency over time
and the material is elastic enough to cope with the strain of lithia-
tion when it is repeatedly charged, leading to cracking, pulverisa-
tion and a low lifecycle.
Dr Melanie Loveridge from the WMG found that a mixture
of silicon and a form of chemically modified graphene which
could resolve these issues and create a viable silicon anode in a
lithium-ion batteries. This could be practically manufactured on
an industrial scale and without the need to use nano-sized silicon
particles.
Separating and manipulating a few connected layers of gra-
phene gave the researchers a few-layer graphene (FLG) mate-
rial. The study has found that FLG can dramatically improve the
performance of larger micron-sized silicon particles when used
in an anode. The researchers created anodes that were a mixture
Recycling worn lithium cathodes in new batteries
By Nick Flaherty
E
ngineers at the University of California San Diego have de-
veloped a way to reuse worn lithium cobalt oxide cathodes
from spent batteries in new cells.
The degraded cathode particles are extracted from the used
battery and then boiled and heat treated. New batteries built with
the recycled cathodes see the original charge storage capacity,
charging time and lifetime, which is a considerable advance. The
lithium cobalt oxide cathodes are the most common in consumer
batteries, but the process also works on
nickel, manganese and cobalt (NMC) cath-
odes used in most electric vehicles.
Less than five percent of used lithium
ion batteries are recycled today, partly as a
result of the difficulties and cost of reusing
the components. Most have the materials
extracted to be re-used from scratch.
“Think about the millions of tons of
lithium ion battery waste in the future, espe-
cially with the rise of electric vehicles, and
the depletion of precious resources like lithium and cobalt, mining
more of these resources will contaminate our water and soil. If we
can sustainably harvest and reuse materials from old batteries,
we can potentially prevent such significant environmental damage
and waste,” said Zheng Chen, professor of nanoengineering at UC
San Diego. “We simply restore the degraded material by putting it
through the same processing steps,” he said.
This could also address economic issues. “The price of lithium,
20 News February 2018 @eeNewsEurope
BATTERIES
of 60% micro silicon particles, 16% FLG, 14%
Sodium/Polyacrylic acid, and 10% carbon
additives, and then examined the performance
(and the changes in structure of the material)
over a 100 charge-discharge cycles.
“The flakes of FLG were mixed throughout
the anode and acted like a set of strong, but
relatively elastic, girders,” said Dr Loveridge,
Senior Research Fellow at the WMG. “These
flakes of FLG increased the resilience and elasticity of the mate-
rial greatly reducing the damage caused by the physical expan-
sion of the silicon during lithiation. The graphene enhances the
long range electrical conductivity of the anode and maintains a
low resistance in a structurally stable composite.”
“More importantly, these FLG flakes can also prove very ef-
fective at preserving the degree of separation between the silicon
particles. Each battery charge cycle increases the chance that
silicon particles become electrochemically welded to each other,”
she said. “This increased agglomeration increasingly reduces and
restricts the electrolyte access to all the particles in the battery
and impedes effective diffusion of lithium ions, which of course
degrades the battery’s life and power output.”
The team has already started using this for a two year project
led by Varta Micro-innovations, along with Cambridge University,
CIC, Lithops and Italian Institute of Technology) into production
of silicon/graphene composites and processing into lithium-ion
batteries for high-energy and high-power applications.
cobalt and nickel has increased significantly. Recovering these
expensive materials could lower battery costs,” he said.
The method involves first collecting cathode particles from
spent lithium ion batteries. Researchers then pressurize the cath-
ode particles in a hot, alkaline, solution containing lithium salt. This
solution can also be recycled and reused to process more batches.
Afterwards, the particles go through a short annealing process in
which they are heated to 800ºC and then cooled very slowly.
Researchers made new cathodes from
the regenerated particles and then tested
them in batteries built in the lab. The new
cathodes showed the same energy storage
capacity, charging time and lifetime as the
originals.
Overall, the recycling process uses
5.9MJ/kg, equivalent to the energy in
about three quarters of a cup of petrol for
a kilogram of cathode material. This is half
the energy required by other lithium battery
recycling processes.
The next step is to optimise this process for industrial scales
and Chen’s team is planning to work with battery companies in
Asia. A key step is the cathode harvesting, which is currently done
by hand, and this would be automated.
The team has filed a provisional patent on this work and is aim-
ing to license it through the university’s Innovation and Commer-
cialization Office.
www.eenewseurope.com
 AUTONOMOUS DRIVING

Daimler tests AI in real-world traffic

By Christoph Hammerschmidt

S
ince traffic conditions vary significantly by country, future one another and can change their display within a short time.
autonomous cars need to be capable of machine learn- This demands more of, for instance, the Multi Purpose Camera
ing. As part of a test drive on five continents, Daimler (MPC) and the quality of digital maps. Speed limits or even data
has tested the conditions of road traffic in five continents. The which only apply at certain times of day are equally as challeng-
result: cars must be able to learn on their own. And, ideally, traf- ing.
fic conditions worldwide should be harmonized. Road and lane markings are not standardised around the
“Zebra” crossings on four-lane highways in China, right turns world either. In China, for example, the zebra crossing has a
from the left lane in Australia, pedestrian traffic on any kind of double meaning. In the city it marks a pedestrian crossing,
road in South Africa or temporary driving ban in the immediate on the motorway, on the other hand, it shows the minimum
vicinity of stopping school buses in the USA – on every conti- distance between vehicles driving behind one another. On
nent, human and electronic drivers
face challenges which will have an
influence on the driving charac-
teristics of future autonomous
vehicles. Automated and autono-
mous vehicles have to know about
these country-specific particulari-
ties and understand them in their
respective context in order to be
able to make the correct driving
decisions.
To study these challenges and
ways to overcome them, Daimler
organized its “Intelligent World
Drive” throughout Germany, China,
Australia, South Africa and the US.
The drive recently ended in Las
Vegas, USA. As test vehicles the
company used semi-automated
sedans of its S class.
The differences in traffic regula-
tions, driving habits and driv-
ing cultures across the various
countries gave an insight into the
complexity of global challenges
in the development of automated and autonomous driving multi-lane Interstates and Freeways in the USA there are often
functions. In particular the national particularities in terms of dedicated lanes for car pools of at least two people. They can
infrastructure, traffic regulations and the conduct of other road be separated from the other lanes by two uninterrupted yellow
users place very different requirements on the sensors and lines or by metal guard rails. It can be difficult for the vehicle
algorithms of the vehicles. “The Intelligent World Drive makes sensors to detect and correctly interpret them as special lanes.
it clear that autonomous driving requires global development In the USA there are also what are known as Botts’ Dots. These
activities and test drives” says Ola Källenius, Member of the dots for lane marking are plastic or ceramic and also put special
Board of Management of Daimler AG responsible for Group demands on lane detection. California is thus the first federal
Research and Mercedes-Benz Cars Development. “Automated state planning to abolish the Botts’ Dots and to standardise the
and autonomous vehicles need international learning material lane markings for future autonomous driving.
from actual road traffic in order to understand traffic situations Testing cars against country-specific features is the only way
and to be prepared for different scenarios.” to develop learning-capable systems and algorithms which
The differences in the countries give a small insight into the are in a position to cope with a multitude of situations reliably
complexity of global challenges in the development of automat- and safely, Daimler summarizes. The résumé of the Intelligent
ed and autonomous driving functions. In particular the national World Drive: To enable more highly automated and autonomous
distinctions in terms of infrastructure, traffic regulations and driving functions to be tested in real road traffic and then to
the conduct of other road users place very different require- enter series use, further international harmonisation of the legal
ments on the sensors and algorithms of the vehicle. In Australia, framework is necessary. There is a need for action, especially
electronic displays with variable speed limits are used. Special in relation to international agreements on road traffic law, which
displays indicate the current speed limit. They are equipped set the binding framework for national legislation and which
with bright white LEDs, a red LED ring and a yellow LED warn- currently still compulsorily require a driver. Further changes are
ing lamp and can also depict simple symbols and letters as important with regard to vehicle certification and data storage.
well as speed limits. In some cases they are positioned next to

www.eenewseurope.com eeNews Europe February 2018 News 21

 NEWS & TECHNOLOGY
Modular motion tracking glove with haptics for AR/VR
By Julien Happich
A
imed at virtual reality/augmented re-
ality (VR/AR) and robotics, the smart
glove developed by researchers
from the Tyndall National Institute has been
designed as a haptic human computer
interface system, complete with multiple
sensors and actuators distributed across
each finger.
To be showcased at CES, the smart
glove was developed as part of an innova-
tive project between the Wireless Sensor
Networks group at Tyndall National Institute, Ireland and project
leads Telecommunications Software & Systems Group (TSSG)
in Waterford Institute of Technology. Bray based, Product and
Interaction Design Consultancy, Design Partners, were engaged
as human factors engineering and industrial design experts for
the project.
“The smart glove for HCI is a culmination of over a decade
of research in the development of motion sensing for the hu-
man body and incorporates the latest sensing technologies
needed to bridge the human and digital worlds of Augmented/
Virtual Reality and robotics” wrote Dr Brendan O’Flynn, Head of
Tyndall’s Wireless Sensor Networks group, in an email exchange
with eeNews Europe.
“As opposed to traditional video gaming, where the visual
sense is the only significant sense that plays a role in the ap-
plication, in Virtual Reality (VR) and Augmented Reality (AR)
applications, the vestibular, proprioception and tactile senses
are important. VR/AR systems thus demand accurate and
low-latency Human Computer Interface (HCI) systems to avoid
motion sickness effects”.
ZnO TFTs revealed as next-gen in-screen force sensors
By Julien Happich
R
esearchers at the University of Cali-
fornia San Diego and the University
of Texas at Austin have demon-
strated a novel transparent force sensing
technology that can be readily integrated
into any type of display, as a one-micron
thin layer deposited on top of the back-
plane, using commonly used materials and
processes.
Until now, adding 3D touch (touch loca- A ZnO touchscreen array fixed onto a
tion and force) meant additional process- carrier wafer for testing with a display
ing electronics and sensing layers. But driver from Synaptics.
the researchers who leveraged the known
piezoelectric properties of ZnO to create force sensing thin film
transistors have now demonstrated that they could seamlessly
integrate their force sensors onto the Indium gallium zinc oxide
[IGZO] active matrix of most displays (the materials and deposi-
tion processes being similar). The force sensors can replace the
capacitive sensing layer as a whole since each pressure sensor
(transistor) also acts as a switch.
In a paper titled “Improved Performance of Zinc Oxide Thin
Film Transistor Pressure Sensors and a Demonstration of a
Commercial Chip Compatibility with the New Force Sens-
22 News February 2018 @eeNewsEurope
SENSORS
As well as all the required sensors and
actuators, the modular smart glove incor-
porates novel software, data-analytics, data
fusion algorithms, auto-calibration regimes
and power control regimes, to optimise the
user experience. It has been designed to be
easy to use, wear and wash.
“Embedded data fusion algorithms and
auto-calibration techniques ensure precision
analysis of fingers motion at an output rate
from 100 to 700Hz, faster than that of any
competitor. Because the algorithms run on the microsystem
within the glove no additional processing power is required on
the gaming console making it compatible with lower-cost infra-
structure”, noted O’Flynn.
The smart glove integrates 12 inertial sensors to account
for all the degrees of freedom of the hand. It integrates ten
tactile feedback actuators which can simulate tactile events in
the virtual world. In addition, it integrates 11 IR-LEDs for 3D-
positioning camera-based tracking that also aim to improve the
accuracy of the joint angles and absolute position estimation.
Wireless connectivity is provided by Wi-Fi and Bluetooth
standards to ensure compatibility with all gaming and AR/VR
platforms, with Unity-compatible outputs.
Based on standard USB connectivity, the modular construc-
tion allows for low-cost manufacture and the easy repair or
replacement of finger segments to adapt for different gaming
experiences, so users can build their own system in an organic
fashion. This means the one-size-fits-all plug-and-play system
can easily be adaptable to both right and left hands for sports
coaching, musicianship or any other learning environment.
ing Technology” published in the Journal
of Advanced Materials Technologies, first
author Siarhei Vishniakou describes how the
ZnO TFTs’ piezoelectric characteristics and
transistor performance had been optimized
through oxygen-rich deposition conditions.
For a 16x16 transparent pressure sensor
array, the paper reports a sensitivity of 4nA
kPa−1 and a latency below 1ms (an order of
magnitude faster than today’s touchscreens
response).
Taking Apple’s 3D-Touch technology as
a commercial benchmark, Vishniakou who
founded the Californian startup Dimensional Touch and cur-
rently serves as its Chief Technology Officer observes that his
technology could be implemented at a fraction of the cost of
Apple’s 3D-Touch, without the 13 grams penalty noted for the
iPhone 7. What’s more, the ZnO TFTs operate reliably even on
thin flexible substrates. Although it has proven the new ZnO
TFTs force sensors on a 1”x1” device, the company is confident
it can add its pixel-resolution force-sensing capability into any
type of display from 5’’ diagonal to 27’’ diagonal and higher, us-
ing existing manufacturing equipment.
www.eenewseurope.com

Nanowires for customisable, stretchable supercapacitors
By Nick Flaherty
R
esearchers in Singapore have created a customizable,
fabric-like power source that can be cut, folded or
stretched without losing its charge.
Led by Professor Chen Xiaodong of the School of Materi-
als Science & Engineering at Nanyang Technological University
(NTU), the team created a customisable supercapacitor where
the structure and shape can be changed after it is manufac-
tured, while retaining its function as a power source. Existing
stretchable supercapacitors are made
into predetermined designs and
structures, but this design can be
stretched in multiple directions so that
it is less likely to be mismatched when
joined to other components.
The supercapacitor has a hon-
eycomb-like structure built from a
strengthened manganese dioxide
nanowire composite that stores four
times more charge than most existing
stretchable supercapacitors. When
stretched to four times its original
length it also maintains nearly 97 per
cent of the charge, even after 10,000 stretching cycles. While
manganese dioxide is a common material for supercapacitors,
the long nanowire structure, strengthened with a network of car-
bon nanotubes and nanocellulose fibres, allows the electrodes
to withstand the strains of the customisation process.
Pairing the supercapacitor with a sensor and placed on the
human elbow, it performed better than existing stretchable su-
KAIST stamps vertical microLEDs onto flexible substrates
By Julien Happich
R
esearchers from the Korea Advanced Institute of Science
& Technology (KAIST) have developed a novel anisotro-
pic conductive film (ACF)-based transfer and intercon-
nection technology to transfer an array of vertically-grown inor-
ganic LEDs to a densely connected flexible substrate, yielding
what they call flexible vertical micro LEDs of f-VLEDs.
In a first step, an array of individual vertical LEDs is picked
up by a carrier stamp, then the carrier stamp is pressed into
contact with a precisely aligned bottom flexible electrode sub-
strate through the anisotropic conductive
film by compression bonding. Because the
ACF film contains conductive micro-parti-
cles, the compression bonding step simulta-
neously transfers the VLEDs and selectively
connects them to the underlying flexible
substrate. Such flexible micro LEDs could
be designed into high resolution wearable
displays and for wearable lighting.
For their research, the scientists led by
Professor Keon Jae Lee from the Department of Materials Sci-
ence and Engineering have designed the necessary transfer
equipment. They fabricated a 50x50 f-VLED array just about
5µm thin, reporting an optical power density of 30mW/mm2 or
three times higher than that of lateral micro LEDs, while improv-
ing the LEDs’ thermal reliability and lifetime by reducing heat
www.eenewseurope.com
FLEXIBLE ELECTRONICS
percapacitors, producing a stable stream of signals even when
the arm was swinging, says Chen. The data was then transmit-
ted wirelessly to external devices, such as one that captures a
patient’s heart rate. The team believes the supercapacitor could
be easily mass-produced as it would rely on existing manufac-
turing technologies at SGD$0.13 (USD$0.10) to produce 1 cm2
of the material and has filed a patent for the technology.
“A reliable and editable supercapacitor is important for devel-
opment of the wearable electronics
industry. It also opens up all sorts
of possibilities in the realm of the
‘Internet-of-Things’ when wearable
electronics can reliably power them-
selves and connect and communicate
with appliances in the home and other
environments,” said Chen.
The NTU team also worked with
Dr. Loh Xian Jun, Senior Scientist and
Head of the Soft Materials Depart-
ment at the Institute of Materials
Research and Engineering (IMRE),
Agency for Science, Technology
and Research (A*STAR). “Customisable and versatile, these
interconnected, fabric-like power sources are able to offer a
plug-and-play functionality while maintaining good perfor-
mance. Being highly stretchable, these flexible power sources
are promising next-generation ‘fabric’ energy storage devices
that could be integrated into wearable electronics,” he said.
generation within the thin film.
With the help of Professor Daesoo Kim from the Department
of Biological Sciences, the new f-VLEDs were then applied to
optogenetics, whereby light pulses are used to stimulate neu-
rons within localized cortical areas of the brain. In a paper titled
“Optogenetic Control of Body Movements via Flexible Vertical
Light-Emitting Diodes on Brain Surface ” published in the Feb-
ruary 2018 issue of Nano Energy, the researchers describe how
they inserted the flexible VLEDs between the skull and the brain
surface of a mouse to control her behaviour,
illuminating motor neurons on two-dimen-
sional cortical areas located deep below the
brain surface.
Brain science aside, other biomedical
applications envisaged by the researchers
include phototherapeutic treatment and
contact lens biosensors.
Back in 2016, Professor Lee has spun-off
a startup called FRONICS (abbreviated from
Flexible electRONICS) to commercialize its transfer technolo-
gies to companies willing to design their own IoT products,
based on flexible micro LED technology. According to the com-
pany’s website, the anisotropic conductive film-based transfer
and interconnection technology is applicable to both GaN and
GaAs LEDs and can support resolutions from 400 to 1300ppi.
eeNews Europe February 2018 News 23
 NEWS & TECHNOLOGY
Israeli startup turns luminaires into
wireless power chargers
By Julien Happich
A
t CES in Las Vegas,
Israeli startup Wi-Charge
unveiled the first ever
solution of room lighting com-
bined with wireless charging,
designed for a café environment.
The FDA-approved solution,
dubbed LIGHT, can deliver up to
2.5W of power at a distance of
up to 5 meters, simultaneously
recharging a number of mobile
and multimedia battery-powered
applications across a room,
totalling up to 50Wh/day.
Nowadays, when you speak about wireless charging, you
immediately think about either inductive coupling as it is imple- CEO says.
mented in charging mats or RF-based solutions as demonstrat-
ed by the like of Powercast or Energous. But Wi-Charge claims
its solution is both safer (radiation-free, interference-free) and of a percent. Everything I have spotted on public reports agrees
more energy efficient, since a precisely focused infrared laser
beam is all what it takes to transfer power between a charging
hotspot and client devices equipped with an add-on photovol-
taic cell.
Reached by eeNews Europe, Wi-Charge’s Co-Founder and
CEO Victor Vaisleib accepted to share more about the technol-
ogy and his company’s business roadmap.
“The term ‘wireless power’ is used for a very broad range of
different solutions without pointing out particular benefits for
the users”, acknowledges Vaisleib, “we like to describe what
we offer as ‘autonomous charging’, that is, recharging devices
without any user intervention.
According to the CEO, to be practical, a wireless power
infrastructure needs to be able to
deliver sufficient power over a suf-
ficient distance, without asking users
to get their devices to a specific
power charger or to a charging mat.
Components also need to be small.
Wi-Charge says it complies with all
these requirements, being much
more efficient than other long-dis-
tance RF-based wireless charging
solutions.
Appreciating the physics of wire-
less charging, Vaisleib notes: “Send-
ing energy over a distance, a beam
always diverges, which is a bad thing A wall-plug transmitter beaming out power to different with a particular pattern that allows
as too little power is going in the
devices.
right direction, and too much power
splits and goes elsewhere. With RF, only relatively humble dis-
tances can be achieved, with a power capacity about a fraction
of a Watt, limited by the maximum level of exposure allowed for
safety. Shorter wavelength beams maintain their integrity better. appropriate power levels.
With an infrared laser, we have a practically non-diverging beam
able to deliver its entire power content onto a small receiver”.
Trying to establish energy efficiency comparisons with the
24 News February 2018 @eeNewsEurope
WIRELESS CHARGING
likes of Powercast or Energous
proves difficult since manufactur-
ers never publish their specs. But
proceeding to a succinct reverse
engineering from bits of informa-
tion found in a recent FCC filing
from Energous, Vaisleib’s theo-
retical analysis yields that a 10W
transmitter would deliver about
0.1W of power at a distance of
about 75cm.
“Even with an ideal hardware
implementation, a simple analysis
of RF beam propagation in air points to inefficiency, without
even considering hardware power conversion inefficiencies” the
“With ultrasound, you get a few percentages of the initial
power reaching the target device, for RF, you get only a fraction
with that. If RF worked, we’d be happy to work with it. The
fundamental idea to adopt infrared was not a quick process,
we had it more than five years in development” told us Vaisleib
during a phone interview.
In stark contrast, the company says it can deliver several
watts of power over distances up to 10 meters, with very small
components, offering a solution that can be readily deployed as
a wireless charging infrastructure.
So, what’s in a transmitter, and how does a transmitter pre-
cisely identify and target devices to be charged in a room? We
asked.
“For aiming, as in all laser-based technologies or projectors,
nothing beats micro-mirrors. They
are the most cost-effective way to
aim a laser, and depending on the
application we can multiplex beams
or use different mirrors or both. For
larger devices such as smartphones,
we may focus a single beam whereas
for smaller devices such as sen-
sors or webcams, we can use one
beam sequentially polled to multiple
devices” revealed Vaisleib.
As for the identification and acqui-
sition of targets, the CEO explained
that each target receiver blinks back
the transmitter to discriminate the
power needs of individual receivers.
Built into the power transmitter, arrays of pin-diodes are ar-
ranged to triangulate and locate the different sources of signal,
mapping the receivers in the room so they can be targeted with
But you need an add-on receiver including the photovoltaic
cell, and it could take some time to convince end-users to buy
into this sort of infrastructure, we objected.
www.eenewseurope.com

“Partners are approaching us to integrate our solution into
all types of smart devices, for the wireless delivery of power
to wireless cameras, door locks, smart speakers and other
devices that are not yet as widespread
as smartphones and whose adop-
tion is stalled by battery life” the CEO
answered.
“We see OEMs come to us to inte-
grate the wireless power receivers into
their devices with a wireless power
supply that could be a wall-plug
transmitter beaming out to different
devices” he added.
First, the company wants to extend
its wireless power infrastructure Different implementations of Wi-Charge’s
through B2B partnerships in corpo-
transmitters and receivers.
rate offices, or public places such
as coffee shops and restaurants, where the receivers could
be integrated into different form factors. That could be smart-
phone cases or dongles for customers to use. Interestingly, one
practical example Vaisleib gave us was that of wireless charging
mats. In order to be integrated into furniture, those need to be
wired. But typically in coffee shops and restaurants, about 70 to
75% of actual seating area and tables is free standing, meaning
they can’t be wired easily. But you could have ceiling-mounted
transmitters beaming their power to charging mats integrated
into cable-free furniture.
“We are not announcing partnerships yet, but we are en-
gaged with several manufacturers. After a number of years,
once we have built an infrastructure, we see a mixture of busi-
ness models. Selling the hardware may not be the best of them.
We could offer monthly subscriptions for maintenance and
upgrades, in some business models, we may retain ownership
of the hardware and collect valuable data”.
“With a map of our infrastructure, we could tip people where
to go to charge their phone. The data collected would be sub-
ordinated to the needs of customers. We can tell if your battery
has depleted by 40%, end-users could decide what kind of
profile charge they want for their battery so it can last longer”.
Ultra-compact, fully
regulated and isolated
1 and 3 Watt
DC/DC converters.
Reliable. Available. Now.
www.eenewseurope.com
WIRELESS CHARGING
With such an infrastructure, mobile operators may want to
offer tier services only to their subscribers, for example offering
fast charging to one type of subscriber or less charge top-up
to others. A coffee shop or restaurant
could operate specific charge restric-
tions per table or per customers based
on loyalty programs. Power delivery
could be completely software-configu-
rable, so a casino operator could want
to provide only trickle charge to its
customers, so they would stay longer.
Of course, integration with lumi-
naires is an obvious choice in the long
term and eventually the company
wants to go that route.
“It fits well with what we do and
many lighting manufacturers come
to us with that request”, agreed the CEO, “but the technology
needs to mature a bit, it needs to be accepted by customers
before we can target consumers directly. That’s why the first
transmitter we showed at CES was in a bulb form factor with
LEDs, acknowledging the potential for luminaire integration”.
Asked if the laser beam from the power transmitter could
also serve to encode data for light communication (LiFi-style),
Vaisleib doesn’t want to confuse potential partners.
“If you look at our early patents, we also patented data con-
nection with those links, and technically, this possibility exists,
but commercially, we never found it beneficial as for now it is
still way inferior to radio links”.
If LiFi became mainstream in 20 years from now, then surely
Wi-Charge could integrate it, but for now it considers the ap-
plication too niche.
On its webpage, the company envisions the proliferation of
remote charging hotspots in homes, offices and in public areas
in a way similar to the adoption of WiFi, freeing mobile devices
from all cable connections. It lists in its product catalogue sev-
eral types of power transmitters, with power delivery capacities
ranging from 500mW from a 137x137x59mm unit to 2,500mW
for the LED luminaire version.
TRN 1(SM) und TRN 3(SM) Series
• Package occupies only 1 cm2
on the PCB
• Available in SIP package or
for SMD mount
• Wide input voltage ranges up
to 75 VDC includes the low voltage
range 4.5–13.2 VDC
• Operating temperature range
–40 °C to +90 °C
• I/O-isolation 1600 VDC,
continuous short circuit protection
www.tracopower.com
eeNews Europe February 2018 News 25
 NEWS & TECHNOLOGY
Wavy-channel TFTs drive more current, could increase pixel density
By Julien Happich
I
n a paper titled “Wavy Architecture Thin-Film Transistor for
Ultrahigh Resolution Flexible Displays”, published in the
Small journal, researchers from the King Abdullah Univer-
sity of Science and Technology (KAUST) unveiled a vertically
enhanced TFT architecture, where the channel presents multiple
vertical fins stretching longitudinally across the transistor’s
source and drain.
The 3D wavy structure, with 2µm tall and 3µm wide “fins”
repeated at a 6µm pitch give the channel a 70% larger opera-
tional width when compared with coplanar devices, allowing the
wavy-channel TFTs (WC-TFT) to drive currents 70% higher per
unit chip area.
By expanding the transistor’s width vertically, through the
implementation of grooved trenches, turn-on voltage or the OFF
current values remain the same as traditional TFTs while benefit-
ing from higher drive currents, the paper reports.
The researchers proved their new concept by fabricating
WC-TFTs on a silicon substrate with ZnO/Al2O3/Al layers for
the gate stack. Once peeled off and transferred onto a flexible
polydimethylsiloxane (PDMS) substrate, the transistors oper-
ated well with no degradation in their electrical performance or
cracks in the gate stack even when flexed down to a bending
radius of 5mm. “ON” current was only slightly lower, 10% lower
than when bent to a radius of 1cm.
Device schematics for both coplanar
TFT architecture (top) and the wavy The peeled off WC TFTs transferred onto
channel TFT architecture (bottom). a PDMS substrate.
KAIST weaves OLED fibres for wearables
By Julien Happich
R
esearchers from the Korea Advanced
Institute of Science & Technology (KAIST)
have produced highly efficient fibre-based
OLEDs by applying the successive layers of a
functional OLED stack around ultra-thin fibres
through a dip-coating process.
The researchers identified that solution
process planar OLEDs can be applied to fibres
without any reduction in performance through the
technology. The fibre OLEDs obtained boasted a
luminance in excess of 10,000 cd/m2 with a cur-
rent efficiency of 11cd/A. Fibre-based OLEDs woven
Even operated under tensile strains of up to
into knitted clothes
4.3%, the fibre OLEDs retained more than 90% of
their current efficiency. The researchers then wove and knitted
the fibre OLED into textiles and knitted clothes, and demon-
26 News February 2018 @eeNewsEurope
FLEXIBLE ELECTRONICS
The new transistors were also tested to drive a red LED
mounted on the flexible substrate. Compared to traditional pla-
nar TFTs, the WC TFTs were able to drive the LED at twice the
optical output power, 3 versus 1.5mW, under similar gate and
drain bias conditions for both devices.
Conversely, the WC TFTs could be scaled down so as to oc-
cupy 70% lower area compared with planar TFTs while deliver-
ing the same drive current, hence making it possible to achieve
a higher pixel fill factor for active-matrix LCD or OLED displays
of higher pixel density.
The researchers anticipate that further design optimization
of the WC TFT could lead to more than twice the drive current
per unit chip of area compared to planar TFTs, if given tighter
fin pitches and higher aspect ratio trenches. Although they
chose ZnO as a channel material because they could deposit
it uniformly via atomic layer deposition (ALD), the research-
ers note that their wavy architecture is compatible with other
channel deposition methods such as DC and radio frequency
(RF) magnetron sputtering, making it material agnostic and ap-
plicable to channel materials including indium gallium zinc oxide
(IGZO) and indium
zinc oxide (IGZO).
Side view SEM of the planar (top) and
wavy channel (bottom) TFTs with 2µm
tall 3µm wide “fins” at a 6µm fin pitch.
For the WTFT, the 3D wavy pattern
means a 70% larger operational width
when compared with coplanar devices.
strated the OLEDs remained functional, making
them applicable to flexible wearable displays.
The low temperature coating process (at
around 105℃) is applicable to many types of
fibres, with diameters ranging from 300 down to
90µm, highlight the researchers.
Professor Choi said, “Existing fiber-based
wearable displays had limitations for applicabil-
ity due to their low performance. However, this
technology can fabricate OLEDs with high perfor-
mance on fibers. This simple, low-cost process
opens a way to commercialize fiber-based wear-
able displays.”
This research led by a PhD candidate Seonil
Kwon was published online in the international journal for nano-
science, Nano Letters last December.
www.eenewseurope.com
 8-bit MCUs

8-bit PIC® and AVR® Microcontrollers

www.microchip.com/8bitEU
 Overview

As a designer of innovative products, you have a desire to make technology smart, more efficient and accessible to
everyone. Microchip has a passion for developing products and tools that make it easier for you to solve your design
problems and adapt to future needs. Our current lineup of PIC® and AVR® MCUs is the pinnacle of innovation in the
8-bit embedded space and incorporates the latest technologies to enhance system performance while reducing power
consumption and development time.

The PIC and AVR MCU brands represent the two dominant architectures in the embedded design universe. With more than
45 years of combined experience developing commercially available and cost-effective MCUs, Microchip is the supplier of
choice for thousands of customers worldwide. With a strong history of innovation in 8-bit MCUs, Microchip’s extensive
portfolio has the right product for any application.

History of Innovation

2 www.microchip.com/8bit
Unified Strength
Microchip’s portfolio of more than 1,200 8-bit PIC® and AVR® MCUs is not only the industry’s largest, but also offers superior
flexibility and performance in areas of utmost importance to modern embedded designers.

Leadership in Embedded World-class Development Longevity and Legacy

Technologies
• Core Independent Peripherals with
minimal CPU overhead
• Interconnectivity of analog and
digital peripherals
• Industry’s lowest power solutions
(XLP and picoPower® technologies)
• Robust touch sensing technologies
(mTouch® and QTouch®
technologies)
• Widest operating range: 1.6V to 5.5V
• Robustness in hostile environments
Experience
• Separate and dedicated tool
chains for both PIC and AVR MCU
development
• Intuitive graphical configuration
tools, IDEs, compilers and
development boards
• Extensive portfolio of libraries and
support tools
• Industry recognition for having the
best-in-class product non-
obsolescence operating model
• Best-in-class legacy support,
ensuring easy package, pin-out and
software migration

Core Independent Peripherals

Today’s cost-sensitive embedded designs require a paradigm shift away from the traditional “Bytes and Megahertz” approach.
Every embedded system contains a combination of common functions, such as power conversion, motor drive, sensor or
human interface and signal generation. By crafting the PIC and AVR MCU portfolios to suit your evolving needs, we can
support a wide variety of functions and applications.

Today’s PIC and AVR MCUs include on-chip Core

Independent Peripherals (CIPs) that can operate without
constant interaction from the CPU and have the ability
to communicate directly with other peripherals to create
flexible feedback loops. These “core independent”
blocks of configurable hardware intelligence require little
to no code, consume very little power and are much
smaller than the RAM and Flash needed to implement
the same function in software. Additionally, many
simultaneous functions can be enabled into a single
MCU. This leads to flexible, power-efficient 8-bit MCU
designs with the capability to perform the same tasks as
much larger and more expensive devices.

Core Independent Peripherals enable 8-bit PIC and AVR

microcontrollers to retain their class-leading low-power
performance while meeting or exceeding your design
requirements for functionality and performance.

8-bit PIC and AVR Microcontrollers 3

 Applications

A Solution for Every Application

Microchip’s 8-bit MCUs provide high-performance and power-efficient processing and are an ideal solution for a wide range
of applications, from control logic to fully integrated systems involving USB, Ethernet and LCD. Every device includes a range
of powerful peripherals including motor control timers, communication interfaces, powerful ADC and advanced on-chip safety
features.

Microchip’s online design centers provide all the necessary information to get started on a new design within a specific
application segment. Items found on the design centers include supporting products, application notes, reference designs
and development tools. Here are some design centers to help you with your next 8-bit design.

Automotive Displays Wireless Connectivity

8-bit microcontrollers are used in a
wide range of automotive applications
including remote keyless entry, lighting
control and instrument cluster to name
a few.
www.microchip.com/lin
www.microchip.com/can
www.microchip.com/automotive
Microchip microcontrollers offer
integrated LCD solutions that directly
drive segmented displays with letters,
numbers, characters and icons.
They are used in a wide variety of
applications, ranging from meters to
portable medical devices to thermostats
to exercise equipment.
www.microchip.com/lcd
Microchip helps you design cost-
effective, highly reliable wireless
products.
Microchip wireless technologies enable
innovative, scalable and dedicated
designs that fit small footprints,
consume very little power and operate
in rugged environments.
www.microchip.com/wireless

Motor Control Medical Touch and Gesture

Microchip’s PIC and AVR
® ®
microcontrollers contain innovative
motor control PWM peripherals
including complimentary waveforms
and dedicated time bases.
www.microchip.com/motorcontrol
Medical applications can benefit from
MCU peripheral features such as analog
signal conditioning, LCD control and
mTouch® input sensing.
www.microchip.com/medical
Microchip’s MCUs allow you to
integrate touch-sensing interfaces,
thereby reducing total system costs.
www.microchip.com/touch

4 www.microchip.com/8bit
Home Appliance
Microchip helps implement the new
features and functionality required for
your next appliance design with MCUs
integrated with fault-detecting hardware
for safety-critical applications. Microchip
MCUs may be used for motor control,
analog sensor measurements, front
panel keypad control and LCD/LED
displays.
www.microchip.com/homeappliance
Lighting
With advanced peripheral integration
and support for all lighting technologies,
a scalable Microchip solution provides
significant flexibility and innovation that
expands lighting product capabilities
and provides product differentiation.
www.microchip.com/lighting
Smart Energy
Microchip is well-positioned to
address various requirements of smart
metering systems with a combination
of discrete and highly integrated
system-on-chip solutions built around
the same processing platforms. This
will reduce time-to-market and enable
an unmatched level of flexibility in
addressing new requirements.
www.microchip.com/smartenergy

Ethernet Low Power USB

Microchip’s Ethernet solutions are Microchip offers low-power solutions Microchip offers USB solutions capable
designed to support the growing needs with eXtreme Low Power (XLP) of full-speed USB operation with the
of this market with a full complement technology with sleep currents down to PIC® and AVR® device families.
of robust, highly integrated Ethernet 9 nA. Our products are equipped with www.microchip.com/usb
devices. peripherals like USB, LCD and analog
www.microchip.com/ethernet which enables solutions with smaller
physical footprints and reduced bill of
material costs.
www.microchip.com/lowpower
8-bit MCU Quick Reference
Our flexible 8-bit PIC and AVR MCU portfolio offers a number of product families with varying levels of intelligent peripheral integration and
operating capabilities, enabling you to find the best MCU for your specific application and is summarized in the Quick Reference Guides.

The 8-bit MCU portfolio’s true strength lies in the vast array of flexible hardware peripherals
that are available to increase capability in any control system. Working together, our
selection of Core Independent Peripherals, Intelligent Analog, user interface peripherals
and on-board communications can enable system functions on MCUs with minimal
code footprint, reduced power consumption and accelerated time-to-market. Common
application functions like power and motor control, environmental sensing, system
management and user interface can be combined onto a single MCU to develop an
extremely cost-effective solution. For more details about Microchip products, please visit
our website at www.microchip.com/8bit.
8-bit PIC and AVR Microcontrollers 5
 Quick Reference Guide

8-bit PIC® Microcontroller Peripheral Integration

Program Flash Memory (KB)

Intelligent Analog Waveform C

Product
Family

ADC (# of bits)

DAC (# of bits)

HC I/O (mA)

10-bit PWM

16-bit PWM
SlopeComp
Data EE (B)

CCP/ECCP
Pin Count

HSComp
Comp

COG
PRG

ZCD
OPA
PIC10(L)F3XX 6 384–896 B HEF 8 ü

PIC16(L)F151X/2X 28–64 3.5–28 HEF 10 ü

PIC12LF1552 8 3.5 HEF 10

PIC16LF155X/6X 14–20 7–14 HEF 10 (2) ü

PIC16(L)F145X 14–20 14 HEF 10 ü ü

PIC1X(L)F157X 8–20 1.75–14 HEF 10 ü 5 ü

PIC16(L)F153XX 8–48 3.5–28 HEF 10 ü 5 ü ü 4

PIC1X(HV)F752/53 8–14 1.75–3.5 – 10 ü 5/9 50 ü ü ü ü

PIC1X(L)F1612/3 8–14 3.5 HEF 10 ü 8 ü ü

PIC16(L)F161X 14–20 7–14 HEF 10 ü 8 100 ü ü ü

PIC16(L)F170X 14–20 3.5–14 HEF 10 ü 5/8 ü ü ü ü ü

PIC16(L)F171X 28–40 7–28 HEF 10 ü 5/8 ü ü ü ü ü

PIC16(L)F176X/7X 14–40 7–28 HEF 10 ü 5/10 100 ü ü ü ü ü ü ü ü

PIC16(L)F183XX 8–20 3.5–14 256 10 ü 5 ü ü

PIC16(L)F188XX 28–40 7–56 256 10 (3) ü 5 ü ü ü

PIC16(L)F191XX 28–64 14–56 256 12(3) ü 5 ü ü ü

PIC18(L)FXXK40 28–64 16–128 256–1K 10 (3) ü 5 ü ü ü

PIC18(L)FXXK42 28–48 16–128 256–1K 12(3) ü 5 ü ü 4

PIC18(L)FXXJ94 64–100 32–128 – 12 ü ü

PIC18(L)FXXK50 20–40 8–32 256 10 ü 5 ü

PIC18(L)FXXK83(4) 28 32–64 1K 12(3) ü 5 ü ü ü

PIC18(L)FXXK90 60–80 32–128 1K 12 ü ü

Note 1: In addition to standard 8-bit and 16-bit timers 2: Independent Dual ADC Modules 3. ADCC: Analog-to-Digital Converter with Computation 4. CAN capable 5. LCD with Charge Pump

6 www.microchip.com/8bit
 Peripheral Function Focus
Logic Safety and Low Power and
Control Timing and Measurements (1) Communications User Interface
and Math Monitoring System Flexibility

UART with Protocols

16-bit PWM (16-bit)

EUSART/AUSART

mTouch® Sensing

IDLE/DOZE/PMD
USB with ACT
NCO (20-bit)

LIN Capable
SMT (24-bit)

CRC/SCAN
HLT (8-bit)

MathACC
TEMP/TS

DIA/MAP
AngTMR

DMA/VI
I2C/SPI
WWDT

HCVD
RTCC

MULT
CWG

DSM
NCO

CLC

LCD

PPS
HLT

ü ü ü ü ü ü

ü 2 2 ü ü

ü ü ü ü

ü ü 2 ü ü ü

ü ü ü ü ü ü ü

ü ü ü ü ü ü ü

ü ü ü ü ü ü ü ü 2 2 ü ü ü ü

ü ü ü

ü ü ü ü ü ü ü ü

ü ü ü ü ü ü ü ü ü ü ü ü ü ü ü

ü ü ü ü ü ü ü ü ü

ü ü ü ü ü ü ü ü ü ü

ü ü ü ü ü ü ü ü ü ü ü

ü ü ü ü ü ü ü 2 ü ü ü ü

ü ü ü ü ü ü ü ü ü ü ü ü 2 ü ü ü ü ü

ü ü ü ü ü ü ü ü ü ü ü ü ü ü(5) ü ü ü

ü ü ü ü ü ü ü ü 5 2 ü ü ü ü ü

ü ü ü ü ü ü ü ü ü ü ü ü ü 1 1 2 ü ü ü ü ü ü ü

ü ü 4 2 ü ü ü ü(5) ü ü

ü ü ü ü ü ü ü

ü ü ü ü ü ü ü ü ü ü 2 ü ü ü ü ü ü ü

ü ü 2 ü ü ü ü ü

8-bit PIC and AVR Microcontrollers 7


INTELLIGENT ANALOG: Sensor Interfacing and Signal Conditioning
ADC: Analog-to-Digital Converter
ADC2/ADCC: Analog-to-Digital
Converter with Computation
Comp: Comparator
DAC: Digital-to-Analog Converter
HC I/O: High-Current I/O
HSComp: High-Speed Comparator
OPA: Operational Amplifier
PRG: Programmable Ramp Generator
SlopeComp: Slope Compensation
VREF: Voltage Reference
ZCD: Zero Cross Detect
WAVEFORM CONTROL: PWM Drive and Waveform Generation
CCP/ECCP: (Enhanced) Capture
Compare PWM
COG:
Complementary Output Generator
CWG: Complementary Waveform
Generator
DSM: Data Signal Modulator
NCO: Numerically Controlled
Oscillator and 16-/20-bit Timer/
Counter
PWM: Pulse Width Modulation
16-bit PWM: Standalone 16-bit
PWM and 16-bit Timer/Counter
TIMING AND MEASUREMENTS: Signal Measurement with Timing and Counter Control
AngTMR: Angular Timer
HLT: Hardware Limit Timer and
8-bit Timer/Counter
NCO:
Numerically Controller Oscillator and
16-/20-bit Timer/Counter
RTCC: Real-Time Clock/Calendar
SMT: 24-bit Signal Measurement
Timer and 24-bit Timer/Counter
TEMP: Temperature Indicator
TS: Temperature Sensor
8-/16-bit Timer
16-bit PWM: Standalone 16-bit
PWM and 16-bit Timer/Counter
8 www.microchip.com/8bit
8-bit PIC® MCU Terminology
General purpose 8-/10-/12-bit ADC
General purpose 10-/12-bit ADC with automated analog signal analysis
(ex. oversampling, averaging, etc.)
General purpose rail-to-rail comparator
Programmable voltage reference with multiple internal and external connections
Up to 50 mA or 100 mA current drive on select I/O pins
General purpose rail-to-rail comparator with < 50 ns response time
General purpose op amp for internal and external signal source conditioning
Analog ramp generator (with slope compensation) for current/voltage mode power supplies
Slope compensation for Peak Current Mode power supplies
Stable fixed voltage reference for use with integrated analog peripherals
AC high-voltage zero-crossing detection for simplifying TRIAC control, synchronized switching
control and timing
1. CCP/ECCP: 10-bit PWM control with 16-bit capture and compare
2. ECCP: Addition of auto shutdown control
Automated complementary output with control of key parameters such as programmable rising/
falling edge events, polarity, phase, precision
dead-band, blanking and auto shutdown
Automated complementary output with control of key parameters such as
dead-band and auto shutdown
1. Modulates up to two carrier signals with digital data to create custom carrier synchronized output
waveforms
2. LED dimming engine functionality via interconnection with 10-/16-bit PWM, DSM and op amp
1. Precision linear frequency generator (@ 50% duty cycle) with 0.0001% step size of source input
clock frequency
2. General purpose 16-/20-bit timer/counter
General purpose 10-bit PWM control
1. High-resolution 16-bit PWM with edge- and center-aligned modes
2. General purpose 16-bit timer/counter
Phase angle timer for measurement and control of rotational and periodic events (ex. motor, AC
mains, TRIAC, etc.)
1. Hardware monitoring for missed periodic events and fault detection
2. General purpose 8-bit timer/counter with external reset capabilities
1. Precision linear frequency generator (@ 50% duty cycle) with 0.0001% step size of source input
clock frequency
2. General purpose 16-/20-bit timer/counter
Maintains accurate clock and calendar timing with external 32.768 kHz crystal
1. Accurate measurement of any digital signal including period, duty cycle, time of flight;
instantaneous vs. average measurements
2. General purpose 24-bit timer/counter
Provides relative temperature measurements utilizing the ADC
Provides linear relative temperature measurements utilizing the ADC with two factory-calibrated
reference values
General purpose 8-/16-bit timer/counter
1. High-resolution 16-bit PWM with edge- and center-aligned modes
2. General purpose 16-bit timer/counter
Learn more about 8-bit PIC Microcontrollers at www.microchip.com/8bit.
Learn more about Core Independent Peripherals (CIP) at www.microchip.com/CIP.
 LOGIC AND MATH: Customizable Logic and Math Functions
CLC: Configurable Logic Cell
MULT: Hardware Multiplier
MathACC: Math Accelerator
SAFETY AND MONITORING: Hardware Monitoring and Fault Detection
CRC/SCAN: Cyclical Redundancy
Check with Memory Scan
HLT: Hardware Limit Timer and
8-bit Timer/Counter
WWDT:
Windowed Watch Dog Timer
COMMUNICATIONS: General, Industrial, Lighting and Automotive
ACT: Active Clock Tuning for
Crystal-Free USB
CAN: Controller Area Network
LIN: Local Interconnect Network
EUSART/AUSART: Enhanced/
Addressable Universal
Asynchronous Receiver Transceiver
I2C: Inter-Integrated Circuit
SPI: Serial Peripheral Interface
UART: Universal Asynchronous
Receiver Transmitter
USB: Universal Serial Bus
USER INTERFACE: Capacitive Touch Sensing and LCD Control
HCVD: Hardware Capacitive
Voltage Divider
LCD: Liquid Crystal Display
LCD w/Charge Pump: Liquid
Crystal Display with Charge Pump
mTouch: Microchip Proprietary
Capacitive Touch Technology
LOW POWER AND SYSTEM FLEXIBILITY: XLP Low-Power Technology, Peripheral and Interconnects
DIA: Device Information Area
DMA: Direct Memory Access
DOZE: Power Saving Mode
HEF: High-Endurance Flash
IDLE: Power Saving Mode
MAP: Memory Access Partition
PMD: Peripheral Module Disable
PPS: Peripheral Pin Select
VI: Vectored Interrupts
XLP: eXtreme Low Power
Technology
PDF version available for download at www.microchip.com/8bitquickreference.
8-bit PIC and AVR Microcontrollers
1. Integrated combinational and sequential logic
2. Customer interconnection and re-routing of digital peripherals
MULTIPLY function of two 8-bit values with 16-bit result
1. MULTIPLY, ADD, ACCUMULATE functions of 8-/16-bit values with 35-bit result
2. Calculates a 16-bit PID function based on configurable Kp, Ki, Kd constants with a 34-bit result
1. Automatically calculates CRC checksum of Program/DataEE memory for NVM integrity
2. General purpose 16-bit CRC for use with memory and communications data
1. Hardware monitoring for missed periodic events and fault detection of external hardware
2. General purpose 8-bit timer/counter with external reset capabilities
System supervisory circuit that generates a reset when software timing anomalies are detected within
a configurable critical window
1. Auto-tuning of internal oscillator when connected to USB host (eliminates need for external crystal)
2. Tunes internal oscillator to match accuracy of external clock source
Industrial- and automotive-centric communication bus
1. Industrial- and automotive-centric communication bus
2. Support for LIN when using the EUSART
1. General purpose serial communications
2. Support for LIN when using the EUSART
General purpose 2-wire serial communications
General purpose 4-wire serial communications
Supports LIN master and slave, DMX, DALI and device protocols
Support for full-speed USB 2.0 device profiles
Simplifies implementation and reduces overhead of mTouch® sensing applications
Highly integrated segmented LCD controller
The LCD charge pump provides proper bias voltage and contrast for the LCD regardless of the Vdd
level.
1. Capacitive sensing for touch buttons and sliders
2. Capacitive sensing for system measurements and detection (ex. water level, intrusion detection,
etc.)
Dedicated memory area for data storage of temp sensor factory calibration values, factory ID and
FVR values for ADC and COMP
Moves data between memories and peripherals without CPU overhead, improving overall system
performance and efficiency
Ability to run the CPU core slower than the system clock used by the internal peripherals
128B Non-volatile data storage with high-endurance 100k E/W cycles
Ability to put the CPU core to sleep while the internal peripherals continue to operate from the
system clock
Customizable Flash partitioning with bootloader write protection option
Peripheral power disable hardware to minimize power consumption of unused peripherals
I/O pin remapping of digital peripherals for greater design flexibility and optimized
board layout
Offers faster and more predictable interrupt response times, with lower software overhead
XLP technology devices with extreme low-power operation modes for battery/low-power applications
9
 Quick Reference Guide

8-bit AVR® Microcontroller Peripheral Integration

Intelligent Analog Waveform C

Program Flash Memory (KB)

Product

Temperature Sensor
ADC (# of channels)
Family

ADC Gain Stage

Internal Voltage
ADC (# of bits)

DAC (# of bits)
Comparators

16-bit PWM
SRAM (KB)

8-bit PWM
Pin Count

Reference
ATtiny4/5/9/10 6 0.5–1 0.032 10³ 4(3) ü 2
ATtiny102/104 8/14 1 0.032 10 5/8 ü ü 2
ATtiny13A 8–20 1 0.064 10 4 ü 2
ATtiny20/40 12–20 2/4 0.128/0.256 10 8/12 ü ü 2 2
ATtiny24A/44A/84A 14–20 2–8 Up to 0.512 10 8 ü ü ü ü 2 2
ATtiny25(V)/45(V)/85(V) 8–20 2–8 Up to 0.512 10 4 ü ü ü ü 4
ATtiny48/88 28–32 4/8 Up to 0.512 10 8 ü ü ü 1 1
ATtiny87/167 20–32 8/16 0.512 10 11 ü ü ü 1 2
ATtiny261A/461A/861A 20–32 2–8 Up to 0.512 10 11 ü ü ü ü
ATtiny21x/41x/81x/161x/321x 8–24 2–32 Up to 2 12 12 ü 8 ü ü 2
ATtiny441/841 14–20 4/8 Up to 0.512 10 12 ü ü ü 1 2
ATtiny1634 20 16 1 10 12 ü ü ü 2 2
ATtiny2313A 20 2 0.128 – – ü ü 2 2
ATmega8A/16A/32A 28–44 8–32 1–2 10 8 ü 2 1
ATmega8U2/16U2/32U2 32 8–32 0.5–1 – – ü ü ü 4 6
ATmega16U4/32U4 32 16/32 1/2 10 12 ü ü ü 5
ATmega48PB/88PB/168PB/328PB 32 4–32 0.5–2 10 8 ü ü ü 4 2/6(6)
ATmega64A/128A 64 64–128 4 10 8 ü ü ü 2 6
ATmega164PA/324PA/644PA/1284P 44 16–128 1–16 10 8 ü ü ü 4 2/2/4
ATmega165PA/325PA/645P 44 16–64 1–4 10 8 ü ü 4 6
ATmega169PA/329PA/649P 64 16–64 1–4 10 8 ü ü 2 2
ATmega324PB 44 32 2 10 8 ü ü 2 2
ATmega640/1280/2560/1281/2561 64–100 64–256 8 10 8/16 ü ü ü 4 6/12
ATmega3290PA/6490P 100 32–64 2–4 10 8 ü ü ü 2 2
ATmega3250PA/6450P 100 32–64 2–4 10 8 ü ü ü 2 2
ATxmega A1U Family 100 64–128 4–8 12 16 ü ü 12 ü ü 8 ü
ATxmega A3U Family 64 64–256 4–16 12 16 ü ü 12 ü ü 7 ü
ATxmega A4U Family 44–49 16–128 2–8 12 12 ü ü 12 ü ü 5 ü
ATxmega B1/B3 Family 64–100 64–128 4–8 12 8 ü ü ü ü 2/3 ü
ATxmega C3/D3 Family 64 32–384 4–32 12 16 ü ü ü ü 5 ü
ATxmega C4/D4 Family 44–49 16–128 2–8 12 12 ü ü ü ü 4 ü
ATxmega E5 Family 32 8–32 1–4 12 16 ü ü 12 ü ü 3 ü
1: LIN port also 2: Peripheral Touch Controller 3: Only on the ATtiny5/10 4: Not on the ATtiny212/214/412/414/416 5: Only on the ATmega1281/2561 6: Only on the ATmega328PB 7: Only on the C3 an

10 www.microchip.com/8bit
 Peripheral Function Focus
Logic,
Timing and Safety and User
Control Crypto and Communications System Flexibility
Measurements Monitoring Interface
Math

picoPower® Technology
External Bus Interface
8-bit Timer/Counters

12-bit Timer Counter

16-bit Timer/Counter
Quadrature Decoder

QTouch® Technology
Waveform Extension

QTouch Technology
Real-Time Counter

Crypto (AES/DES)

DMA Channels
Event System
SleepWalking
Sleep Modes
with PTC (2)
IRCOM
USART
MULT

UART
WDT

LCD
BOD
CRC
POR

USB
CCL

SPI
I2C
1 ü ü ü 4
2 ü ü 1 4
ü ü ü ü 3 ü
1 1 ü ü ü 1 1 ü 4
1 1 ü ü ü ü 1 1 ü 4 ü
2 ü ü ü ü 1 1 ü 3
1 1 ü ü ü 1 1 3 ü
1 1 ü ü ü ü 1 (1)
1 2 4
1 1 ü ü ü 1 1 ü 4 ü
ü 1 1 ü ü ü ü ü ü 1 (1)
1 1 ü (4) ü ü 3 ü
1 2 ü ü ü 2 1 1 4 ü
1 1 ü ü ü ü 2 1 ü 4 ü
1 1 ü ü ü ü 1 1 2 3 ü
ü 2 1 ü ü ü ü 1 1 1 ü 5
ü 2 3 ü ü ü ü 2 ü 2 2 ü 6
1 1 ü ü ü ü 1 ü 1 6
ü 2 1/3(6) ü ü ü ü 1/2(6) 1/2(6) 1/2(6) ü 1(6) 6
2 2 ü ü ü ü 2 1 1 ü 6
ü 2 1/1/2 ü ü ü ü 2 1 1 ü 6 ü
ü 2 3 ü ü ü ü 3 2 2 ü 6 ü
ü 2 1 ü ü ü ü 1 1 1 ü ü 5
ü 2 1 ü ü ü ü 1 1 1 ü ü 5
ü 2 4 ü ü ü ü 2/4 1 1 ü ü(5) 6
ü 2 1 ü ü ü ü 1 1 1 ü ü 5
ü 2 1 ü ü ü ü 1 1 1 ü 5
ü ü ü 8 ü ü ü ü ü ü 8 ü 4 4 ü ü ü 4 ü 5 ü
ü ü ü 7 ü ü ü ü ü ü 7 ü 2 3 ü ü 4 ü 5 ü
ü ü ü 5 ü ü ü ü ü ü 5 ü 2 2 ü ü 4 ü 5 ü
ü ü ü 2/3 ü ü ü ü ü ü 1/2 ü 1 1 ü ü ü 2 ü 5 ü
ü ü ü 5 ü ü ü ü ü 3 ü (7)
2 2 ü ü ü 5 ü
ü ü ü 4 ü ü ü ü ü 2 ü(7) 2 2 ü ü ü 5 ü
ü ü ü 3 ü ü ü ü ü ü 2 1 1 ü ü 4 ü 5 ü
nd C4

8-bit PIC and AVR Microcontrollers 11

 8-bit AVR® MCU Terminology

INTELLIGENT ANALOG: Sensor Interfacing and Signal Conditioning

ADC: Analog-to-Digital Converter General purpose 10-/12-bit ADC
ADC Gain Stage: Analog-to-Digital Programmable gain stage, providing amplification steps on the differential
Converter Gain Stage input voltage
Comp: Comparator General purpose rail-to-rail comparator
Programmable voltage reference with multiple internal and external
DAC: Digital-to-Analog Converter
connections
VREF: Voltage Reference Stable fixed voltage reference for use with integrated analog peripherals
WAVEFORM CONTROL: PWM Drive and Waveform Generation
PWM: Pulse Width Modulation General purpose 10-bit PWM control
16-bit PWM: Standalone 16-bit 1. High-resolution 16-bit PWM with edge- and center-aligned modes
PWM and 16-bit Timer/Counter 2. General purpose 16-bit timer/counter
TIMING AND MEASUREMENTS: Signal Measurement with Timing and Counter Control
8-/12-/16-bit Timer General purpose 8-/12-/16-bit timer/counter
LOGIC, CRYPTO AND MATH: Customizable Logic and Math Functions
1. Integrated combinational and sequential logic
CCL: Configurable Custom Logic
2. Customer interconnection and re-routing of digital peripherals
MULT: Hardware Multiplier MULTIPLY function of two 8-bit values with 16-bit result
Data encryption and decryption can be easily performed for both
Crypto (AES/DES)
internally stored data or for small external data packets
SAFETY AND MONITORING: Hardware Monitoring and Fault Detection
CRC/SCAN:
Automatically calculates CRC checksum of Program/DataEE memory for
Cyclical Redundancy Check
NVM integrity
with Memory Scan

Peripheral Spotlight
Peripheral Touch Controller
The powerful tiny ONE series (tiny416/817/1617) MCUs feature the Peripheral Touch
Controller (PTC), which enables you to easily add capacitive buttons and sliders
to your embedded application. The PTC autonomously acquires and processes
capacitive touch inputs without external components. On-chip hardware based noise
filtering, moisture tolerance and zero-drift temperature compensation ensure reliable
sensor operation in the harshest of operating climates. Additionally, the PTC’s core-
independent operation gives you the ability to reduce your design’s power consumption
with the wake-on-touch or wake-on-proximity features. No matter the requirements,
your next touch-based solution just became easier with the Peripheral Touch Controller.

12 www.microchip.com/8bit
 COMMUNICATIONS: General, Industrial, Lighting and Automotive
USART: Universal Asynchronous 1. General purpose serial communications
Receiver Transceiver 2. Support for LIN/IrDA®
I2C: Inter-Integrated Circuit General purpose 2-wire serial communications
SPI: Serial Peripheral Interface General purpose 4-wire serial communications
IRCOM: Infrared Communication Encodes and decodes data according to the IrDA communication
Module protocol
USER INTERFACE: Capacitive Touch Sensing and LCD Control
LCD: Liquid Crystal Display Highly integrated segmented LCD controller
QTouch®: Microchip Proprietary
Provides a simple-to-use solution to realize touch-sensitive interfaces
Touch Technology
QTouch with PTC: QTouch with Provides a simple-to-use solution to realize touch-sensitive interfaces
Peripheral Touch Controller with a Peripheral Touch Controller
LOW POWER AND SYSTEM FLEXIBILITY: Low-Power Technology, Peripheral and Interconnects
Moves data between memories and peripherals without CPU overhead,
DMA: Direct Memory Access
improving overall system performance and efficiency
Flexible routing of peripheral events, ability to control peripheral
Event System
independent from the CPU
Highly flexible module for interfacing external memories and memory-
External Bus Interface
addressable peripherals
picoPower® Technology Low-power technology
Low-power saving modes, IDLE, power-down, power-save, standby and
Sleep Modes
extended standby
SleepWalking Ability to put the CPU core to sleep until a relevant event occurs

Peripheral Spotlight
Analog-to-Digital Converter with Computation
Many embedded applications require the measurement of analog voltages, along
with a significant amount of post-processing and signal analysis. Typically, these
tasks place a heavy burden on MCU hardware and can significantly lengthen
your code development cycle. The Analog-to-Digital Converter with Computation
(ADC2) peripheral, available on several PIC® MCU families, implements a suite of
post-processing features in hardware, including low-pass filtering, oversampling
and averaging. The ADC2 also automates signal acquisition tasks and adds new
capabilities, such as continuous mode retrigger (which waits for a certain threshold
before triggering a conversion) and auto conversion (which allows triggering a conversion based on variety of internal and
external sources) to significantly reduce CPU load and software development time. With the ADC2, you don’t have to spend
valuable cycles performing expensive computations; this Core Independent Peripheral does the heavy lifting for you.

8-bit PIC and AVR Microcontrollers 13

 8-bit PIC® and AVR® MCU Families

8-bit PIC MCU Product Families

Microchip’s 8-bit PIC microcontrollers fall into three product architecture categories providing a variety of options for any
application requirement. All device families have low power capability, flexible Flash program memory and instructions and
data on separate buses.

PIC10 MCUs
PIC10 microcontrollers have a simple instruction set for ease of use and quick development. They offer the smallest form
factor that can be easily implemented into the most space constrained designs. This microcontroller family includes an internal
oscillator, a comparator, ADC and data memory for data-logging, giving engineers the ability to add smarts in various entry-
level applications. Its low cost and easy implementation can also help designers to replace multiple discrete components with
an integrated MCU solution, resulting in additional savings in board space.

PIC12/PIC16 MCUs
PIC12/PIC16 microcontrollers are ideal for many applications that require a higher level of embedded control and more
memory. These peripheral-rich devices feature various serial analog and digital peripherals, such as: SPI, I2C, USART, LCD
and ADCs. With low sleep and dynamic current consumption, these MCUs are highly suited for energy efficient and battery
powered applications.

PIC18 MCUs
Offering the largest pin count and memory size, PIC18 microcontrollers combine the maximum level of performance and
integration with the ease-of-use of an 8-bit architecture. With up to 16 MIPS of processing power, PIC18 microcontrollers
feature advanced peripherals such as CAN, USB, Ethernet, touch sensing and LCD display drivers. The PIC18 family has a
flexible range of self-programming Flash memory from 4 KB to 128 KB and up to 8 KB RAM. Many PIC18 products conserve
power with XLP technology and specialized clocking and sleep modes for battery applications.

8-bit AVR Product Families

Microchip’s 8-bit AVR MCUs are an excellent choice for a wide variety of embedded systems. These devices offer a unique
combination of performance, power efficiency and design flexibility. Optimized to reduce development time, they are based on
the industry’s most code-efficient architecture for C and assembly programming.

tinyAVR® MCUs
The tinyAVR MCU family is optimized for applications that require performance, power efficiency and ease of use in a small package.
All tinyAVR devices are based on the same architecture and compatible with other AVR® devices. The integrated ADC, DAC,
comparators, EEPROM memory and brown-out detector let you build applications without adding external components. tinyAVR
devices also offer Flash memory and on-chip debug for fast, secure and cost-effective in-circuit upgrades that significantly cut your
time to market.

megaAVR®
The megaAVR family is the ideal choice for designs that need some extra muscle. For applications requiring large amounts
of code, megaAVR devices offer substantial program and data memories with performance up to 20 MIPS. Meanwhile,
innovative picoPower technology helps minimize power consumption. All megaAVR devices offer self-programmability for fast,
secure and cost-effective in-circuit upgrades. You can even upgrade the Flash memory while running your application.

XMEGA® MCUs
The XMEGA family of MCUs delivers the best possible combination of real-time performance, high integration and low power
consumption for 8/16-bit MCU applications.

14 www.microchip.com/8bit
 Development Tools

Software Development
Dedicated PIC® and AVR® MCU Tool Chains
Microchip Technology offers two best-in-class tool chains for PIC and AVR MCUs. These separate and dedicated development
ecosystems will help streamline your next design, from prototype to production. Microchip offers easy-to-use software and
hardware development resources to accelerate your next PIC or AVR MCU-based design. The PIC microcontrollers are finding
their way into new applications like smartphones, audio accessories, video gaming peripherals and advanced medical devices.
Microchip provides solutions for the entire performance range of 8-bit microcontrollers, with easy-to-use development tools,
complete technical documentation and post design-in support through a global sales and distribution network.

PIC MCUs AVR MCUs

Intuitive Graphical
Configuration Tools

Free IDE and

Compilers

Simplified
Development
Environments

Rapid Prototyping Options

Microchip’s MPLAB® Xpress and Xplained Development boards make it easy to get started building your application with PIC
and AVR MCUs. Both board series feature integrated programmers to help you get up and running as quickly as possible while
minimizing startup cost. Xpress and Xplained boards are available with a wide variety of on-board MCUs, making them an
excellent entry point for those who are new to PIC or AVR MCU architectures.

MPLAB Xpress Evaluation Board (DM164140) (PIC MCUs)

The Xpress Evaluation Board features an on-board drag-and-drop programmer, a mikroBUSTM expansion header and
the popular PIC16F18345 MCU for easy development with the latest generation of PIC MCUs. The board integrates
seamlessly with the free, cloud-based MPLAB Xpress IDE (www.microchip.com/xpress).

ATTiny817 Xplained Mini (ATTINY817-XMINI) (AVR MCUs)

The tiny817 Xplained Mini Evaluation Kit is a hardware platform for evaluating tiny817, tiny816, tiny814 and
tiny417 microcontrollers. The evaluation kit comes with a fully integrated debugger that provides seamless integration
with Atmel START and Atmel Studio development environment.

Other Popular Development Boards

PIC MCUs AVR MCUs
MPLAB Xpress Evaluation Boards (DM164141/2/3) Xplained Nano (ATTINY-XNANO)
8-bit Curiosity Development Board (DM164137) Xplained Pro (ATMEGA324PB-XPRO)
Explorer 8 Development Board (DM160228) Xplained Mini (ATMEGA168PB-XMINI) (ATMEGA328PB-XMINI)
PICDEM™ Lab II Development Platform (DM163046) STK600 Development Board (ATSTK600)

8-bit PIC and AVR Microcontrollers 15

Support Training
Microchip is committed to supporting its customers in If additional training interests you, Microchip offers several
developing products faster and more efficiently. We maintain a resources including in-depth technical training and reference
worldwide network of field applications engineers and technical material, self-paced tutorials and significant online resources.
support ready to provide product and system assistance. For • Overview of Technical Training Resources:
more information, please visit www.microchip.com: www.microchip.com/training
• Technical Support: www.microchip.com/support • MASTERs Conferences:
• Evaluation samples of any Microchip device: www.microchip.com/masters
www.microchip.com/sample • Developer Help Website:
• Knowledge base and peer help: www.microchip.com/developerhelp
www.microchip.com/forums • Technical Training Centers:
• Sales and Global Distribution: www.microchip.com/sales www.microchip.com/seminars

Sales Office Listing

AMERICAS EUROPE EUROPE ASIA/PACIFIC ASIA/PACIFIC
Atlanta, GA Austria - Wels Israel - Ra’anana Australia - Sydney China - Zhuhai
Tel: 678-957-9614 Tel: 43-7242-2244-39 Tel: 972-9-744-7705 Tel: 61-2-9868-6733 Tel: 86-756-321-0040
Austin, TX Denmark - Copenhagen Italy - Milan China - Beijing India - Bangalore
Tel: 512-257-3370 Tel: 45-4450-2828 Tel: 39-0331-742611 Tel: 86-10-8569-7000 Tel: 91-80-3090-4444
Boston, MA Finland - Espoo Italy - Padova China - Chengdu India - New Delhi
Tel: 774-760-0087 Tel: 358-9-4520-820 Tel: 39-049-7625286 Tel: 86-28-8665-5511 Tel: 91-11-4160-8631
Chandler, AZ (HQ) France - Paris Netherlands - Drunen China - Chongqing India - Pune
Tel: 480-792-7200 Tel: 33-1-69-53-63-20 Tel: 31-416-690399 Tel: 86-23-8980-9588 Tel: 91-20-3019-1500
Chicago, IL France - Saint Cloud Norway - Trondheim China - Dongguan Japan - Osaka
Tel: 630-285-0071 Tel: 33-1-30-60-70-00 Tel: 47-7289-7561 Tel: 86-769-8702-9880 Tel: 81-6-6152-7160
Dallas, TX Germany - Garching Poland - Warsaw China - Guangzhou Japan - Tokyo
Tel: 972-818-7423 Tel: 49-8931-9700 Tel: 48-22-3325737 Tel: 86-20-8755-8029 Tel: 81-3-6880-3770
Detroit, MI Germany - Haan Romania - Bucharest China - Hangzhou Korea - Daegu
Tel: 248-848-4000 Tel: 49-2129-3766-400 Tel: 40-21-407-87-50 Tel: 86-571-8792-8115 Tel: 82-53-744-4301
Houston, TX Germany - Heilbronn Spain - Madrid China - Hong Kong SAR Korea - Seoul
Tel: 281-894-5983 Tel: 49-7131-67-3636 Tel: 34-91-708-08-90 Tel: 852-2943-5100 Tel: 82-2-554-7200
Indianapolis, IN Germany - Karlsruhe Sweden - Gothenberg China - Nanjing Malaysia - Kuala Lumpur
Tel: 317-773-8323 Tel: 49-721-62537-0 Tel: 46-31-704-60-40 Tel: 86-25-8473-2460 Tel: 60-3-6201-9857
Tel: 317-536-2380 Germany - Munich UK - Wokingham China - Qingdao Malaysia - Penang
Los Angeles, CA Tel: 49-89-627-144-0 Tel: 44-118-921-5800 Tel: 86-532-8502-7355 Tel: 60-4-227-8870
Tel: 949-462-9523 Germany - Rosenheim China - Shanghai Philippines - Manila
Tel: 951-273-7800 Tel: 49-8031-354-560 Tel: 86-21-3326-8000 Tel: 63-2-634-9065
Raleigh, NC China - Shenyang Singapore
Tel: 919-844-7510 Tel: 86-24-2334-2829 Tel: 65-6334-8870
New York, NY China - Shenzhen Taiwan - Hsin Chu
Tel: 631-435-6000 Tel: 86-755-8864-2200 Tel: 886-3-577-8366
San Jose, CA China - Wuhan Taiwan - Kaohsiung
Tel: 408-735-9110 Tel: 86-27-5980-5300 Tel: 886-7-213-7830
Tel: 408-436-4270
China - Xiamen Taiwan - Taipei
Canada - Toronto Tel: 86-592-2388138 Tel: 886-2-2508-8600
Tel: 905-695-1980
China - Xian Thailand - Bangkok
Tel: 86-29-8833-7252 Tel: 66-2-694-1351
10/28/16

www.microchip.com

Microchip Technology Inc. | 2355 W. Chandler Blvd. | Chandler AZ, 85224-6199

The Microchip name and logo, the Microchip logo, AVR, megaAVR, MPLAB, PIC, picoPower, QTouch, tinyAVR and XMEGA are registered trademarks and PICDEM is a trademark of Microchip
Technology Incorporated in the U.S.A. and other countries. mTouch is a registered trademark of Microchip Technology Inc in the U.S.A. All other trademarks mentioned herein are property of their
respective companies. © 2017, Microchip Technology Incorporated. All Rights Reserved. DS30009630L. ML3120Eng08/17

Nanolaminate boost for flexible screens and printed electronics
By Nick Flaherty
R
esearchers at Georgia Tech in the US believe they have
come up with a way to build organic thin film transistors
that are more stable and reliable. This could be a signifi-
cant move for flexible screens and printed electronics.
The nanostructured gate dielectric is composed of a fluo-
ropolymer layer followed by a nanolaminate
made from two metal oxide materials. This
serves as gate dielectric and simultaneously
protects the organic semiconductor – which
had previously been vulnerable to damage from
the environment – and enables the transistors
to operate with significantly higher stability than
inorganic versions, including underwater.
“We have now proven a geometry that yields
lifetime performance that for the first time es-
tablish that organic circuits can be as stable as
devices produced with conventional inorganic technologies,”
said Bernard Kippelen, the Joseph M. Pettit professor in Geor-
gia Tech’s School of Electrical and Computer Engineering (ECE)
and director of Georgia Tech’s Center for Organic Photonics and
Electronics (COPE). “This could be the tipping point for organic
thin-film transistors, addressing long-standing concerns about
the stability of organic-based printable devices.”
A key part of the development is that the dielectric layer uses
two components, a fluoropolymer and a metal-oxide layer.
“When we first developed this architecture, this metal oxide
layer was aluminium oxide, which is susceptible to damage
from humidity,” said Canek Fuentes-Hernandez, senior research
scientist. “We developed complex nanolaminate barriers which
could be produced at temperatures below 110 degrees Celsius
and that when used as gate dielectric, enabled transistors to
sustain being immersed in water near its boiling point.”
The architecture uses alternating layers of aluminium oxide
and hafnium oxide, five of each repeated 30 times on top of
the fluoropolymer, to make the dielectric. The oxide layers are
Europe wants modular LED luminaires for a circular economy
By Julien Happich
A
s part of the European Commission’s Hori-
zon 2020 work programme, the Repro-light
project (Re-usable and re-configurable parts
for sustainable LED based lighting systems - www.
repro-light.eu) aspires to initiate a transformation in
the European LED lighting industry, by creating the
‘Luminaire of the Future’.
Led by representatives and driving forces from the European
lighting industry, as well as manufacturers, experts on lighting
sustainability and the Social Sciences, the Repro-light consor-
tium aims to harness innovative technologies and materials to
implement a never attempted before modular luminaire archi-
tecture, a smart production scheme and also the development
of a reconfigurable customised LED luminaires.
The LED market is currently mid-transformation, shifting its
focus from energy efficiency to function values. This project is
perfectly timed to capitalise and lead this transformation, as it is
set to change the reputation of the LED luminaire from a dispos-
able object, into a customisable and sustainable product, with a
www.eenewseurope.com
FLEXIBLE ELECTRONICS
produced with atomic layer deposition (ALD) and the resulting
nanolaminate is 50nm thick and essentially immune to the ef-
fects of humidity.
“While we knew this architecture yielded good barrier proper-
ties, we were blown away by how stably transistors operated
with the new architecture,” said Fuentes-Her-
nandez. “The performance of these transistors
remained virtually unchanged even when we
operated them for hundreds of hours and at
elevated temperatures of 75°C. This was by far
the most stable organic-based transistor we
had ever fabricated.”
For the laboratory demonstration, the
researchers used a glass substrate, but many
other flexible materials – including polymers
and even paper – could also be used.
In the lab, the researchers used standard ALD growth
techniques but newer processes such as spatial ALD that uses
multiple heads with nozzles delivering the precursors could
speed up production and allow the devices to be scaled up in
size. “ALD has now reached a level of maturity at which it has
become a scalable industrial process, and we think this will al-
low a new phase in the development of organic thin-film transis-
tors,” said Kippelen.
The nanolaminate technique could allow development of
inexpensive paper-based devices, such as smart tickets, that
would use antennas, displays and memory fabricated on paper
through low-cost processes. It can also be applied to flexible
OLED displays.
The demonstration used a model organic semiconductor with
a relatively low carrier mobility of 1.6cm2/Vs. As a next step, the
researchers are looking at organic semiconductors that provide
higher charge mobility and testing the nanolaminate under dif-
ferent bending conditions, across longer time periods, and in
other device platforms such as photodetectors.
high function value.
The consortium is made up of European leading
experts including German LED luminaire manu-
facturer Trilux, but also BJB GmbH, Italian design
and prototyping SME Grado Zero Espace S.r.l.,
and Rohner Engineering. Also taking part are in-
novative members of the lighting industry such as
Austrian lighting design and research consultancy Bartenbach
GmbH and founder, owner and publisher of LED professional,
Luger Research. Experts on lighting sustainability and Life Cycle
Assessment such as the Catalonia Institute of Energy Research
(IREC) and Mondragon University who are prominent in Social
Sciences complete the consortium.
The need for projects like Repro-light has been highlighted in
several recent white papers, including “Serviceable Luminaires
in a Circular Economy” published by The European Lighting
Association. This white paper highlights the value of a circular
economy, and it is just such an economy the ‘Luminaire of the
Future’ is design to be part of.
eeNews Europe February 2018 News 27
 DESIGN & PRODUCTS 3D INTEGRATION & CIRCUIT PACKAGING

How tunable human-centric lighting is

placing greater thermal demands on LEDs
By Mike Edwards

T
unable lighting — or human-centric lighting —
is one of the biggest trends hitting the lighting
industry, and it’s easy to see why. Both re-
search and user feedback shows the right light style,
colour, warmth and brightness can have tangible
benefits to people in different contexts affecting
behaviour, mood and concentration.
For example, in education, a cool blue or white
light can help boost students’ (and teachers’) at-
tentiveness after lunch when enthusiasm and energy Fig. 1: Packaged versus chip-scale packaged (CSP) LED.
tend to dip. Conversely, a warmer orange light will
help calm a classroom of rowdy children. the encapsulant, which is there for protection, so typically it
In hospitality, restaurants frequently take advantage of basic must escape via the back of the LED, travelling through multiple
tunable lighting, creating different ambiences at different times layers of material that spread the heat to mitigate the danger.
of day for guests — without needing a complex, expensive
multimode lighting system. As seen on figure 2, the effect of a sub-mount or package
And in countries that experience extremely high tempera- base is to spread the heat from the LED over a larger area
tures in the summer particularly the Middle East and extremely before it reaches the metal-clad printed circuit board (MCPCB).
low temperatures in the winter, like Switzerland or Canada, This means the MCPCB is not challenged to deal with the cool-
more and more buildings feature tunable lighting that offers ing the point heat source of the LED junction.
bluer light in the summer to create a cooler atmosphere and But with CSPs, manufacturers have stripped out the sub-
orange light in the winter to create a feeling of warmth. mount or package base (the finished CSP must be no more
While the move towards tunable lighting benefits occupants than 20% larger than the chip itself to qualify as a CSP), which
and customers, the technical challenges involved with creat- means that the LED now produces the same amount of heat but
ing LEDs that deliver tunable lighting is setting LED engineers in a much smaller environment. And as a double whammy, the
challenges around cost, performance, reliability and manufac- sub-mount/package base, which the CSP LED has now lost,
turability. used to act as a great heat spreader, therefore, reducing the
intensity of the heat and the risk.
Turning towards CSP LEDs
To support tunable lighting requires multiple LEDs
emitting different wavelengths. This requires many
more LEDs in each bulb. One of the essential modi-
fications that engineers must make to the usually
deployed standard ‘packaged’ LED is to strip out
certain parts to make the LED component smaller,
thus creating a chip-scale packaging (CSP) LED
where each device is less than 1.2x than its core
LED functional element.

Creating a CSP itself is by no means the only

challenge — another is how to deal with the intense
heat problem caused by the small size. Some CSPs
now exceed 10W/mm2. What most engineers know,
but consumers do not, is that LEDs tend to be only
around 40% efficient, which means the LED wastes
60% of its input power as heat. As with any elec-
tronic device, too much heat can negatively affect Fig. 2: Heat flow through a standard packaged LED.
performance criteria such as light quality and can
cause serious damage to the fabric of the LED so needs remov- The result is an LED that’s also an intense point heat source
ing as quickly as possible. that poses risks the light quality and the lifespan of any LED
With standard packaged LEDs, removing this heat is a much bulb it is added to. And nobody wants to buy a lighting system
easier task than with CSPs. Heat cannot escape with through that doesn’t live up to its full potential.
As shown on figure 3, heat transport though a MCPCB is a
Mike Edwards is VP of sales and marketing at Cambridge combination of radial spreading in the copper and axial conduc-
Nanotherm – www.camnano.com tion through the copper, dielectric and aluminium.

28 News February 2018 @eeNewsEurope www.eenewseurope.com


Fig. 3: Heat flow through a CSP LED.
To a very simplistic approximation, optimum heat flux is ob-
tained when these vectors are equal. This means if the dielectric
of the MCPCB is not a good thermal conductor, then the copper
tracking must be 70–105µm thick (2 or 3 Oz) and large in area to
provide effective cooling.
A second problem with removing the sub-mount/package
base is that it shifts the responsibility of thermal management
from the LED packaging manufacturers onto the luminaire
integrators and overall light-engine designers. This is something
have never had to deal with in their MCPCB solutions up until
now.
Traditional CSP thermal
management techniques
Traditionally, the only option open to level 2 inte-
grators was to mount the CSP LEDs onto better
thermally performing PCBs — usually metal-clad
PCBs (MCPCBs) that offer high thermal conductiv-
ity. MCPCBs are typically constructed from a solid
sheet of aluminium (occasionally copper) usually
1.5mm thick with a 70µm-thin sheet of copper glued
on with an epoxy dielectric.
This epoxy is often filled with particles of a
thermally conductive material like aluminium nitride
(AIN) to increase its thermal performance without Fig. 4: Heat flow through a Nanotherm MCPCB.
affecting the electrically isolating properties of the
adhesive.
At best, this approach to manufacturing MCPCBs results
in a thermal transfer away from these point heat sources of
around 100W/mK in the best-performing products — often far
less in practice. Whilst this performance is perfectly acceptable
for most low-count LED modules the challenge with CSPs is
exacerbated by the number of LEDs in close proximity and the
way they emit heat.
CSPs are a point heat source — their small size mean they
quickly saturate any substrate that doesn’t have the requisite
thermal conductivity, quickly leaving the LED vulnerable to over-
heating and failure.
There’s clearly a requirement for a more cost-effective board-
level thermal management alternative that can deal with the
thermal and optical profile of CSPs, while maintaining manu-
facturability and the ability to use completely standard printed
circuit board assembly lines.
www.eenewseurope.com
3D INTEGRATION & CIRCUIT PACKAGING
MCPCBs at acceptable prices
for the first time
Late last year and during January 2018 was a
tipping point for the LED industry. Now, there
is an alternative to the standard process of
creating a metal-clad printed circuit board
(MCPCB) — an alternative that used to only
be available for high-power LED applications
because of its price, but now is affordable
for more general tunable lighting thanks to
scale of manufacture and years of research
and innovation driving down the cost of unit
production.
Using an electro-chemical oxidation (ECO)
process, Cambridge Nanotherm converts the
surface of aluminium into a layer of thermally conductive, but
electrically isolating, nano-grain alumina (Al2O3). Because of its
method of creation, this “nanoceramic” is atomically bonded
to the rest of the board, which very simply provides superior
thermal conductivity.
While alumina doesn’t necessarily offer the best thermal
conductivity as a constituent element, the extreme thinness of
the aluminium oxide layer creates results in a thermal resis-
tance from the top side of the copper to the underside of the
aluminium of 0.31°C.cm2/W, and potentially lower, which is ideal
for tunable lighting, and relaxes design constraints considerably.
In simple terms, the best designs can now dissipate heat effi-
ciently enough to allow more individual LEDs to be combined in
the same array, increasing tuneable light output without greatly
increasing the focal spot size of the bulb.
Figure 4 illustrates how for a standard Nanotherm MCPCB,
optimum cooling of a point heat source is achieved by a disc of
copper 3.5mm in diameter.
While a larger area of copper will help further, the law of
diminishing returns applies. Thicker copper is of no benefit (un-
less the copper area is very large) and increases cost, weight
and prohibits circuit tracks with narrow gaps that are essential
for connecting to tiny CSP LEDs.
While tunable lighting drives the need for CSP LEDs, its up-
take and affordability are being held back by challenges — un-
less the industry adopts better thermal management. But given
the enormous demand for, and potential for, tunable lighting, the
LED industry is responding with innovation after innovation. And
in this regard, nanoceramic MCPCBs are leading the way.
eeNews Europe February 2018 News 29
 DESIGN & PRODUCTS 3D INTEGRATION & CIRCUIT PACKAGING

Atomically-thin memristors could yield

ultra-high density memory
By Julien Happich

A
pplying 2D materials to ambient conditions.
the memristor concept, a They also highlight that atomristor
team of multidisciplinary devices offer distinct advantages in terms
researchers from US and Chinese of ultimate vertical scaling, down to an
universities have demonstrated atomic layer with forming-free operation.
the scalability of nonvolatile By replacing the metal electrodes with
resistance switching down to graphene, the entire memory cell could be
atomically-thin devices. scaled below 2nm.
Reporting their results in ACS’ “Moreover, the transparency of gra-
Nano Letters in a paper titled phene and the unique spectroscopic
“Atomristor: Nonvolatile Resis- features of 2D materials affords direct
tance Switching in Atomic Sheets optical characterization for in situ studies
of Transition Metal Dichalcogen- and in-line manufacturing testing”, they
ides”, the researchers extrapolate write in the paper.
that the “atomristors”, as they Data endurance is still something that
describe their devices, could need to be improved, but memory reten-
yield memristor densities in the tion was tested up to a week, which could
range of 1015/mm3, which would be already sufficient for certain neuro-
translate to a theoretical areal Artistic rendering of a voltage-induced memory morphic applications involving short and
density of 6.4Tbit/in2 for single- effect in monolayer nanomaterials layered into medium term plasticity. In fact, due to the
bit single-level memory storage. “atomristors”. Credit: Cockrell School of Engineering, sub-nanometre thinness of the monolay-
Although it had been observed The University of Texas at Austin ers, they extrapolate that at a loose pitch
that a number of solution-pro- of
cessed multilayer two-dimensional (2D) material morphologies 10nm, an atomristor
could yield a nonvolatile resistance switching (NVRS) behaviour, density of 1015/mm3 could
where a device’s resistance can be modulated between a high- be achieved. Those could
resistance state (HRS) and a low-resistance state (LRS) and be used to mimic human
retain those states without power, it is the first time that such a synapses (about 109/mm3) in
NVRS behaviour is observed in atomically-thin vertical metal–in- neural networks.
sulator–metal (MIM) devices, claim the researchers. According to the authors,
To build their “atomristors”, the scientists prepared synthetic designed as a single-bit
atomic sheets of transitional metal dichalcogenide (TMD) such single-level memory device,
as MoS2, MoSe2, WS2 and WSe2, using standard chemical vapor these stacked atomristors
deposition (CVD) and metal–organic CVD processes. They would yield a theoretical
then transferred the atomically-thin sheets to sandwich them areal density of 6.4Tbit/in2.
between different types of electrodes including silver, gold and Another potential applica-
even graphene. Nonvolatile resistance switching was observed tion that the researchers Optical image of the atomristor
in all cases. The crossbar devices consisted of TMD atomic anticipate for their atomris- structure.
sheets between top and bottom electrodes, on top of a Si/SiO2 tors is zero-static power
device substrate. radio frequency (RF) switching. In fact,
In a particular implementation, in one of their experiments, they demon-
with 2D MoS2, the atomristor had strated a monolayer switch operating to
a high-resistance state (low cur- 50GHz with acceptable insertion loss of
rents measured) until a 1V bias about 1dB and an isolation in excess of
was applied to set the atomic- 12dB. The authors believe the atomris-
layer switch to a low-resistance tors could be further optimized through
state. The device retained it scaling to operate at 100s of THz. They
resistance value until a nega- conclude that in frequency switching
tive bias was applied to reset applications, the high breaking strain and
it. In this vertical metal–insula- ease of integration of 2D materials on soft
tor–metal configuration, which substrates would make the atomristors
lends itself well to 3D integration, Schematic illustration of the atomristor. The top and suitable candidates for the manufacture
the researchers measured an on/ bottom electrodes (TE and BE) could be gold, while of flexible nonvolatile digital and analog/
off ratio over 104 and truly zero- the TMD could be MoS2. RF switches.
static power for data retention in

30 News February 2018 @eeNewsEurope www.eenewseurope.com

 3D INTEGRATION & CIRCUIT PACKAGING

System-in-Package EDA solution for

heterogeneous chip integration Dresden, Germany, 11 – 12 April 2018
Advanced Semiconductor Engineering, (ASE), and Cadence smartsystemsintegration.com
Design Systems have jointly developed a System-in-Package
(SiP) EDA solution to address the challenges of designing and
verifying Fan-Out Chip-on-substrate (FOCos) multi-die pack-
ages.
Previously, IC packaging engineers leveraged standard EDA
design tools coupled with a set of loosely defined rules to Discover innovations.
Find solutions. Share visions.
At Smart Systems Integration, the international
conference and exhibition on integration issues
of miniaturized systems.

Creating a smarter future.

lay out their packages. However, this approach has many

limitations when designing today’s advanced multi-die pack-
ages. To provide a more holistic approach to the design and
verification of SiP and advanced fan- out packages, ASE and
Cadence collaborated closely to develop a design kit, meth-
odology, and streamlined and automated reference flow using
enhanced Cadence IC packaging and verification tools, all
tailored for ASE’s advanced IC package technologies.
The solution consists of the SiP-id (System-in-Package -
intelligent design) design kit, an enhanced reference flow
including IC packaging and verification tools from Cadence,
and a new methodology that aggregates the requirements of
wafer-, package- and system-level design into a unified and
automated flow.
By deploying the SiP-id methodology, designers can reduce
design iterations and greatly improve throughput as compared
to existing advanced packaging EDA tools, reducing the time Pre- & after-conference field trip
needed to design and verify ultra-complex SiP packages.
In a typical use case with high-pin-count dies, packaging
engineers using SiP-id and the accompanying reference flow Several keynotes by industry specialists
and methodology were able to reduce time from more than
six hours to only 17 minutes, compared to existing tools with Conference dinner on a paddle steamer
manual operation, claim the two companies.
More and more of our customers are looking at multi-die
advanced-package technologies to solve their next- genera- Co-organizer:
tion design challenges, said Tom Beckley, senior vice presi-
dent and general manager of the Custom IC & PCB Group
at Cadence. Advanced packaging extends Moore’s Law and
plays directly into our System Design Enablement strategy, so
collaborating with ASE to fulfill their vision for SiP is a natural
fit for us. We expect the results of this effort to mutually ben-
Part of the activities of:
efit Cadence and ASE customers by providing a methodology
optimized for SiP design.
Cadence
www.cadence.com

www.eenewseurope.com eeNews Europe February 2018 News 31

 DESIGN & PRODUCTS WIRELESS NETWORKING

What needs to happen for 5G to be a reality?

By Richard Edgar

T
oday, we rely on networks to ensure all our devices and nearly 3Gbps, while Vodafone said it had achieved its first 5G
gadgets are connected all the time, enabling us to do all data connection in Italy.
the things we want to do–send emails, read online, chat 5G will run on a new high-spectrum band that uses higher
to friends, stream videos and so much more. frequency signals than those currently used with 4G. It is likely
As this becomes ever further entrenched in our lifestyles it that 5G will sit on frequency bands up to 6GHz–4G occu-
becomes increasingly important–no one wants to experience pies frequency bands up to 20MHz. This means that the new
delays; they simply want their email to be sent, their message spectrum band will have less congestion–this is vital when you
received and their favourite TV show to start streaming imme- consider the Internet of Things (IoT) and the expectation that
diately. Whilst 4G has served us this will grow to reach 75 billion
well, that doesn’t stop us looking devices by 2025, combined with
to the future–hello 5G! the belief that mobile data use
Some may question why we will continue to rocket.
need 5G when 4G is sufficient
for most current requirements. The challenges
However, 5G will provide even currently facing 5G
faster data speeds, and more The internet is quickly becoming
importantly, it will, over time, saturated and a new standard is
offer enough capacity for users needed to meet future demand.
to perform every function they 5G won’t ‘just’ happen and there
want without loss of connection are a number of challenges that
or drop in speed, no matter how need to be overcome in the next
many people are connected at few years. In many ways these
the same time. Indeed 5G could are similar to the challenges that
even provide ‘always on’ con- faced 4G when as it was first be-
nectivity. ing deployed. The biggest chal-
Furthermore, when it comes lenge that needs to be overcome
to latency and increasing user is spectrum availability but cost
demand and expectation, 4G and network architecture also
simply won’t be able to keep up need to be addressed.
for much longer. 5G’s proposed Spectrum availability is always
ultra-low-latency could be as low a challenge; however, it remains
as between 1ms and 10ms, com- crucial to the success of 5G.
pared to 4G’s 40ms to 60ms. It It’s well known that the radio
would have the ability to provide frequencies used for 3G and 4G
the real-time responses users are are overcrowded and therefore
starting to crave. For example, new spectrum is needed to cope
it would enable a sports specta- with increasing demand. 5G will
tor at a stadium to watch a live need to use the proposed high
stream of an alternative camera frequency 6GHz bands to be
angle of the same game, with no able to deliver the desired data
delay. speeds. Allocation and manage-
So what needs to happen for ment of spectrum is down to
5G to become a reality? governments and to date, there has been little harmonisation
The initial standard for 5G has just been approved and fur- across countries. This will need to change for 5G services to
ther iterations will be finalised by 2020. Business applications work globally.
based on 5G will appear very shortly and it’s fair to assume that Additionally, to ensure the 5G spectrum doesn’t run out, it
we will then start to see consumer access a couple of years might be necessary to develop ways of using spectrum more
later. Analyst house Ovum suggests there will be 24 million 5G smartly. For example, by only assigning the amount needed for
subscribers by 2021. each task.
The 2018 Winter Olympics in PyeongChang, South Korea As with 4G, cost is a big consideration. However, until the
is expected to provide one of the first large-scale tests for 5G, technology needed for 5G is developed and tested, no one will
showcasing the types of services of which it will be capable. really know the true cost of deployment of the network infra-
This includes new kinds of video services to transform the structure.
experience of watching live events. However, between now and Equally, if the full benefits of 5G are to be realised, the price
2020 there will be numerous trials around the world. Recently, of 5G capable handsets and services for businesses and the
EE conducted its UK-first 5G trial and achieved speeds of consumer must be affordable. Just because they will be able to
download more data, and faster, it does not mean that they will
Richard Edgar is Director of Communications Technology at be prepared to pay a huge bill every month in excess of what
Ensigma, Imagination Technologies - www.imgtec.com/ensigma they now pay for 4G.

32 News February 2018 @eeNewsEurope www.eenewseurope.com


An additional consideration is whether devices will actu-
ally be ready for 5G. Potentially, we could implement this new,
super-fast network and not have any compatible user equip-
ment. This is a big concern for the 2018 Winter Olympics.
Byeong-Moo Lee, VP Olympic Business, KT has commented,
“We don’t yet know whether we will have sufficient devices. We
are working with all the major vendors to get as many as we
can. Our priority is smartphones, but we do have some tablets
and wearables in mind too.”
Will there ever be a need for 6G?
The pace at which IoT is growing is astronomical. Soon every
piece of technology will have a wireless connection, creating
a lot of traffic and congestion on the network–from washing
machines and fridges, through to farm animal sensors, fitness
bands and garbage bins as well as autonomous cars to name
but a few. Add to the mix consumer expectation and demand,
communication networks have got to keep up–simply, consum-
ers don’t care whether they use Wi-Fi or Bluetooth, for example,
they just want their devices to work in real-time.
TE Connectivity and Credo demo 112 Gbit/s single lane connectivity
By Jean-Pierre Joosting
T
E Connectivity and Credo have teamed up to demon-
strate single lane connectivity of 112 Gbps over a chip
to module (IO) channel and over a backplane channel
at DesignCon 2018. As industry groups such as OIF and IEEE
meet to discuss feasibility of these types of links at these next-
generation data rates, TE and Credo are showing performance
on actual hardware.
The IO channel demonstration uses the OSFP IO connector
from TE with 16nm, lower power, high performance 112G PAM4
SerDes technology from Credo to exhibit operation over a 10-
inch printed circuit board (PCB) channel. The “O” is for “octal”
— it is being designed to use eight electrical lanes to deliver
400 GbE — and “SFP” is for “small form factor pluggable.” The
channel is driven by Credo’s 112G PAM4 SerDes, which is oper-
ating over a total ball-to-ball channel loss of >15 dB. The demo
shows bit error rate (BER) performance of better than 1e-7. The
www.eenewseurope.com
WIRELESS NETWORKING
Unlike when previous generation networks, were implement-
ed, such as 1G, 2G and 3G, with 5G it is vital that the correct in-
frastructure is put in place right away. This will mean that rather
than being replaced in the not-to-distant-future it can simply be
upgraded. It’s also important that a real emphasis is placed on
software and not just hardware.
Typically, mobile data technologies were built around hard-
ware. With 5G a different approach is being taken, with the
focus on software which, compared to hardware, can be easily
updated. If this done correctly, there will be no need for 6G.
However, only time will tell if this is the case.
Looking to a 5G future
While the reality is that it is still some way off, 5G is coming and
when it does arrive, people will wonder how they lived without
it. 5G has the potential to change the world as we know it. It will
dramatically increase the speed at which data is transferred,
improve response times and provide enough capacity for the
billions of devices that will be connected in the IoT, paving the
way for innovative new technologies and services.
OSFP IO connector is a state-of-the-art 8 channel IO connector
currently adopted for 400 Gbps applications. The demonstra-
tion establishes the OSFP connector performance as an 800
Gbps-capable IO system.
The backplane demonstration uses TE’s latest STRADA
Whisper orthogonal backplane connector operating with a total
channel loss of 20 dB at 28 GHz in a PCB-based direct plug
orthogonal (DPO) architecture. Driven by Credo’s 112G PAM4
SerDes, the demo shows BER performance levels that fully
enable the adoption of STRADA Whisper solutions for the next
wave of networking equipment.
“Credo’s proven low power silicon expertise being ad-
vanced to 112G SerDes is a key technology milestone. Credo
is enabling the industry to progress quickly to 112 Gbps serial
electrical signaling, which may pave the way for accelerating the
deployment of 112G single lane, end-to-end network connectiv-
ity. Enabling faster data, in smaller spaces at a potentially lower
cost,” said TE’s Nathan Tracy, technologist, member of system
architecture team and industry standards manager.
eeNews Europe February 2018 News 33
 DESIGN & PRODUCTS WIRELESS NETWORKING

Why would the US consider nationalising 5G?

By Jean-Pierre Joosting

A
xios is reporting that the Trump administration is con- ties and eighties when many telecommunications companies
sidering nationalising the 5G network in response to were government-owned.
perceived threats from China and the advances it has 5G as a technology is also in a rapid state of flux, could a US
made in 5G and AI. Federal effort to build a 5G network be flexible enough to keep
To back up their claims, Axios obtained a powerpoint pre- abreast? There is precedent with the military where consumer
sentation and memo produced by a senior National Security mobile phone technology outpaced military radio to such an
Council official. extent that soldiers relied on mobile phones to communicate
According to Axios, the documents lay out two paths, the in Middle East around the time of the second Iraq war. If this
first is for the US government to build a centralised 5G net- comes to fruition, expect the rest of the world to move on and
work within three years and rent it out to service
providers, the second is for wireless companies
to build their own networks and compete with
each other but have to adhere to US standards
that dictate what equipment to use and on how it
is installed.
However, if the stated aim is to protect the
network against China, the first option would
be quite likely as wireless providers would most
likely source equipment that meets global stan-
dards of which China plays a big part, if not a
dominant role.
The second option is what is already happen-
ing with the exception that providers are free to
source equipment globally. Wireless providers
are already in a competitive race to implement
5G and gain market share. However, the bet-
ter second option could be mitigated in terms
of security by excluding Chinese products and
setting standards for the USA and its allies. This
presumes other countries would want to sign up.
There are significant problems to a purely
government run 5G network.
There is an issue of trust as users data can be
spied on by any government agency, much like
the fear before the advent of mobile that com-
panies like AT&T were working with government
to log and store voice calls. The network might
be secure from China, though this would be
debatable as governments globally already have
a poor record when it comes to data security,
but it would potentially hand the US government a lot of control advance more.
over user data and access. Would international companies pull Nationalisation would lead to stifled innovation and countries
back from the US due to data protection concerns, especially if such as China would get even further ahead. The approach
they could not encrypt their data. Even if the US run 5G network seems to be blatant attempt to take control of the communica-
used encryption, government agencies would have the keys. tions system. However, the second approach that sets stan-
Certainly, international companies in China are wary of the Chi- dards and requirements would divide the world but keep the
nese government and its potential ability to access their data. networks privately run – akin to industrial protectionism prac-
Such a project would be a huge government infrastructure ticed by many countries including China and the EU.
project subject to the usual cost and time overruns associated Though it is debatable that Chinese Telecommunications
with so many government run projects. It is entirely possible companies are truly private they tend to operate like private
that the outcome would be very different than envisaged. The companies, but with the benefits of government support at
project would not be market driven, killing innovation and critical times and with R&D. Might, it not be better for other
potentially missing out on new possibilities. Wireless exper- countries to support R&D and encourage a startup culture to
tise would need to be acquired and it is not given that such compete with China, rather than try and shut down the market
a project could attract the right people. The wireless industry to private enterprise through nationalisation. Afterall, China ap-
has boomed over the last two decades driven by collaboration, pears to be replicating the success Europe developed with the
standards and openness. Does anyone remember, in many GSM standard, which came to dominate mobile in the nineties,
countries, waiting months for a landline telephone in the seven- through support, standards and R&D.

34 News February 2018 @eeNewsEurope www.eenewseurope.com


Inverse-design and 3D printing results
in millimetre-wave metadevices
By Jean-Pierre Joosting
A
team at Northwestern University have come up with an
approach that starts with a function and asks what struc-
ture is needed to achieve the desired result – inverse
design – enabling them to create metadevices at millimetre-
wave frequencies.
The researchers used inverse design principles and a basic
3D printer purchased from Amazon, technologies
not commonly used in their field which enabled
the team to create highly efficient, non-resonant,
broadband metadevices at millimetre-wave
frequencies that could prove revolutionary for
consumer products, defence, and telecommu-
nications, including next-generation 5G wireless
networks.
Inverse design starts with a basic premise
about function and then finds a structure to en-
able it. Examples include wafer-thin eyeglasses
or a smartphone camera so small it is invisible
to the naked eye or an aerodynamic sensor that
can conform to the exact slope and angle of a jet
airplane wing.
These visions of future technologies emerge
from a technical paper titled “Inverse-designed The 3D-printed inverse-
broadband all-dielectric electromagnetic metade- designed meta-grating for a
vices” published online in Scientific Reports.
15° polarization splitter.
“I feel like we’re really on the verge of some-
thing big,” said Koray Aydin, assistant professor of electrical
engineering and computer science at the McCormick School
of Engineering, who is leading the research efforts in inverse-
designed metadevices. “There’s a lot that needs to be done in
the research part, but we’re going in the right direction.”
Using computer modelling, optimization software and
complex algorithms, the team set out to build metadevices that
could bend or focus millimetre waves but that avoided problems
with conventional approaches, such as low efficiency, narrow
bandwidth and the bulkiness of the devices.
“What we’ve achieved here is a new way of creating electro-
www.eenewseurope.com
WIRELESS NETWORKING
magnetic devices that achieve certain functions that conven-
tionally seemed impossible to do,” said Prem Kumar, professor
of electrical engineering and computer science in McCormick
and of physics and astronomy in the Weinberg College of Arts
and Sciences. Kumar likened the process to machine learning
and said it could produce unexpected outcomes, for example
the functionality over a broad bandwidth.
Francois Callewaert, a McCormick School of
Engineering graduate student who works with
Aydin, developed the inverse design algorithm
and performed the numerical simulations. Vesselin
Velev, a physics and astronomy graduate student
who works with Kumar, helped with the detailed
millimetre-wave measurements.
When the algorithm realised the design it was
of a complex geometric shape, one that was not
known or even intuitive. The problem facing the
researchers was how does one make such an ob-
ject? Conventional manufacturing methods would
be difficult and expensive. The answer was addi-
tive, or 3D, printing.“This is the heart of the study,”
Aydin said. “We’re the first to combine these two
to make working devices.”
Kumar agreed. “The important thing to me is the
multidisciplinary nature of it,” he said. “We can de-
sign a lens in a way that it doesn’t look like a lens.”
Another great strength of their process, Aydin said, was that
it was imminently scalable from the microwave to the visible
frequency range because of the flexibility of 3D printing.
“It is an exciting result,” said Alan V. Sahakian, the John
A. Dever Chair and Professor of Electrical Engineering and
Computer Science. “Where in the past somebody might have
done a long analysis trying to approximate the behaviour, here
we essentially input the behaviour we wanted into a computer
and the computer optimizes a structure that has that behaviour
and then it comes out the other end of this three-dimensional
printer.
eeNews Europe February 2018 News 35
 DESIGN & PRODUCTS
10x16x1.2-mm SiP delivers
multimode LTE-M / NB-IoT
The nRF91 series low power cellular IoT device developed by
Nordic Semiconductor is designed to be easily accessible for
cellular connectivity. Measuring
only 10x16x1.2mm, the global
multimode LTE-M / NB-IoT
System-in-Package (SiP)
device integrates a complete
low power cellular IoT system
including a modem, trans-
ceiver, RF front end, dedicated
application processor, Flash
memory, power management, and crystal and passive compo-
nents. The nRF91 SiP leverages Qorvo’s state-of-the-art, proven
RF front-end, advanced packaging, and MicroShield technol-
ogy to deliver a unique SiP that combines high performance
with low power consumption. The nRF91 series supports global
operation with a single SiP variant thanks to the combination of
Nordic’s multimode LTE-M / NB-IoT modem, SAW-less trans-
ceiver, and a custom RF front-end from Qorvo. The integration of
an Arm Cortex-M33 processor and Arm CryptoCell-310 security
IP enables an entire low power cellular IoT application to be
implemented on an nRF91 SiP. The built-in host processor fea-
tures TrustZone for Armv8-M, helping secure application data,
firmware, and peripherals using an isolated, trusted execution
environment across the CPU and system.
Nordic
www.nordicsemi.com
Conrad promises simple M2M application
development with IoT kit
Developed in collaboration with Elco Industrie Automation
and Microtronics, Conrad’s IoT starter kit C-Control 10 aims
to simplify entry into
the Internet of Things
(IoT) market, Industry
4.0, and digitization for
enterprises and devel-
opers. It consists of the
IoT board, an integrated
M2M SIM, several sen-
sors, general purpose
analog and digital
inputs, USB, SPI, UART, I2C and GPIO as well as a Bluetooth
interface. In addition to an optional temperature recording, any
sensors and actuators can be connected to the board. Data will
be transferred to the Conrad Connect platform via GSM, which
makes it accessible from the cloud. The functionality ranges
from monitoring and diagnostics to maintenance and entire au-
tomation applications. The 39x32x6mm module has its own op-
erating system and is ready for immediate use. User applications
can be designed immediately using readily available tools. Data
transmission is AES-encrypted and with server authentication.
The GSM module is implemented as an ultra-low power version
to save energy. A LiPo battery with 520mAh serves as the power
supply, which makes the box suitable for mobile use. The 32-bit
ARM Cortex-M3 based MCU supports low-power operation.
Conrad Electronic
www.conrad.com
36 News February 2018 @eeNewsEurope
WIRELESS NETWORKING
Stand-alone Wi-Fi module in a
29.0x13.5x2.66mm SMT package
The PAN9420 launched by Panasonic Industry Europe device
is a 2.4GHz 802.11 b/g/n fully embedded Wi-Fi Module with
integrated
stack and API
that minimizes
firmware de-
velopment and
includes a full
security suite.
Featuring a
fully shielded
case, integrated crystal oscillators and a chip antenna, the
PAN9420 module combines a high-performance CPU, high-
sensitivity wireless radio, baseband processor, medium access
controller, encryption unit, boot ROM with patching capability,
internal SRAM and in-system programmable flash memory. The
integrated memory enables storing web content such as HTML
pages or image data. The parallel support of access point and
infrastructure mode allows easy setup of simultaneous Wi-Fi
connections from the PAN9420 to smart devices and home
network routers. The pre-programmed Wi-Fi SoC firmware
enables Ad-hoc mode (Wi-Fi Direct) applications. With the
transparent mode, raw data can be sent from the UART to the
air interface to smart devices, web servers or PC applications.
The PAN9420 comes in a 29.0x13.5x2.66mm SMT package.
Panasonic Industry Europe
https://eu.industrial.panasonic.com
Bluetooth 5.0 ICs realize high-performance,
long range and low power
Toshiba Electronics Europe has added the TC35680FSG and
TC35681FSG to its line-up of ICs compliant with the Bluetooth
low energy Ver. 5.0 com-
munications standard. The
ICs enable the high-speed
features 2M PHY and Coded
PHY (500 kbps and 125 kbps)
added to the Bluetooth Ver.
5.0 standard and support all
of the required data rates. Re-
ceiver sensitivity is an indus-
try leading -105 dBm at 125 kbps, and a built-in high efficiency
power amplifier in the transmission block delivers up to +8 dBm
transmission power. All these features enable long range com-
munication with low current consumption. Based on an ARM
Cortex-M0 processor, the ICs incorporate a 256 kB Mask ROM
that supports the Bluetooth baseband process, and 144 kB of
RAM for processing Bluetooth application programs and data.
They support both the HCI and GATT profiles, including server
and client functions. See also: Casambi and Danlers collaborate
on Bluetooth mesh connected daylight sensors
Additionally, 18-port GPIOs interfaces, which can be set to 2
channels each for SPI, I2C, and UART are also included, allow-
ing full structuring of systems that connect to various peripheral
devices. These GPIOs can be configured as a wakeup function,
4-channel PWM, 5-channel ADC interface and many others.
Toshiba Electronics Europe
www.toshiba.semicon-storage.com
www.eenewseurope.com
 WIRELESS NETWORKING

Dual-mode mobile M2M Raspberry Pi and LimeSDR Partnership brings cellular

terminal targets logistics open-source DVB project data into IoT platform
Thuraya Telecommunications Company has The Raspberry Pi Zero and LimeSDR Advanced RF Technologies, Inc., (ADRF)
launched the Thuraya Tracking and Moni- Mini have been used together to has joind the Strategic Alliance Partner

toring (T2M) service in conjunction with the

launch of the first product release of the
year, the T2M-DUAL terminal. A mobile,
dual-mode device for superior machine-
to-machine (M2M) communications and
remote asset tracking and monitoring, the
T2M-DUAL enables the simultaneous col-
lection of data from multiple points includ-
ing location information, data from external
sensors and peripheral devices, and input
gathered from vehicle or heavy equipment
CANBus. The robust and interoperable
T2M-DUALallows for the integration of any
third-party applications and has several
built-in smart features such as geo-fencing,
network selection based on least cost rout-
ing, internal battery backup and location
and sensor data. With dual-mode capabili-
ties, the terminal ensures that connectivity
is automatically maintained as assets move
between regions covered by GSM and
those that lie beyond the confines of cellular
networks – under the strong and secure
coverage of Thuraya’s satellite network.
Thuraya Telecommunications Company
www.thuraya.com
create what is claimed to be the
world’s smallest open-sourced DVB
transmitter.
The DVB transmitter was created by
Evariste Okcestbon and is capable of
streaming live images from a PiCam
using avc2ts, dvb2iq and limetx
software to encode and transmit
H.264 video on the 1.2 GHz amateur
TV band. The two boards are con-
nected together using the USB ports.
The Raspberry Pi Zero measures
65 x 30 mm and the LimeSDR Mini
measuring 69 x 31 mm, giving a total
solution that is pocket sized. The
signals broadcast from the DVB pair-
ing are received by a Raspberry Pi 2
connected to an RTL SDR stick and
HDMI monitor. The receiver solution
runs rtl_sdr, leandvb, kisspectrum,
ts2es and hello_video software.
Lime
www.limemicro.com
Program set up by Sirqul – a collaboration
that enhances the capabilities of the Sirqul
IoT platform by incorporating data from
the cellular connections ADRF provides.
Sirqul is an engagement-as-a-service IoT
provider that offers software and bundled
set of vertical solutions, which combine
with mesh networks to track and identify
individuals’ behavior, location, and intent.
This can range from identifying the exact
whereabouts of a fan at a sports stadium
to deliver food to enabling autonomous
robots to navigate industrial warehouses
and factories, as well as be monitored.
ADRF’s technology enables Sirqul to
incorporate data from 3G and 4G adding
redundancy to the network, while creating
new opportunities for venues and build-
ings that aren’t equipped to provide WiFi.
Sirqul will be able to leverage ADRF’s
expertise in deploying successful wireless
systems and mission-critical technologies
in public arenas and smart buildings.
Advanced RF Technologies
www.adrftech.com

Your Innova�ve New Products Deserve Our Innova�ve New Crystals

• Ultra-Miniature components crucial for IoT

applica�ons.
Small - 2.0mm x 1.6mm • Maximise PCB Real Estate
20.000MHz to 54.000MHz • Smaller crystals need lower power requirements
• Off-The-Shelf products developed with leading chip
manufacturers for Bluetooth, Zigbee and WiFi
applica�ons.
• Proven Reliability
Smaller - 1.6mm x 1.2mm
• Highly compe��ve global pricing from AEL Crystals Ltd
24.000MHz to 54.000MHz
Talk to our highly experienced field sales engineers, and let’s
see how we may assist you in the development of your next
innova�ve product.
Smallest - 1.2mm x 1.0mm
36.000MHz to 54.000MHz
Unit28, The ioCentre - Salbrook Road - Salfords - Surrey - UK
email: sales@aelcrystals.co.uk - Web: www.aelcrystals.co.uk
ael Tel: +44 1293 789200

www.eenewseurope.com eeNews Europe February 2018 News 37

 DESIGN & PRODUCTS
Model-level debugging: a bridge between
control engineering and software engineering
By S. Tucker Taft
A “model-level” debugger is a debugger that can provide a
side-by-side view of a simulation model of a control system (for
example defined using Simulink) with the source code produced
by an automatic code generator, and allow the developer to
use the model instead of the generated source code for setting
breakpoints, viewing and updating signal values, and stepping
through execution. This is similar to how a “source-level” de-
bugger supports debugging using the source code, freeing the
user from worrying about the details of the assembly-level code.
A model-level debugger that is based on an underlying
source-level debugger can also be used for testing systems that
combine auto-generated and hand-written code, potentially on
both the host and on some final target. Furthermore, presum-
ing a separate simulator for the model, the developer can use
such a debugger to perform back-to-back comparisons against
signal values logged during simulation, for an individual block,
or for a model as a whole, while delving into the details of a par-
ticular subsystem as needed. By displaying the model together
with the generated source code, a model-level debugger pro-
vides a uniquely productive bridge between control engineering
and software engineering.
Screen capture of the QGen model-level debugger.
Interacting with a model-level debugger
A model-level debugger lets you monitor and control the execu-
tion of code generated from simulation models, either on the
host (“Software in the Loop”, or SIL) or on an embedded target
machine (“Processor in the Loop”, or PIL). Such a debugger
gives you full control through model-level breakpoints and
signal-value displays, helping to connect the software elements
to the corresponding modeling elements.
You can combine auto-generated and hand-written code.
Both can be debugged together, and a model-level debugger
can display the model and its signal values whenever it reaches
sections of code that were auto-generated. If the debugger
stops in hand-written code, then only the source code would
be displayed. But if a breakpoint is reached in auto-generated
code, then the source code line, and the associated block
S. Tucker Taft is Director of Language Research at AdaCore –
www.adacore.com – He can be reached at taft@adacore.com
38 News February 2018 @eeNewsEurope
PROGRAMMING & DEBUGGING TOOLS
within the model, can both be displayed. In effect, model-level
debugging keeps the simulation model execution in sync with
code execution, by using traceability information built by the
automatic code generator. You can thus analyze the model’s
behavior while stepping through the generated source code, or
even the final assembly code.
A model-level debugger can also let you alter the values of
signals and redirect program execution to force the model to
reach a certain state at any point. This effect can be difficult
to achieve by merely altering the inputs to the model. If you
suspect a problem in one part of the model, you can probe it
further without needing to leave the debugger to change the
model or craft new input streams.
An important capability of a model-level debugger is to log
signal values during a debugger run, and then trace them back
to the original block(s) within the model in the simulator. This
can help resolve unexpected behavior and let you make any
desired changes in the simulator at the identified point in the
model.
Cross development using a
model-level debugger
Many source-level debuggers support “cross” debugging,
where a program can be debugged on a separate target ma-
chine, using a network or other connection between the host
(where the debugger is running) and the target (where some
kind of small-footprint “agent” is running). This same capability
is present in a model-level debugger that incorporates such a
cross-debugger. By using the normal cross-compiler toolchain
and target interfacing capabilities, together with an appropriate
auto code-generator, you can directly cross-compile and run
the auto-generated code on a supported platform while debug-
ging at the model level from the host and observing the execu-
tion behavior of the target.
Examples of use of a model-level debugger
The simulation model can be displayed and browsed within an
IDE, as a read-only view showing all simulation blocks, subsys-
tems, and sub-model references.
Displaying the model in an Integrated Development
Environment (IDE)
www.eenewseurope.com

You can define the model’s inputs via using the standard
simulator capabilities, or even using completely separate
models. A model-level debugger can record all inputs provided
to the model and create a debugging session where the same
inputs are sent, under control of the debugger, to the generated
code. The behavior of the model and the generated code with
these inputs can then be observed and debugged in sync.
Creating input for the target using the simulator
Just as you can insert a breakpoint in the source code, a
model-level debugger lets you insert a breakpoint directly from
the model view. Execution will subsequently pause when reach-
ing the code corresponding to the block containing the break-
point.
Inserting a breakpoint on a given block or model reference
By maintaining exact traceability between code variables and
their corresponding signals in the simulation model, a model-
level debugger can display signal values as they are updated.
It can grey out blocks that have been executed, to show which
signals have been updated and which still have values from the
previous computation cycle.
Displaying values for signals dynamically.
A model-level debugger provides a powerful way to analyze
the specific behavior of a given simulation model. By changing
signal values for one or more iterations you can direct the model
into a desired state of interest to the debugging session.
www.eenewseurope.com
PROGRAMMING & DEBUGGING TOOLS
Changing signal values during the execution.
A model-level debugger can store in a file all or some signal
values obtained during a debug session. When opening this file
in the simulator, you then have links to the corresponding simu-
lation block of each signal. This helps trace problems identified
at the code level to corresponding model-level blocks, where
you can investigate them further in the simulator.
Logging signals and highlighting corresponding blocks in the
simulator.
A model-level debugger shows you how a model-based
system performs on the final target, even when combined with
hand-written code. You can observe and understand the behav-
ior of the system at multiple levels - the model, the source code,
or the assembly code - by stopping at critical points, examining
signal values and seeing how they change over time, and stop-
ping within hand-written code as desired.
You can thus monitor, test, and debug an entire program, and
bridge the gap from the control engineering view of the world
of blocks and signals, to the software engineering view of the
world of functions and variables.
A model-level debugger can help you isolate and identify
problems where they are most easily captured, by combining
the full power of a low-level debugger with the model-level view
of the whole system.
AdaCore has developed a model-level debugger based on
the principles outlined above, as part of its QGen Model-Based
Development Suite. This is helpful for users who prefer a model-
based view of the world, to nevertheless debug problems that
only show up on the final target. Something that was unexpect-
ed, is that such a debugger turns out to be helpful for users who
are new to model-based development, by showing them the
correlation between models and auto-generated code, and con-
necting modeling concepts with programming concepts, which
modeling neophytes may find more accessible. And finally, the
ability to see the same system from multiple viewpoints allows
system engineers and software engineers to work together to
ensure the overall quality of the delivered model-based system.
eeNews Europe February 2018 News 39
 DESIGN & PRODUCTS PROGRAMMING & DEBUGGING TOOLS

Artificial intelligence chips could spill out

of data centers, onto desks
By James Morra

N
o one knows what the future of artificial intelligence will challenge Nvidia yet.
look like and no one knows what computer architecture Nvidia pushes its appliances and workstations as a power-
will take it there. For years, Nvidia has been trying to ful package for deep learning. It sold one of the first bundles to
expand the market for its graphics chips, which are the cur- Avitas Systems, an automated inspection startup that coaches
rent gold standard for training and running algorithms based software to identify corrosion and other defects using photo-
on deep learning. And it has often used tactics unrelated to the graphs of things like underwater pipelines and critical equip-
performance of its chips. ment in facilities like power plants and oil refineries.
Last year, the company released the DGX Station to enable “You might have researchers or data scientists that are do-
software engineers to experiment with software libraries used ing a lot of experimentation, refining their models,” said Tony
in artificial intelligence and improve algorithms before sending Paikeday, director of product marketing for Nvidia’s DGX Sys-
them to the cloud, where the software is trained on enormous tems. “At that stage of the development lifecycle, we found that
amounts of data. The workstation contains chips based on developers prefer not to feel encumbered by a resource sitting
Nvidia’s Volta architecture and provides
480 trillion floating-point operations per
second, or teraflops.
The DGX workstation shares the same
software stack as the DGX-1 appliance,
a miniature supercomputer that provides
960 teraflops of performance. That way,
software engineers can swiftly swap
software between Nvidia’s workstations
and appliances, which can be installed
in data centers where training typically
happens.
Nvidia introduced both products to
tighten its grip over the artificial intel-
ligence market and promote its Volta
architecture, which contains custom
tensor cores for handling deep learning.
But according to one industry executive,
the company’s rivals could use the same
strategy to push their custom chips onto
software engineers.
“They say, in the early phases of
designing neural networks, we don’t
want to go to data centers,” said Jin Kim, chief data science way out in a data center. They want it to sit close to where they
officer for machine learning chip startup Wave Computing. “We sit.”
want a workstation right next to us for experimentation, taking “We did this because we wanted to offer a proof point to
elements of existing neural networks and putting them together the marketplace,” said Paikeday, adding that Nvidia looks
like Lego blocks.” forward to original equipment manufacturers selling their own
He declined to disclose whether Wave Computing plans to workstations using its graphics chips and software language.
release its own workstation. But the company, which has raised “We wanted to set a blueprint for them to follow,” he said in an
$117 million over the last nine years, has been putting the fin- interview with Electronic Design.
ishing touches on an appliance equipped with its dataflow pro- Other companies may also follow Nvidia’s blueprints. In addi-
cessing unit (DPU), which supports lower precision operations tion to Wave Computing, both Graphcore and Intel are working
that consume less power and memory than traditional chips. on server appliances that could potentially be paired with a
When it is finished, the appliance is projected to provide per- workstation. Startups like Groq and Cerebras Systems could
formance of 2.9 quadrillion operations per second for machine imitate them, building boxes with custom silicon and the mas-
learning workloads. Wave Computing has also built a special sive amounts of memory required for training.
compiler that translates code into a form that its silicon can There are other hints. Over the last year, Graphcore has
understand. The company designed its coarse-grained recon- raised $110 million from investors that include Dell, the second
figurable array chips to have 16,384 cores. largest supplier of workstations as well as a prolific purveyor of
Wave Computing is acutely aware that software engineers server infrastructure. The company claims that its custom hard-
are asking for workstations to experiment with algorithms ware can be used to shorten the training phase of deep learning
outside of the data center, said Kim. Other startups have almost from days to hours.
certainly gotten the same requests. But none have ventured to Industry analysts say it is also possible that Dell will acquire

40 News February 2018 @eeNewsEurope www.eenewseurope.com

 PROGRAMMING & DEBUGGING TOOLS

Graphcore, giving it custom chips to install in its servers and gateways. Hewlett
Packard, the largest supplier of workstations and another major maker of servers, is
also considered a potential destination for startups like Groq and Cerebras Systems,
which are still operating under the radar.
That could affect where these companies stand in the market for deep learning
chipsets, which the research firm Tractica predicts could grow from $513 million in
2016 to $12.2 billion by 2025. Nvidia estimates that the market for computer chips
used in training could generate $15 billion in 2020, in contrast to $11 billion for infer-
encing, which requires less powerful chips.
Paikeday could not discuss specific numbers, but he said that Nvidia almost
always sells its workstations paired with an appliance. That could change as it con-
DC-DC Converters
tinues to charm small companies with the power and portability of the Station. For
instance, Avitas Systems takes the system out to where it collects information, editing Transformers &
algorithms in the field.
“I also think training will move out of the data center,” Kim, Wave’s chief data
science officer, said in an interview. “There are small enterprises that want to have
Inductors
early stage training done locally, especially when using methods like transfer learn-
ing, where you don’t need data centers full of compute resources to train and deploy
reasonably accurate models.”
Transfer learning is a technique for training software on small platters of data. It in-
DC-DC Converters
volves removing the layers of a neural network trained for a specific task, like identify- 2V to 10,000 VDC Outputs
ing faces in a photograph, while preserving lower levels of the network, which handle 1-300 Watt Modules
more primitive tasks like pattern matching. Using the lower levels as the foundation, • MIL/COTS/Industrial Models
the higher levels can be trained on smaller amounts of data to complete more special- • Regulated/Isolated/Adjustable
ized tasks, like spotting skin cancer. Programmable Standard Models
• New High Input Voltages to 900VDC
Many industries like finance and healthcare are starting to experiment with transfer
• AS9100C Facility/US Manufactured
learning, said Kim. And since it requires less computing power and less data than • Military Upgrades and
training from scratch, engineers could wing it with appliances or workstations instead Custom Modules
of renting cloud infrastructure. That could be a faster growing market than data cen-
ters, said Kim.
This article was originally published on Electronic Design –
Transformers
www.ElectronicDesign.com & Inductors
Surface Mount &
Imagination pushes developers to create CNNs on GPUs Thru Hole
• Ultra Miniature Designs
Imagination Technologies’ PowerVR CLDNN software development kit has been • MIL-PRF 27/MIL-PRF 21308
designed to makes it easy for developers to create Convolutional Neural Networks • DSCC Approved Manufacturing
(CNNs) using PowerVR GPUs. Both an API and SDK are included, as well as an im- • Audio/Pulse/Power/EMI
age for flashing on to an Acer Chromebook R13 for Multiplex Models Available
hardware development. • For Critical Applications/Pico
Imagination launched its first neural network ac- Modules, Over 45 Years’ Experience
celerator, PowerVR Series2NX in September 2017.
For full characteristics of these and the
Future versions of the API and SDK library will entire PICO product line, see PICO’s
enable developers to develop neural network ap- Full line catalog at
plications that will be able to target Imagination’s
GPU and NNA hardware without having to rewrite
www.picoelectronics.com
code. The PowerVR CLDNN API is Imagination’s
first AI-focused API. It provides functions to create network layers for constructing
and running a neural network on PowerVR hardware. By using specialist OpenCL
PICO ELECTRONICS, Inc.
kernels, it enables developers to focus on their neural network creation with fewer 143 Sparks Ave., Pelham, New York 10803
s
overheads. The API also performs low-level hardware-specific optimizations, en- Call Toll Free 800-431-1064
FAX 914-738-8225
abling better-optimized graphs than a custom user OpenCL implementation. The
PowerVR CLDNN SDK demonstrates how a neural network can be deployed to E Mail: info@picoelectronics.com
PowerVR hardware through the CLDNN API. It includes helper functions such as file
loading, dynamic library initialisation and OpenCL content management. A refer- Pico Representatives
ence manual is also included, as well as the source code for sample applications Germany
ELBV/Electra Bauemente Vertrieb
that demonstrate how to use the PowerVR CLDNN API. An image has also been E mail: info@elbv.de
created that developers can flash to an Acer Chromebook R13, which contains a Phone: 49 089 460205442
Fax: 49 089 460205442
PowerVR GX6250 GPU via the Mediatek MT-8173 SoC. By downloading the image,
England
developers can access everything they need to write neural network applications on Ginsbury Electronics Ltd
the device. E-mail: rbennett@ginsbury.co.uk
Imagination Technologies Phone: 44 163 429800
Fax: 44 163 4290904
www.imgtec.com

www.eenewseurope.com eeNews Europe February 2018 News 41

 DESIGN & PRODUCTS
Codasip’s Studio: 7th edition for the easy
configuration of RISC-V processors
RISC-V embedded processor IP supplier Codasip has launched
the 7th generation of its Studio, IP-design and customization
software that allows
for fast configuration
and optimization of
RISC-V processors and
customer-proprietary
processor architec-
tures. Studio 7 can
be used for processor
prototyping for a spe-
cific application domain, fast design space exploration, or the
development of custom extensions using Codasip’s architecture
description CodAL language. The tool then generates hardware
and corresponding SDKs that are aware of the custom exten-
sions, including Verilog or VHDL RTL and System Verilog UVM
environments, testbenches and synthesis scripts, full compiler
toolchain including advanced profiling and debugging tools, and
both cycle-accurate and fast instruction-accurate simulation
tools. Some of the new features included with Studio 7 include
native support for industry-standard AMBA interfaces, allow-
ing for easy replacement of other processor cores while reusing
existing, proven peripheral IP. IEEE 1149-7-compatible 2-wire
JTAG minimizes pin-count. The tool also features improved
clock-gating for low-power requirements.
Codasip
www.codasip.com
Lauterbach extends tool support
to Renesas RH850
Microprocessor tool vendor Lauterbach GmbH has extended
the kernel awareness for the embOS Real-time Operating
System (RTOS) from SEGGER
Microcontroller to the RH850 family
of microprocessors from Rene-
sas. Debug tool family TRACE32,
already supports embOS on ARM,
PowerPC, RX, SH and NIOS-II
families. The embOS awareness
plugin for TRACE32 allows the de-
veloper to visualise RTOS resourc-
es and objects such as task lists,
mailboxes, timers and semaphores.
Developers are free to investigate interrupt routines, drivers
and application code all from within the familiar environment of
TRACE32. When the awareness is configured, extra features
become available, for instance the setting of task aware break-
points. All members of the RH850 family provide dedicated
counter registers which can be accessed non-intrusively by the
TRACE32 debugger. These can be configured to display mini-
mum, maximum and mean runtimes for a user marked block
of code or the runtimes of various tasks in the embOS system.
If the target provides off-chip trace capabilities, TRACE32 can
record processor cycles and can be configured to collect data
on task switches. Using this information, a detailed analysis of
the program history, including task switches, can be viewed..
Lauterbach GmbH
www.lauterbach.com
42 News February 2018 @eeNewsEurope
PROGRAMMING TOOLS
Qt Company, Green Hills collaborate on
automotive HMI platforms
The Qt Company and Green Hills Software will continue their
partnership and enhance the integrated automotive digital
cockpit, first
debuted at
CES this
year. The two
companies
will further im-
prove the hu-
man machine
interface (HMI)
development platforms behind the digital cockpit concept. The
digital cockpit consists of a number of different HMI platforms,
each having a unique security level. These HMIs are powered
by instrument clusters using different operating systems, which
in turn make it harder to offer a single unified cockpit experi-
ence. Green Hills and Qt have worked together to overcome
this problem by integrating the Qt Automotive Suite framework
with Green Hills’ INTEGRITY Multivisor virtualisation tools.
This software combination helps integrate the HMI interfaces
across multiple operating systems operating a different safety
levels into one multicore solution, while remaining compliant
to stringent automotive safety standards. The combination of
Green Hills INTEGRITY RTOS and Qt Safe Renderer enables
safety-critical display elements to be certified to ISO 26262.
The concept cockpit demo shown at CES was built with the
Qt Automotive Suite to demonstrate interdomain graphics and
data sharing.
Green Hills
www.ghs.com
Open automotive sound design platform
Harman’s AudioworX has been released as a new standard-
setting audio framework for innovative and configurable
in-vehicle user experiences. Built upon the company’s exper-
tise in user experience (UX) design,
acoustics and signal processing,
AudioworX aims to help automak-
ers differentiate their brands through
highly personalized, passenger-cen-
tric solutions. The Harman Audio-
worX framework comprises three
main components: The Operating
System: A unified platform enabling
rapid design and quick porting on diverse hardware, accom-
plishing more in less time with a high degree of flexibility, ef-
ficiency and re-use. The Library: An extensive digital signal pro-
cessing (DSP) library consisting of more than 200 components
that accelerates the delivery of rich new audio features while
fast-tracking system tunings. The Tuning Tools: A compre-
hensive tuning tool kit that provides smart, seamless, intuitive
configuration capabilities, allowing precise algorithm and audio
control of both audio and NVH algorithms. This control makes
it easier to achieve expert system tuning for advanced automo-
tive sound design, both for sound systems and in-vehicle noise
management.
Harman
www.harman.com
www.eenewseurope.com

videantis claims 1000x performance boost on
deep learning applications
videantis GmbH ‘s v-MP6000UDX visual processing architecture
is claimed to increase deep learning algorithm performance by
up to three orders of magnitude, while maintaining software
compatibility with the company’s previous
v-MP4000HDX processor architecture.
The company continues to see a lot
of demand for smart sensing systems
that combine deep learning with other
computer vision and video processing
techniques such as SLAM or structure
from motion, wide-angle lens correction,
and video compression.
It claims to be the only company that
can run all these tasks on a single unified processing archi-
tecture, simplifying SOC design and integration while easing
software design and reducing unused dark silicon. “We’ve qui-
etly been working on our deep learning solution together with a
few select customers for quite some time and are now ready to
announce this exciting new technology to the broader market,”
says Hans-Joachim Stolberg, CEO at videantis.
“To efficiently run deep convolutional nets in real-time re-
quires new performance levels and careful optimization, which
we’ve addressed with both a new processor architecture and
a new optimization tool. Compared to other solutions on the
market, we took great care to create an architecture that truly
processes all layers of CNNs on a single architecture rather
than adding standalone accelerators where the performance
Ceva goes non-DSP with neural processor
Ceva Inc., well known as a supplier of DSP intellectual property
circuits, has announced its first non-DSP processor family –
the NeuPro for neural network inferencing applications. Ceva
(Mountain View, Calif.) looked at more than
120 neural network software architectures
as it devised its hardware, according to
Liran Bar, director of product marketing for
Ceva’s imaging and vision DSP production
line. These include the well-known AlexNet,
GoogleNet and ResNet. Although the NeuPro
family members are well suited to image-
based applications, where Ceva also offers
the DSP-based Ceva-XM, it also addresses
multiple other classification and pattern recognition applications
and is described as being suitable for AI at the edge. These
applications include natural language processing, real-time
translation, authentication, workflow management and other
learning-based applications. NeuPro is not intended to do the
training of neural networks, which is usually done off-line in a
data center with weights then downloaded to edge equipment.
The NeuPro architecture is based on a sea of 8bit by 8bit MACs
that can be paired up to perfrom 16bit by 8bit calculations and
16bit by 16bit calculations. There are four AI processors:
the NP500 is the smallest processor, including 512 8bit MAC
units and targeting IoT, wearables and cameras;
the NP1000 includes 1024 MAC units and targets mid-range
smartphones, ADAS, industrial applications and AR/VR head-
sets; the NP2000 includes 2048 MAC units and targets high-end
smartphones, surveillance, robots and drones;
www.eenewseurope.com
MISCELLANEOUS
breaks on the data transfers in between.” “Using some clever
design features, the v-MP6000UDX architecture we’re announc-
ing today increases throughput on key neural network imple-
mentations by roughly 3 orders of magnitude, while remaining
extremely low power and compatible with our v-MP4000HDX
architecture. This compatibility ensures
a seamless upgrade path for our cus-
tomers toward adding deep learning
capabilities to their systems, without
having to rewrite the computer vision
software they’ve already developed for
our architecture”, Stolberg continued.
The v-MP6000UDX processor architec-
ture includes an extended instruction
set optimized for running convolutional
neural nets, increases the multiply-accumulate throughput
per core eightfold to 64 MACs per core, and extends the
number of cores from typically 8 to up to 256. Alongside the
new architecture, videantis also announced v-CNNDesigner,
a new tool that enables easy porting of neural networks that
have been designed and trained using frameworks such as
TensorFlow or Caffe. v-CNNDesigner analyzes, optimizes, and
parallelizes trained neural networks for efficient processing
on the v-MP6000UDX architecture. Using this tool, the task of
implementing a neural network is fully automated and it just
takes minutes to get CNNs running on the low power videantis
processing architecture.
Videantis
www.videantis.com
the NP4000 includes 4096 MAC units for high-performance
edge processing in enterprise surveillance and autonomous
driving.
“The trend in AI is towards lower 8bit or even 4bit or binary net-
works, but these lower resolution cases
have not been proved beyond academic
exercises. We support 8bit and 16bit
fixed point data. The 16bit accuracy is for
retraining purposes,” said Bar.
To provide a seamless handover between
vision processing and neural network
analytics the NeuPro includes a vision
processing unit (VPU) based on Ceva-
XM. The VPU also handles the CDNN
(Ceva Deep Neural Network) software and provides software-
based support for new advances in AI workloads. CDNN is
Ceva’s neural network software framework that allows develop-
ers to generate and port their proprietary neural networks to
the processor. CDNN supports the full gamut of layer types and
network topologies.
The NeuPro hardware architecture is designed to reduce band-
width to external memory via DDR and includes convolution,
and fully connected activiation and pooling layers. It also per-
forms “on-the-fly” pooling and activation pipeline processing.
The IP cores are being aimed at 16nm manufacturing process
technologies and have been benchmarked as providing 30
times the performance of the Ceva-XM4 and 20 times the per-
formance of the Ceva-XM6 when running ResNet-50.
Ceva Inc.
www.ceva-dsp.com
eeNews Europe February 2018 News 43
 DESIGN & PRODUCTS MISCELLANEOUS

Temperature and humidity data loggers Snap-on cover lids ease electronics
can monitor up to eight sensors enclosures assembly
Teracom’s TCW210-TH data loggers are aimed at remote tem- Pixus Technologies now offers covers for instrumentation en-
perature and humidity monitoring applications. Each TCW210- closures with snap-on lids for easy assembly. The Pixus enclo-
TH can monitor up to sure lids feature
8 temperature and/or a side edge fold
temperature-humidity with half shears
sensors of various that allow a solid
brands and types (up to or perforated en-
24 different parameters closure cover to
each) via either 1-Wire snap into place.
interface and/or the The covers have
more robust MODBUS side notches for
RTU over RS485. For fitting into rails
example, four sensors can connect via MODBUS and another with 160mm,
four can connect via 1-Wire, all at the same time. Utilizing 280mm or
an Ethernet interface for web and network connectivity, the 280mm depths. This lid design provides easy and cost-effec-
device monitors real time data for temperature, humidity and tive assembly. EMC gaskets are also included. The snap-on
dew point, and records all monitored parameters using internal lids are available for Pixus’ Vario, RiCase, Mobile, and EMC
FLASH memory. The records can be made on set time intervals subrack enclosure systems. Pixus provides subracks and elec-
and/or on alarm conditions. Memory capacity allows at least 36 tronics enclosures for various applications, including Industrial,
days logged with data recorded once every minute. The log file Energy, Communications, Mil/Aero, Medical, Transportation,
can be periodically uploaded on a dedicated server by HTTP Research/Physics, and other markets. The company also
Post. The device is supported by Android and iOS applications, provides backplanes, chassis platforms, and components for
as well as by Teracom’s remote and monitoring software TC embedded computing systems.
Monitor, which provides a free license for up to 10 items. The Pixus Technologies
logger utilizes a circular buffer in FLASH memory (up to 70,000 www.pixustechnologies.com
records). When it is full, the new data overwrites the oldest one.
Teracom Systems
www.teracomsystems.com Glass-based 8.0x8.0mm capacitive fingerprint
sensor is transparent
Multi-Gig Ethernet for copper Japan Display Inc. (JDI) has developed a transparent glass-
based capacitive fingerprint sensor, relying on the proprietary
in-vehicle networks capacitive multi-touch
Announcing a collaboration with Nvidia to provide Multi-Gig technology it uses for
networking support for the company’s DRIVE Xavier and its LCD product lines.
DRIVE Pegasus high- JDI’s glass-based
performance computing capacitive fingerprint
platforms, Aquantia has sensor applies the
launched its AQceler- basic touch functional-
ate Automotive prod- ity technology that is
uct line, including the integrated in its in-cell
AQV107 Multi-Gig PHY, Pixel Eyes displays
the AQVC107 PCIe which have been used
Multi-Gig MAC+PHY in smartphone and
Ethernet controller and digital still camera applications. With Pixel Eyes, the touch
the AQVC100 PCIe Multi-Gig controller (MAC only). All sup- functionality is integrated into the glass substrate by detecting
porting data rates up to 10GbE over copper, the devices aim to the changes in capacitance that occur when a finger touches
foster the development of autonomous vehicles with Multi-Gig the LCD surface. The glass substrate identifies the area
In-Vehicle Network (IVN) able to rapidly share data between the touched by the finger through the detection of the changes
sensors and the compute nodes for a safe and secure driving in capacitance. This technology has evolved within JDI to the
experience. Moving to Level 4 and Level 5 autonomous driving extent that the company has succeeded in enhancing it to
requires a new network architecture that can support the high- detect the changes in capacitance caused by the recesses and
bandwidth requirements for the IVN to support the increasing ridges of an individual’s fingerprints. This further technological
number of sensors and high-resolution cameras. Redundancy development made it possible to form JDI’s new capacitive
of all the function-critical components and systems is neces- fingerprint sensor on a transparent glass substrate. JDI will
sary to provide the utmost levels of safety. Earlier this month, expand the lineup to include larger or smaller sizes as well as
Aquantia also announced a collaboration with Molex on a 10 the 8x8mm sensor now proposed at a pixel density of 508dpi
Gbps Automotive Ethernet Network at CES. for a 160x160 resolution.
Aquantia Japan Display Inc.
www.aquantia.com www.j-display.com

44 News February 2018 @eeNewsEurope www.eenewseurope.com


JPEG 2000 Audio/Video over IP OEM boards
Silex Inside (formerly Barco Silex), has extended its family
of Audio/Video over IP OEM boards to include JPEG 2000
encoding and decoding. The company has enhanced its Viper
OEM solution with the
support of JPEG 2000
codec for the high-end
AV market. The integra-
tion of JPEG 2000 adds
new possibilities to the
Viper family by supporting
higher bandwidth video
signals including 4K 60fps
4:4:4. The Viper OEM solution with VC-2 HQ remains use-
able for applications where the video format is limited to 4K60
4:2:0. Thanks to the JPEG 2000 codec, the video signal can be
further compressed while keeping the latency below one video
frame. The latency of a complete AV over IP setup with Viper
including the transmitter and receiver remains very low, around
10ms. By improving the compression ratio of the Viper plat-
form, video formats up to 4K 60fps 4:4:4 now can be transport-
ed over a single 1G Ethernet cable with visually lossless quality.
The new platform includes all the features that were already
part of the original VC-2 HQ version, namely, video wall exten-
sion, stream encryption, forward error correction for increased
reliability, image processing features and more.
Silex Inside
www.silexinside.com
Constant-current LED drivers in low-profile
DFN package for edge lighting
Diodes Incorporated has extended its families of linear LED
drivers to include the BCR420UFD and BCR421UFD devices
in the ultra-low profile
DFN2020 package, only
0.6mm high, making
them well-suited for 12V
and 24V LED edge-light-
ing applications.
Like other members of
the BCR4xxU series, the
new devices can deliver
an adjustable constant
current between 10mA
and 200mA (BCR420UFD) or 350mA (BCR421UFD) with ±10%
tolerance, which in the case of the BCR421UFD can be con-
trolled using a PWM signal of up to 25kHz and a duty cycle of
1% to provide LED dimming functionality. Both devices include
a negative temperature coefficient, which lowers the LED driver
current as the internal temperature rises, thereby protecting
and extending the lifetime of the LEDs. In addition, the linear
topology used is based on an NPN emitter-follower with emitter
resistor, which means the devices exhibit virtually no EMI, mak-
ing them fit for sensitive applications such as medical lighting.
With a wide operating voltage range of between 1.4V and 40V
and total power dissipation of up to 1.7W, the devices can be
used to drive longer strings of LEDs while still providing suf-
ficient protection from transients in a 24V system.
Diodes
www.diodes.com
www.eenewseurope.com
MISCELLANEOUS
Flat panel controller allows
custom monitors up to 4K UHD
The CRTtoLCD-91 from eCOUNT embedded is a new flat panel
controller for the development of customer-specific monitors,
displays
and
control
panels
that sup-
ports flat
screens
with up
to 4K
UHD
resolution. As the first industry solution of this kind, it also offers
cloud connectivity plus an integrated RFID reader for screen
unlock authentication. The 4K UHD CRTtoLCD-91 flat panel
controller is based on the Himax HX6310-A scaler chip and
supports flat panel displays with 4K-UHD up to 3840×2160 at
60Hz. Thanks to content-dependent backlight control and new
scaling algorithms, displays - even of non-native 4K content -
are exceptionally brilliant and sharp. An optional industry-first
HF RFID reader with NFC support and WiFi cloud connection
can be plugged onto the controller board. Later this year, a UHF
RFID reader with a range of up to 10 meters will be added as an
option for logistics applications. With a comprehensive range of
on-screen display functions, as well as extended configuration
and scripting options for OEM-specific designs, the intelligent
LCD controller board also provides all features OEMs need to
develop custom monitors, displays and control panels. For the
connection of flatpanel displays, there are 8 VbyOne lanes at
3.75 Gbps/lane, or 4 eDP 1.2a lanes at 5.4 Gbps/lane. RGB
color formats up to 30-bit depth and CYbCr (420/422/444) are
supported. Graphics and HDCP 1.4/2.2 protected content is
handled via DisplayPort 1.4 or DVI/HDMI 2.0. Additional interfac-
es include Dolby 7.1 audio and analog stereo sound, dual USB
for touch functionality as well as keyboard/mouse and 1x USB
for client connection to embedded computer systems.
eCount Embedded
www.ecount-embedded.com
Multi-die emitter LED delivers 350lm at 2W
Designed for high bay and street lighting applications,
Lumileds’ upgraded Luxeon 5050 reaches flux levels of 350lm
and 175lm/W at 2W LED drive conditions (4000K, 70CRI at
85°C). The round LES, 4.6mm in
diameter, eases optic design, it
also meets DLC Premium V4.1 re-
quirements, enabling fixture manu-
facturers’ access to high utility
rebates and energy savings. Using
the Luxeon 5050 enables over-
all smaller system sizes, which
means very compact fixtures are
possible. The LED is available in
a range of color temperatures from 2700K to 6500K and CRI
levels (70, 80, 90) to meet a variety of high bay and street light-
ing as well as indoor spotlighting needs.
Lumileds
www.lumileds.com
eeNews Europe February 2018 News 45

DESIGN & PRODUCTS
TMR 3-axis magnetometer
sensitive to 10nT/LSB
TDK Corporation and Asahi Kasei Microdevices Corporation
have jointly developed a highly accurate 3-axis magnetometer
that combines an extremely sensi-
tive tunnel magneto-resistive (TMR)
element developed by TDK and an
advanced electronic compass ASIC
designed by AKM. The new TMR
magnetometer on a chip, measur-
ing just 1.6x1.6x0.6mm, features
the industry’s lowest RMS noise
of just 40nT-rms and a very low
current consumption of only 40µA at an output data rate of
100Hz. Thanks to its high sensitivity of 10 nT/Least Significant
Bit (LSB), the magnetic sensor can detect minute changes in
magnetic fields very precisely, enabling the highly accurate
detection of both position and orientation with the help of the
Earth’s magnetic field or a magnetic field generator. These
unique characteristics make the TMR magnetometer suitable
for compact electronic devices such as smartphones, tablets,
game console controllers and all kinds of wearable devices, as
well as other applications that demand high position and ori-
entation precision such as virtual, augmented or mixed reality
(VR, AR, and MR) or indoor navigation.
TDK
www.tdk.com
Infineon, PMD offer 3D image sensor
for face recognition
Infineon Technologies AG (Munich, Germany) has worked with
its long-time partner PMD Technologies AG (Siegen, Germany)
to develop a time-of-flight
image sensor for use in 3D
face recognition. The Real3
(or IRS238xC) enables a
camera module for integra-
tion in smartphones with a
footprint of less than 12mm
by 8mm, including the
receiving optics and VCSEL
illumination. Two previous
chips from the Infineon-PMD collaboration were the IRS1010C
with a resolution of 160 by 120 pixels (19,200 pixels) and the
IRS1020C with a resolution of 352 by 288 pixels (100k pixels).
Both are dynamically configurable via I2C interface, allowing
adjustment in real time to changing lighting and operating con-
ditions. The chips are delivered as bare die for integration with
camera lens and Infra-Red (IR) illumination source in a camera
module. The Real3 sits in between these two predecessors in
terms of resolution with 38,000 pixels. Each pixel features Sup-
pression of Background Illumination (SBI) circuitry. It is tuned
to work at 940nm wavelength infrared light sources making
the projected light invisible and improving the outdoor perfor-
mance even further. The IRS238xC also integrates a dedicated
function to support Laser-Class-1 safety level of the complete
solution. The ToF camera module is ready for mass production
and does not have to be re-calibrated during use.
PMD Technologies AG
www.pmdtec.com
46 News February 2018 @eeNewsEurope
Reader
MISCELLANEOUS O f fer
Run a software defined radio
for your IoT projects
This month, Lime Microsystems is giving away five Grove
starter kits for LimeSDR Mini, worth 200 Euros each, for
eeNews Europe’s readers to
win. Developed in partnership
with Seeed Studio, the kit is
based on the Grove Platform
and was designed for use with
the Raspberry Pi. It includes
a LimeSDR Mini software de-
fined radio platform for cellu-
lar, IoT, DVB, GNSS and other
RF applications with antennas optimised for 433/868/915 MHz
unlicensed bands, plus a GrovePi+ and an array of Grove sen-
sors for temperature, humidity, light, sound, touch, ultrasonic
range, acceleration and vibration. Other Grove extensions
include a rotary angle sensor, a relay, a buzzer, a red LED as
well as a LED bar. The Raspberry Pi (2, 3 or Zero can be used)
is not included in the pack, nor is the power supply and the
Micro SD card. The kit will get you started learning SDR basics
and developing IoT applications, it is aimed at educational use
and for beginners. By combining the hardware components
with Lime’s ScratchRadio software extension, users will be
able to quickly and intuitively create simple and fun applica-
tions that integrate SDR capabilities and peripheral I/O.
Lime Microsystems
www.limemicro.com
Check the reader offer online at
www.eenewseurope.com
7-segment LED-based SMD display
easy to stack
The DSM7T Series from Visual Communications Company
(VCC) includes blue, red and pure green LEDs designed as thin
numeric displays for a wide range of applications, including
industrial devices, light controls, medical and transportation
equipment. The DSM7T Series makes it easy to connect two
or more seven-segment displays to each other, they can be
mounted side-by-side to save space on the PCB. The SMD
form factor eliminates additional soldering and manual labour.
The devices can be mounted in top-mount or reverse-mount
configurations in blue, red and pure green. Their tape and reel
configuration make it easier than ever for mass assembly and
production. The series is available in three digit heights: 0.28”,
0.39” and 0.56” and each numeric display is compliant with
RoHS and REACH requirements and has earned a dust-protec-
tion rating of IP50.
Visual Communications Company
www.vcclite.com
www.eenewseurope.com
 MISCELLANEOUS

Ceramic / quartz-based UVC LED 0.80 and 1.00mm pitch edge card sockets
outputs 3.8mW at 20mA support up to 56Gbps PAM4
Vishay Intertechnology’s VLMU60CL..-280-125 is a new The 0.80mm pitch socket (HSEC8-DP Series) is a differential
ceramic / quartz-based UVC (short wavelength ultraviolet) pair version of Samtec’s popular Edge Rate 0.80 mm pitch
emitting diode designed sockets. Rated for speeds to
to replace mercury UVC 28 Gbps NRZ/56 Gbps PAM4,
lamps for sterilization, the socket features Edge Rate
sanitation, and purification contacts designed to increase
applications. Offered in a cycle life and decrease cross-
6x6x1.6mm surface-mount talk. The double row socket,
package, the device is with 8 to 56 total pairs and op-
claimed to deliver an ex- tional rugged weld tabs, mates
tremely long lifetime. Com- with standard 1.60mm thick
parable to devices in the PLCC2 package with emission angles cards. Optimized for performance to 28Gbps, the 1.00mm
of ±60°, the new device has an emission angle of ± 62.5° and pitch edge card socket (HSEC1 Series) also features signal
high radiant power to 3.8mW at 20mA without the need for an integrity optimized Edge Rate contacts with alignment pin and
external lens. Built on AlGaN technology, the UVC LED features rugged weld tab options. The socket is available with up to 140
forward current to 40mA, forward voltage down to 4.4V, and a total positions. For specific applications, this edge card socket
wavelength range of 270nm to 290nm. can help mitigate misalignment in the X and Y axes.
Vishay Intertechnology Samtec
www.vishay.com www.samtec.com

USound, ST deliver samples of

worked with niobium-doped lead
MEMS micro-speakers zirconate titanate (Nb-doped PZT)
MEMS micro-speaker startup as a piezoelectric material and
USound GmbH (G raz, Austria) PZT is thought to be the mate-
has delivered its first engineering rial of choice within ST’s PeTra
samples of silicon MEMS micro- (piezoelectric transducer) thin-film-
speakers to lead customers. This piezoelectric technology. USound
puts the collaboration behind a intended to use packaging tech-
previous prediction. Ten months ago nology to integrate ASICs, MEMS
when ST was revealed as USound’s and passive components in one
manufacturing partner ST predicted device with a thin and compact
the piezoelectric MEMS actuator form factor. With less than half
would move into volume production in 3Q17 and would ship in the weight of conventional micro-speakers this should provide
consumer products before the end of 2017. A spokesperson for opportunities for design wins in wearable technology such
ST said mass production is now planned in 2Q18 with a design as headphones and augemented reality/virtual reality head-
win already in hand. USound has a patented micro-speaker that gear. ST said USound plans to demonstrate prototype AR/VR
replaces the conventional balanced-armature electrodynamic glasses containing multiple MEMS speakers per side, to invited
transducers for handsets with a piezoelectric MEMS actuator. guests at ST’s private suite during CES 2018. According to ithe
The expectation is that this will reduce power consumption USound website the company has Moon and Ganymede chip
while improving sound quality. The production delay was due products for in-sound and near-sound applications respectively,
to a minor change of design to further improve performance and is developing a variety of reference designs.
and to extend the range of applications covered by the first USound GmbH
product in the market, the spokesperson said. In R&D USound www.usound.com

Plug-in board expands motion

controller with EtherCAT
Maxon motor is equipping its EPOS4 posi-
tioning controllers with additional connectivi-
ty options so they can now be integrated into
any EtherCAT network as a slave by using
an expansion board. The EPOS4 positioning
controllers are now able to communicate via
EtherCAT using the CoE standard (CAN ap-
plication layer over EtherCAT). The EtherCAT
card can be combined with the EPOS4 controllers in a housing,
as well as with all EPOS4 modules. This expansion for the world
of Ethernet-based communication
protocols further expands the areas of
application for the compact positioning
controllers. For example, in applica-
tions where short cycle times are
important in synchronized multi-axis
systems. EPOS4 controllers are suit-
able for controlling both brushed and
brushless DC motors.
Maxon motor
www.maxonmotor.com

www.eenewseurope.com eeNews Europe February 2018 News 47

 DISTRIBUTION CORNER
RS Components takes DesignSpark Sigfox’ SDR dongle now at Digi-Key
to Barclays Eagle Labs Designed for device and solution makers to speed up develop-
RS Components has formed a strategic alliance with Eagle ment and tests, the Sigfox SDR dongle is now available glob-
Labs, a Barclays initiative to foster innovation in communities. ally through Digi-Key Electronics, allowing for the development
As technology partner, RS of Sigfox devices
will support Eagle Labs with independent from
world-leading engineering Sigfox public net-
equipment and technical work coverage. A
expertise from its RS Lo- configurable ra-
cal innovation centres and dio receiver and
DesignSpark. Launched two transmitter, the
years ago within underutilised Sigfox SDR don-
Barclays high street branches, gle is the hardware platform on which to run the Sigfox Device
there are now 13 Eagle Labs providing spaces that offer open Test Suite, a comprehensive device developer software toolkit
and membership-based access to digital fabrication tools, a including a Network Emulator and a Radio Signal Analyzer. This
LoRaWAN SigFox base station, education workshops, rapid pro- mean users can test end-to-end and bidirectional data trans-
totyping facilities and a vibrant community. port services between a Sigfox device and their application.
RS Components Digi-Key Electronics
www.rs-online.com www.digikey.com

OrCAD Entrepreneur on Arrow.com Infineon AEC-Q100 qualified

accelerates PCB design angle sensor at Rutronik
Arrow Electronics and Cadence Design Systems have expanded The 360° angle sensor TLE5012BD from Infineon with two sen-
their ongoing collaboration, launching the Cadence OrCAD sor ICs in one package has been developed for the detection
Entrepreneur package of angular positions in automotive
for OrCAD Capture applications, e. g. in electrically com-
Cloud on the Arrow. mutated motors, for the measurement
com design center. The of steering angles or the angle detec-
new software tool suite tion in general. The sensor detects
is said to cut in half the alignment of a magnetic field by
the time that it takes a measuring sine and cosine voltages
designer to move from with monolithically integrated Giant
design to prototype, when compared to the traditional process Magneto Resistance (iGMR) elements.
of manually designing and simulating across separate systems. With the aid of an internal auto-calibration algorithm, it records
Designers use OrCAD Entrepreneur to capture their system-level the angular values with high precision over a wide temperature
design in the familiar OrCAD Capture Cloud design environment, range and a long service life. The TLE5012BD is pre-calibrated,
which is accessible directly from Arrow.com. the calibration parameters are stored in laser fuses.
Arrow Electronics Rutronik
www.arrow.com www.rutronik24.com

Mouser signs distribution 3-channel optical encoder in a

for Onion’s IoT offering 7x13mm aluminium housing
Mouser Electronics has signed a global distribution agreement Elesta has added a 3-channel optical encoder in 7x13mm alu-
with Onion Corp., a global provider of highly integrated wireless minium housing to its wide product portfolio, making it one of the
microprocessor modules smallest on the market. The
and Internet of Things (IoT) device is fitted with two preci-
development kits. Through sion ball bearings guarantee-
the agreement, Mouser will ing exceptionally smooth run-
distribute the Omega2+ de- ning and a long service life at
vice, kits, and accessories, speeds of up to 80,000 rpm.
ideal for applications such Inside the miniature housing
as home automation, coding is the patented unique Elesta
education, Wi-Fi media servers, robotics, and networking. The encoder system - the E OI R,
Onion product line revolves around the Omega2+, an easy-to-use, an integrated system operating on the basis of a new optical fo-
expandable IoT computer packed with built-in Wi-Fi connectivity, cusing incidental light method. The encoder delivers 128 pulses at
a MicroSD card slot, and a powerful 580 MHz MIPS processor. any rotation speed, to two outputs that are phase shifted by 90°.
Mouser Willow Technologies
www.mouser.com www.willow.co.uk

48 News February 2018 @eeNewsEurope www.eenewseurope.com

 LAST WORD

Accelerating your quality system’s

actionability in manufacturing
By Bob Miklosey

I
rrespective of company size and another system. These actions are then
scale, quality systems play an essen- individually printed and assembled into
tial role for manufacturers in mitigat- a packet, and physically accompany the
ing risk and meeting compliance with related production units in process as
industry standards like AS9100 (for the part of its traveler. Throughout the entire
aerospace industry). What can differenti- case example above, the manufacturer
ate one manufacturer’s quality system has actually met the explicit require-
from another? How readily they take ac- ments for data management in aero-
tion against a non-conformance and how space manufacturing, but the manner in
frequently they report key quality metrics, which it has been achieved is not in any
which is enabled through a modern way efficient, cost effective, simple, or
manufacturing execution system (MES). expedient.
In absence of MES and having pro- Beyond the hidden costs of manag-
duction information exist in digital form, ing a spectrum of isolated tools varying
a critical event may find itself reported in age and condition as depicted above,
in a historical context and often too late a key quality challenge is the ability to
to properly react to its findings. MES is agilely react and report when a non-
the essential complement to the orga- conformance occurs within this environ-
nization’s quality management system (QMS), where together ment. Factory floor personnel are in effect conditioned to limit
they can accelerate process development, improve production their defect collection as it becomes more of an event and less
control, deliver regulatory compliance, and provide real time vis- of a natural activity while performing their operations.
ibility to management. Because these occurrences may require entry found in
In many manufacturing landscapes, a formal MES system multiple systems or paper forms and can consume a significant
still is not present. Instead, the ‘system’ in place is often a amount of time to report, there may even be rules imposed
combination of point solutions and paper processes. This upon the inspector to only submit issues requiring more than
archipelago of components accidentally evolved through solv- ‘n’ minutes to correct, leaving a majority of in-process quality
ing a given challenge critical at a given moment. Consider the issues unreported.
situation within a particular factory; test results needed to be The result is an inaccurate reflection of quality metrics and
reported to an end customer and a database was created to an inability to report them in an effective timeframe for broader
store its collected data. Material counterfeiting emerged as a correction, prevention, and improvement to occur.
concern where traceability became mandatory and this chal-
lenge was addressed by stock-
room personnel writing each
lot number on a picking list. A
customer required all torque
and other measurements to be
recorded as part of their unit build
history so in response, operators
record parametric data in a web
form. However, most production
areas still follow work instructions
developed in an office applica-
tion.
The quality policy requires in-
spectors to record defects found
on the product onto a tick sheet
and then add details to a ‘toe tag’
traveling with the product for the
benefit of a technician to perform
a repair, and if severe, they then
duplicate the effort to initiate
a non-conformance event in
The experience radically changes when MES is implemented
Bob Miklosey is the Vice President of Product Management at in a holistic approach and one that is more natural for factory
Aegis Software - www.aiscorp.com floor personnel, engineers, and management.

www.eenewseurope.com eeNews Europe February 2018 News 49

 News EUROPE
LAST WORD
PUBLISHER Quality functions, while essential, are
André Rousselot only one aspect to the overall opera-
+32 27400053 tional coverage of MES. When properly
andre.rousselot@electronicseurope.net designed, MES seamlessly considers all
EDITOR-IN-CHIEF activities conducted at an operation, from
Julien Happich setup to execution, whether performed
+33 169819476 by a person or a machine.
julien.happich@electronicseurope.net At a manual process for instance,
the first quality control occurs when the
EDITORS operator’s certification is validated to
Christoph Hammerschmidt ensure they possess the appropriate
+49 8944450209 skills to perform the required tasks. The
chammerschmidt@electronicseurope.net operators are then instructed to prepare
Peter Clarke their workstations and have the required
+44 776 786 55 93 material and tooling validated for use.
peter.clarke@electronicseurope.net Their work begins by either entering
Nick Flaherty a build quantity or inputting a barcode
+44 7710 236368 for serialized unit tracking, where the
nick.flaherty@electronicseurope.net workpiece itself is then validated to
Jean-Pierre Joosting ensure it is properly traveling along its
+44 7800548133 route sequence and possesses a valid
jean-pierre.joosting@electronicseurope.net status to proceed. They then dynami-
Alistair Winning cally interact with paperless instructions,
+44 7834 770342 engaging with them at a granular level
ally.winning@electronicseurope.net through each activity they perform while
CIRCULATION & FINANCE being presented with related detail found
Luc Desimpel in documents, videos, and drawings, and
luc.desimpel@electronicseurope.net prompting them to capture measure-
ments, serialized subassemblies, and
ADVERTISING PRODUCTION & other attributes that are also validated.
REPRINTS A supervisor or peer may then be
Lydia Gijsegom prompted to review and approve any or
lydia.gijsegom@electronicseurope.net all of these activities. The totality of this
ART MANAGER
experience instills quality through suc-
Jean-Paul Speliers
cessful closed-loop checks and preven-
tive measures without degrading produc-
ACCOUNTING tion line performance or ergonomics.
Ricardo Pinto Ferreira Whether a given operation is primar-
ily intended for assembly, processing,
REGIONAL ADVERTISING
inspection, or test, it must be thought
REPRESENTATIVES
of as an opportunity for quality. While
Contact information at:
an operator may be performing a pure
http://www.eenewseurope.com/content/
assembly task, they may uncover an
sales-contacts
abnormality needing to be reported.
eeNews Europe MES enables all personnel to serve in
is published 11 times in 2018
a quality role and capture all in pro-
by European Business Press SA
cess defects in the same manner as
Chaussée de Louvain 533,
an inspector dedicated to a QC step.
1380 Lasne, Belgium
Advanced MES systems further enhance
Tel: +32-2-740 00 50 european the defect collection experience through
Fax: +32-2-740 00 59 business press
graphical onscreen input referencing
email: info@electronics europe.net
the product design CAD and imagery.
www.eenewseurope.com
Technicians diagnosing test failures on a
VAT Registration: BE 461.357.437.
top-level assembly may identify the root
Company Number: 0461357437
cause by relating it to one or multiple
RPM: Nivelles.
defects found within a child assembly.
Volume 20, Issue 2 eeNews P 304128
It is in the hands of the system’s quality
It is free to qualified engineers and managers
administrators and the quality policy they
involved in engineering decisions – see:
define to identify which events automati-
http://www.eenewseurope.com/subscribe
cally trigger a nonconformance for further
© 2018 E.B.P. SA
review and disposition as part of their
All rights reserved. P 304128
corrective and preventive action (CAPA)
process, relieving the need for redundant
50 News February 2018 @eeNewsEurope
data entry in a secondary system and all
of which are naturally occurring through
the operator’s actions.
The byproduct of all these actions
is the rich analytical data living within a
centralized MES database, immediately
available for reporting and action, and
bringing KPIs and quality metrics to
life in real time. Staff outside the office
view current production activity on their
mobile devices.
Managers receive daily quality reports
directly sent to their inbox. Engineers
conduct ad hoc data analyses on drill
down charts and pivot tables at weekly
action meetings. Shop floor person-
nel and visiting guests view dashboards
running overhead displaying real-time
metrics from areas, lines, and equipment.
MES not only changes how data is col-
lected, it engages all organization levels
to take an active role in quality.
A properly designed MES system
readily adjusts to the unique quality chal-
lenges found in aerospace and defense
manufacturing. It adapts to the manufac-
turing process breadth found inside an
enterprise’s manufacturing operations,
which can include machining, forming,
electronics, cable/harness, module,
and large-scale assembly dwelling over
extended periods.
It eliminates the risk of an opera-
tor referencing the wrong BOM or work
instruction revision and actively informs
them of change notices (ECNs) targeted
for the product and the process they are
performing. It follows the product’s com-
plete lifecycle and maintains its unified
history through its initial production, ad
hoc re-routes for repair, refurbishment,
field return, MRO, and any other process
it undergoes.
When MES is holistic in its adoption
and implementation it drives a manufac-
turing culture of in-process quality from
beginning to end in the enterprise. Both
personnel and end customers benefit
from the immediacy of its information and
reaction, where reliance upon its preced-
ing fragmented methods are unthinkable.
Its impact on the overall quality system is
unmistakable, serving in an indispensable
capacity as the in-process contributor to
a quality system for true actionability.
MES is thereby treated as essential
of a pillar to the organization’s systems
as ERP, completing the manufacturers’
digital footprint for attaining the highest
quality output and achieving operational
excellence.
www.eenewseurope.com
 Your Global Link to the Electronics World

www.eenewsanalog.com www.eenewsled.com

www.eenewseurope.com/newsletters www.eenewsautomotive.com

www.eenewspower.com www.mwee.com
REDEXPERT
embedded world Hall 3 Booth 247
REDEXPERT. Würth Elektronik‘s online platform for simple
component selection and performance simulation.
 The world’s most accurate AC loss model
 Filter settings for over 20 electrical and
mechanical parameters
 Inductor simulation and selection for
DC/DC converters
 Ability to compare inductance/current and
temperature rise/DC current using interactive
measurement curves
#REDEXPERT
 Available in seven languages
 Online platform based on measured
values
 No login required
 Order free samples directly
 Direct access to product datasheets
www.we-online.com/redexpert