Data Modelling and Knowledge Engineering

for the Internet of Things

Wei Wang1, Cory Henson2, Payam Barnaghi1

Centre for Communication Systems Research, University of Surrey

Kno.e.sis Center, Wright State University

Galway City, Ireland, October 8-12, 2012

http://knoesis.org/iot-tutorial-ekaw2012/
 Part 1: Introduction
to Internet of “Things”

Image source: CISCO

2
Internet of Things

 ―sensors and actuators embedded in physical

objects — from containers to pacemakers — are
linked through both wired and wireless networks to
the Internet.‖
 ―When objects in the IoT can sense the environment,
interpret the data, and communicate with each
other, they become tools for understanding
complexity and for responding to events and
irregularities swiftly‖
source: http://www.iot2012.org/
―Thing‖ connected to the internet

Source: CISCO 4
Future Internet - A new dimension

55
 Internet of Things - definition

 ―A world where physical objects are seamlessly

integrated into the information network, and where
the physical objects can become active participants
in business processes.‖
 ―Services are available to interact with these ―smart
objects‖ over the Internet, query and change their
state and any information associated with them,
taking into account security and privacy issues. ‘‖.

Source: Stephan Haller, Internet of Things: An integral Part of the Future Internet, SAP Research, 2009.
6
 What ―Things‖ can be connected?

Home/daily-life devices
Business and
Public infrastructure
Health-care
…
 Sensor devices are becoming widely
available
- Programmable devices
- Off-the-shelf gadgets/tools
Application domain
Why is IoT important?
 Observation and measurement data

Adapted from: W3C Semantic Sensor Networks, SSN Ontology presentation, http://www.w3.org/2005/Incubator/ssn/
Data is important and IoT will produce
lots of it!
 Sensors and devices provide data about the physical world objects.
 The observation and measurement data related to an ―object‖ can be
related to an event, situation in the physical world.
 The processing of turning this data into knowledge/ perception and
using it for decision making, automated control, etc. is another important
phase.
 Huge amount of data related to our physical world that need to be
 Published
 Stored (temporary or for longer term)
 Discovered
 Accessed
 Proceeded
 Utilised in different applications
Turning Data into Wisdom
The ―Things‖

 Embedded device + physical world objects

 Sensor nodes (e.g. SunSPOT, TelOSB, WASPmote).
 Mobile devices (e.g. mobile phones, tablets)

 A set of these that provide information about (a

feature of interest of) a physical world object (e.g.
water level in a tank, temperature of a room).
Components related to ―Things‖

 Physical world objects

 e.g. A room, a car, A person;
 Feature of Interest
 e.g.Temperature of the room, Location of the car,
heart-rate of the person;
 Sensors
 e.g. Temperature sensor, GPS, pulse sensor
 Embedded device
 e.g. WASPmote, SunSPOT, …
Sensors

 Active & Passive Sensors

 Energy Efficiency
 Processing capabilities
 Network communications
 hardware platforms
 software platforms
RFID

 Active Tags and Passive Tags

 Applications: supply chain, inventory tracking, tools
collection, etc.
 Limitations:
 Technology
 Reading range
 Physical limitations
 Interference
 Security and Privacy
Hardware components of sensor
nodes
 Controller
 Memory
 Communication device
 Sensors (or actuators)
 Power supply
 Example: Radiation Sensor Board
(Libelium)

Waspmote

Source: Wireless Sensor Networks to Control Radiation Levels, David Gascón, Marcos Yarza, Libelium, April 2011.
Energy consumption of the nodes

 Batteries have small capacity and recharging could

be complex (if not impossible) in some cases.
 The main consumers of the energy are: the
controller, radio, to some extent memory and
depending on the type, the sensor(s).
 A controller can go to:
 ―active‖, ―idle‖ and ―sleep‖
 A radio modem could turn transmitter, receiver, or
both on or off,
 sensors and memory can be also turned on and off.
 Beyond common sensors

 Human as a sensor
 e.g. tweeting real world data and/or events
 Virtual sensors
 e.g. Software agents generating data

Adapted from: The Web of Things, Marko Grobelnik, Carolina Fortuna, Jožef Stefan Institute.
 Actuators

[2]

Stepper Motor [1]

[4] [3]

Image credits:
[1] http://directory.ac/telco-motion.html
[2] http://bruce.pennypacker.org/category/theater/
[3] http://www.busytrade.com/products/1195641/TG-100-Linear-Actuator.html
[4] http://www.arbworx.com/services/fencing-garden-fencing/
 Wireless Sensor Networks (WSN)

Image source: Protocols and Architectures for Wireless Sensor Networks, Protocols and Architectures for Wireless Sensor Networks
Holger Karl, Andreas Willig, chapter 3, Wiley, 2005 .
Wireless Sensor Networks (WSN)-
gateway connection

Gateway
SunSpots
Directory server
Web user/application
Control channel
Information channel
Distributed WSN
What are the main issues?

 Heterogeneity
 Interoperability
 Mobility
 Energy efficiency
 Scalability
 Security
What is important?

 Robustness
 Quality of Service
 Scalability
 Seamless integration
 Security, privacy, Trust
 In-network processing

 Mobile Ad-hoc Networks are supposed to deliver bits from

one end to the other
 WSNs, on the other end, are expected to provide
information, not necessarily original bits
 Gives addition options
 E.g., manipulate or process the data in the network
 Main example: aggregation
 Applying aggregation functions to a obtain an average value of
measurement data
 Typical functions: minimum, maximum, average, sum, …
 Not amenable functions: median
source: Protocols and Architectures for Wireless Sensor Networks, Protocols and Architectures for Wireless Sensor Networks
Holger Karl, Andreas Willig, chapter 3, Wiley, 2005 .
 In-network processing- example

Applying Symbolic Aggregate Approximation (SAX)

SAX Pattern (blue) with word length of 20 and a vocabulary of 10 symbols

over the original sensor time-series data (green)
 Data-centric networking

 In typical networks (including ad hoc networks), network

transactions are addressed to the identities of specific nodes
 A ―node-centric‖ or ―address-centric‖ networking paradigm
 In a redundantly deployed sensor networks, specific source
of an event, alarm, etc. might not be important
 Redundancy: e.g., several nodes can observe the same area
 Thus: focus networking transactions on the data directly
instead of their senders and transmitters ! data-centric
networking
 Principal design change

source: Protocols and Architectures for Wireless Sensor Networks, Protocols and Architectures for Wireless Sensor Networks
Holger Karl, Andreas Willig, chapter 3, Wiley, 2005 .
 Implementation options for
data-centric networking
 Overlay networks & distributed hash tables (DHT)
 Hash table: content-addressable memory
 Retrieve data from an unknown source, like in peer-to-peer networking – with efficient
implementation
 Some disparities remain
 Static key in DHT, dynamic changes in WSN
 DHTs typically ignore issues like hop count or distance between nodes when performing a
lookup operation
 Publish/subscribe
 Different interaction paradigm
 Nodes can publish data, can subscribe to any particular kind of data
 Once data of a certain type has been published, it is delivered to all subscribes
 Subscription and publication are decoupled in time; subscriber and published are agnostic
of each other (decoupled in identity);
 There is concepts of Semantic Sensor Networks- to annotate sensor resources and
observation and measurement data!
Adapted from: Protocols and Architectures for Wireless Sensor Networks, Protocols and Architectures for Wireless Sensor Networks
Holger Karl, Andreas Willig, chapter 3, Wiley, 2005 .
IoT and Semantic technologies

 The sensors (and in general ―Things‖) are increasingly being

integrated into the Internet/Web.
 This can be supported by embedded devices that directly
support IP and web-based connection (e.g. 6LowPAN and
CoAp) or devices that are connected via gateway
components.
 Broadening the IoT to the concept of ―Web of Things‖
 There are already Sensor Web Enablement (SWE)
standards developed by the Open Geospatial Consortium
that are widely being adopted in industry, government and
academia.
 While such frameworks provide some interoperability,
semantic technologies are increasingly seen as key enabler
for integration of IoT data and broader Web information
systems.
Semantics and IoT resources and
data
 Semantics are machine-interpretable metadata (for mark-up), logical
inference mechanisms, query mechanism, linked data solutions
 For IoT this means:
 ontologies for: resource (e.g. sensors), observation and measurement
data (e.g. sensor readings), domain concepts (e.g. unit of measurement,
location), services (e.g. IoT services) and other data sources (e.g. those
available on linked open data)
 Semantic annotation should also supports data represented using
existing forms
 Reasoning /processing to infer relationships and hierarchies between
different resources, data
 Semantics (/ontologies) as meta-data (to describe the IoT
resources/data) / knowledge bases (domain knowledge).
A Few Words
on
Semantic Web

34
 Semantic Web
SSW Introduction (according to Farside)

Concrete Facts
Resource Description Framework
lives in
General Knowledge
Web Ontology Language

has pet
Person Animal
has pet is a

is a

“Now! – That should clear up a few things around here!”

Semantic Web Stack
Linked Open Data
Linked Open Data

~ 50 Billion Statements
SW is moving from academia
to industry
In the last few years, we have seen
many successes …

Apple
Siri

Watson

Knowledge Graph
Google Knowledge Graph
Sensors and the Web

42
Sensors are ubiquitous
Sensors are small and inexpensive
Digitization of the physical world
 Leading to …

 Improved situational
awareness

 Advanced cyber-physical
systems / applications

 Enabling the Internet of

Things
Enabling the Internet of Things

Situational awareness enables:

 Devices/things to function and

adapt within their environment

 Devices/things to work
together
Sensor systems are too
often stovepiped.

 Closed centralized
management of sensing
resources

 Closed inaccessible data

and sensors
We want to set this data free

With freedom comes responsibility

 Discovery, access, and search
 Integration and interpretation
 Scalability
 Drowning in Data

A cross-country flight from New York to Los Angeles on a Boeing

737 plane generates a massive 240 terabytes of data
- GigaOmni Media
Drowning in Data

In the next few years, sensor networks will produce 10-20

time the amount of data generated by social media.
- GigaOmni Media
Drowning in Data
Challenges

To fulfill this vision, there are difficult challenges to overcome such as the
discovery, access, search, integration, and interpretation of sensors and
sensor data at scale

 Discovery finding appropriate sensing resources and data sources

 Access sensing resources and data are open and available
 Search querying for sensor data
 Integration dealing with heterogeneous sensors and sensor data
 Interpretation translating sensor data to knowledge usable by people and
applications
 Scalability dealing with data overload and computational complexity
of interpreting the data
Solution

Semantic Sensor Web

 Internet Computing, July/Aug. 2008
 Uses the Web as platform for
managing sensor resources and data
 Uses semantic technologies for
representing data and knowledge,
integration, and interpretation
Solution

Discovery, access, and search

 Using standard Web services
 OGC Sensor Web Enablement
Solution

Integration
 Using shared domain models / data representation
 OGC Sensor Web Enablement
 W3C Semantic Sensor Networks
Solution

Interpretation
 Abstraction – converting low-level data to high-level knowledge
 Machine Perception – w/ prior knowledge and abductive reasoning
 IntellegO – Ontology of Perception
Solution

Scalability
 Data overload – sensors produce too much data
 Computational complexity of semantic interpretation
 ―Intelligence at the edge‖ – local and distributed integration and
interpretation of sensor data
SSW Adoption and Applications
 Part 2: Semantic Modelling
for the Internet of “Things”

Image source: semanticweb.com; CISCO

60
Recall of the Internet of Things

 A primary goal of interconnecting devices and

collecting/processing data from them is to create
situation awareness and enable applications,
machines, and human users to better understand
their surrounding environments.
 The understanding of a situation, or context,
potentially enables services and applications to
make intelligent decisions and to respond to the
dynamics of their environments.
Barnaghi et al 2012, ―Semantics for the Internet of Things: early progress and back to the future‖
IoT challenges

 Numbers of devices and different users and interactions required.

 Challenge: Scalability
 Heterogeneity of enabling devices and platforms
 Challenge: Interoperability
 Low power sensors, wireless transceivers, communication, and networking for M2M
 Challenge: Efficiency in communications
 Huge volumes of data emerging from the physical world, M2M and new
communications
 Challenge: Processing and mining the data, Providing secure access and preserving and
controlling privacy.
 Timeliness of data
 Challenge: Freshness of the data and supporting temporal requirements in accessing the
data
 Ubiquity
 Challenge: addressing mobility, ad-hoc access and service continuity
 Global access and discovery
 Challenge: Naming, Resolution and discovery
IoT: one paradigm, many visions

Diagram adapted from L. Atzori et al., 2010, ―the Internet of Things: a Survey‖
Semantic oriented vision

 ―The object unique addressing and the representation and

storing of the exchanged information become the most
challenging issues, bringing directly to a ‗‗Semantic oriented‖,
perspective of IoT‖, [Atzori et al., 2010]
 Data collected by different sensors and devices is usually
multi-modal (temperature, light, sound, video, etc.) and diverse
in nature (quality of data can vary with different devices
through time and it is mostly location and time dependent
[Barnaghi et al, 2012]
 some of challenging issues: representation, storage, and
search/discovery/query/addressing, and processing IoT
resources and data.
What is expected?

 Unified access to data: unified descriptions

 Deriving additional knowledge (data mining)
 Reasoning support and association to other entities and
resources
 Self-descriptive data an re-usable knowledge
 In general: Large-scale platforms to support discovery and
access to the resources, to enable autonomous interactions with
the resources, to provide self-descriptive data and association
mechanisms to reason the emerging data and to integrate it
into the existing applications and services.
Semantic technologies and IoT

 There are already Sensor Web Enablement (SWE)

standards developed by the Open Geospatial
Consortium that are widely adopted.
 While such frameworks provide certain levels of
interoperability, semantic technologies are seen as
key enabler for integration of IoT data and and
existing business information systems.
 Semantic technologies provide potential support for:
 Interoperability and machine automation
 IoT resource and data annotation, logical inference, query and
discovery, linked IoT data
Identify IoT domain concepts

 Users
 Physical entities
 Virtual entities
 Devices
 Resource
 Services
 …

Diagram adapted from IoT-A project D2.1

IoT domain concepts - Entity

 Physical entities (or entity of interests): objects in the

physical world, features of interest that are of
interests to users (human users or any digital
artifacts).
 Virtual entities: virtual representation of the physical
entities.
 Entities are the main focus of interactions between
humans and/or software agents.
 This interaction is made possible by a hardware
component called Device.
Definition adapted from De et al, 2012, “Service modeling for the Internet of Things”
IoT domain concepts –
Device, Resource and Service
 A Device mediates the interactions between users and
entities.
 The software component that provides information on the
entity or enables controlling of the device, is called a
Resource.
 A Service provides well-defined and standardised
interfaces, offering all necessary functionalities for
interacting with entities and related processes.

Definition adapted from De et al, 2012, “Service modeling for the Internet of Things”
Other concepts need to considered

 Gateways
 Directories
 Platforms
 Systems
 Subsystems
 …
 Relationships among them
 And links to existing knowledge base and linked data
Don‘t forget the IoT data

 Sensors and devices provide observation and measurement

data about the physical world objects which also need to be
semantically described and can be related to an event,
situation in the physical world.
 The processing of data into knowledge/ perception and using
it for decision making, automated control, etc.
 Huge amount of data from our physical world that need to be
 Annotated
 Published
 Stored (temporary or for longer term)
 Discovered
 Accessed
 Proceeded
 Utilised in different applications
Semantics for IoT resources and data

 Semantics are machine-interpretable metadata, logical inference

mechanisms, query and search mechanism, linked data…
 For IoT this means:
 ontologies for: resource (e.g. sensors), observation and measurement
data (e.g. sensor readings), services (e.g. IoT services), domain concepts
(e.g. unit of measurement, location) and other data sources (e.g. those
available on linked open data)
 Semantic annotation should also supports data represented using existing
forms
 Reasoning/processing to infer relationships between different resources
and services, detecting patterns from IoT data
Characteristics of IoT resources

 Extraordinarily large number

 Limited computing capabilities
 Limited memory
 Resource constrained environments (e.g., battery
life, signal coverage)
 Location is important
 Dynamism in the physical environments
 Unexpected disruption of services
 …
Characteristics of IoT data

 Stream data (depends on time)

 Transient nature
 Almost always related to a phenomenon or quality
in our physical environments
 Large amount
 Quality in many situations cannot be assured (e.g.,
accuracy and precision)
 Abstraction levels (e.g., raw, inferred or derived)
 …
Utilise semantics

 Find all available resources (which can provide data)

and data related to “Room A” (which is an object in
the linked data)?
 What is “Room A”? What is its location? returns “location” data
 What type of data is available for “Room A” or that “location”?
(sensor category types)
 Predefined Rules can be applied based on available
data
 (TempRoom_A > 80°C) AND (SmokeDetectedRoom_A position==TRUE) 
FireEventRoom_A
 Learning these rules needs data mining or pattern recognition techniques
Semantic modelling

 Lightweight: experiences show that a lightweight ontology

model that well balances expressiveness and inference
complexity is more likely to be widely adopted and reused;
also large number of IoT resources and huge amount of data
need efficient processing
 Compatibility: an ontology needs to be consistent with those
well designed, existing ontologies to ensure compatibility
wherever possible.
 Modularity: modular approach to facilitate ontology evolution,
extension and integration with external ontologies.
Existing models for resources and data

 W3C Semantic Sensor Network Incubator Group‘s

SSN ontology (mainly for sensors and sensor
networks, observation and measurement, and
platforms and systems)
 Quantity Kinds and Units
 Used together with the SSN ontology
 based on QUDV model OMG SysML(TM)

 Working group of the SysML 1.2 Revision Task Force

(RTF) and W3C Semantic Sensor Network Incubator
Group
Existing models for services

 OWL-S and WSMO are heavy weight models: practical use?

 Minimal service model
 Deprecated
 Procedure-Oriented Service Model (POSM) and Resource-Oriented
Service Model (ROSM): two different models for different service
technologies
 Defines Operations and Messages
 No profile, no grounding
 SAWSDL: mixture of XML, XML schema, RDF and OWL
 hRESTS and SA-REST: mixture of HTML and reference to a
semantic model; sensor services are not anticipated to have
HTML
 W3C‘S SSN ontology

Diagram adapted from SSN report

Some existing IoT models and ontologies

 FP7 IoT-A project‘s Entity-Resource-Service ontology

A set of ontologies for entities, resources, devices and
services
 Based on the SSN and OWL-S ontology

 FP7 IoT.est project‘s service description framework

A modular approach for designing a description
framework
 A set of ontologies for IoT services, testing and
QoS/QoI
 Technology independent modelling for services
IoT-A resource model

Diagram adapted from IoT-A project D2.1

IoT-A resource description

Diagram adapted from IoT-A project D2.1

IoT-A service model

Diagram adapted from IoT-A project D2.1

IoT-A service description

Diagram adapted from IoT-A project D2.1

Service modelling in IoT.est

Diagrams adapted from Iot.est D3.1

IoT.est service profile highlight

 ServiceType class represents the service technologies: RESTful

and SOAP/WSDL services.
 serviceQos and serviceQoI are defined as subproperty of
serviceParameter; they link to concepts in the QoS/QoI
ontology.
 serviceArea: the area where the service is provided; different
from the sensor observation area
 Links to the IoT resources through ―exposedBy‖ property
 Future extension:
 serviceNetwork, servicePlatform and serviceDeployment
 Service lifecycle, SLA…
Linked data principles

 using URI’s as names for things: Everything is

addressed using unique URI’s.
 using HTTP URI’s to enable people to look up those
names: All the URI’s are accessible via HTTP
interfaces.
 provide useful RDF information related to URI’s
that are looked up by machine or people;
 including RDF statements that link to other URI’s to
enable discovery of other related concepts of the
Web of Data: The URI’s are linked to other URI’s.
Linked data in IoT

 Using URI’s as names for things;

- URI’s for naming M2M resources and data (and also streaming data);

 Using HTTP URI’s to enable people to look up those names;

- Web-level access to low level sensor data and real world resource
descriptions (gateway and middleware solutions);
 Providing useful RDF information related to URI’s that are looked up by
machine or people;
- publishing semantically enriched resource and data descriptions in the
form of linked RDF data;
 Including RDF statements that link to other URI’s to enable discovery of
other related things of the web of data;
- linking and associating the real world data to the existing data on the
Web;
Linked data layer for not only IoT…

Images from Stefan Decker, http://fi-ghent.fi-week.eu/files/2010/10/Linked-Data-scheme1.png; linked data diagram: http://richard.cyganiak.de/2007/10/lod/

Creating and using linked sensor data

http://ccsriottb3.ee.surrey.ac.uk:8080/IOTA/
Sensor discovery using linked sensor data
Semantics in IoT - reality

 If we create an Ontology our data is interoperable

 Reality: there are/could be a number of ontologies for a domain
 Ontology mapping
 Reference ontologies
 Standardisation efforts
 Semantic data will make my data machine-understandable and my system will be
intelligent.
 Reality: it is still meta-data, machines don‘t understand it but can interpret it. It still does need
intelligent processing, reasoning mechanism to process and interpret the data.
 It‘s a Hype! Ontologies and semantic data are too much overhead; we deal with
tiny devices in IoT.
 Reality: Ontologies are a way to share and agree on a common vocabulary and knowledge; at
the same time there are machine-interpretable and represented in interoperable and re-usable
forms;
 You don‘t necessarily need to add semantic metadata in the source- it could be added to the
data at a later stage (e.g. in a gateway);
 Part 3: Semantic Sensor Web
and
Perception

Image source: semanticweb.com; CISCO

93
Introducing the Sensor Web
What is the Sensor Web?

 Sensor Web is an additional layer connecting sensor networks

to the World Wide Web.

 Enables an interoperable usage of sensor resources by

enabling web based discovery, access, tasking, and alerting.

 Enables the advancement of

cyber-physical applications through
improved situation awareness.
Why is the Sensor Web important?

 In general
 Enable tight coupling of the cyber and physical
world

 In relation to IoT
 Enable shared situation awareness (or context)
between devices/things
 Bridging the Cyber-Physical Divide

Psyleron’s Mind-Lamp (Princeton U),

connections between the mind and the
physical world.
MIT’s Fluid Interface Group: wearable
device with a projector for deep
interactions with the environment

Neuro Sky's mind-controlled headset to

play a video game.
Bridging the Cyber-Physical Divide

FitBit Community allows the

automated collection and
sharing of health-related data,
goals, and achievements

Foursquare is an online application which

integrates a persons physical location and
social network.
Community of enthusiasts that share experiences of
self-tracking and measurement.
Bridging the Cyber-Physical Divide

Tweeting Sensors
sensors are becoming social
How do we design the Sensor Web?

 Integration through shared semantics

 OGC Sensor Web Enablement
 W3C SSN ontology and Semantic Annotation

 Interpretation through integration of heterogeneous

data and reasoning with prior knowledge
 Semantic Perception/Abstraction
 Linked Open Data as prior knowledge

 Scale through distributed local interpretation

 ―intelligence at the edge‖
OGC Sensor Web Enablement
Role of OGC SWE
 Vision of Sensor Web

 Quickly discover sensors (secure or public) that can meet my

needs – location, observables, quality, ability to task
 Obtain sensor information in a standard encoding that is
understandable by me and my software
 Readily access sensor observations in a common manner, and in a
form specific to my needs
 Task sensors, when possible, to meet my specific needs
 Subscribe to and receive alerts when a sensor measures a
particular phenomenon
 Principles of Sensor Web

 Sensors will be web accessible

 Sensors and sensor data will be discoverable
 Sensors will be self-describing to humans and software (using a
standard encoding)
 Most sensor observations will be easily accessible in real time
over the web
 OGC SWE Services

 Sensor Observation Service (SOS)

 access sensor information (SensorML) and sensor observations (O&M

 Sensor Planning Service (SPS)

 task sensors or sensor systems

 Sensor Alert Service (SAS)

 asynchronous notification of sensor events (tasks, observation of
phenomena)

 Sensor Registries
 discovery of sensors and sensor data
OGC SWE Services
OGC SWE Languages

 Sensor Model Language (SensorML)

 Models and schema for describing sensor characteristics

 Observation & Measurement (O&M)

 Models and schema for encoding sensor observations
OCG SWE Observation
We want to set this data free

With freedom comes responsibility

 Discovery, access, and search
 Integration and interpretation
Semantic Sensor Web

OGC Sensor Web

Enablement

RDF OWL
 Sensor Web + Semantic Web
Semantic Web
 The web of data where web content is processed by
machines, with human actors at the end of the chain.
 The web as a huge, dynamic, evolving database of
facts, rather than pages, that can be interpreted and
presented in many ways (mashups).
 Fundamental importance of ontologies to describe the
fact that represents the data. RDF(S) emphasises
labelled links as the source of meaning: essentially a
graph model . A label (URI) uniquely identifies a
concept.
 OWL emphasises inference as the source of meaning:
a label also refers to a package of logical axioms
with a proof theory.
 Usually, the two notions of meaning fit.
 Goal to combine information and services for
targeted purpose and new knowledge
Sensor Web
 The internet of things made up of Wireless Sensor
Networks, RFID, stream gauges, orbiting satellites,
weather stations, GPS, traffic sensors, ocean buoys,
animal and fish tags, cameras, habitat monitors,
recording data from the physical world.
 Today there are 4 billion mobile sensing devices plus
even more fixed sensors. The US National Research
Council predicts that this may grow to trillions by 2020,
and they are increasingly connected by internet and
Web protocols.
 Record observations of a wide variety of modalities:
but a big part is time-series‟ of numeric measurements.
 The Open Geospatial Consortium has some web-service
standards for shared data access (Sensor Web
Enablement).
 Goal is to open up access to real-time and archival
data, and to combine in applications.
 So, what is a Semantic Sensor Web?

 Reduce the difficulty and open up sensor networks by:

 Allowing high-level specification of the data collection process;

 Across separately deployed sensor networks;
 Across heterogeneous sensor types; and
 Across heterogeneous sensor network platforms;
 Using high-level descriptions of sensor network capability; and
 Interfacing to data integration methods using similar query and
capability descriptions.

 To create a Web of Real Time Meaning!

 W3C SSN Incubator Group

 SSN-XG commenced: 1 March 2009

 Chairs:
 Amit Sheth, Kno.e.sis Center, Wright State University
 Kerry Taylor, CSIRO
 Amit Parashar  Holger Neuhaus  Laurent Lefort, CSIRO

 Participants: 39 people from 20 organizations, including:

 Universities in: US, Germany, Finland, Spain, Britain, Ireland
 Multinationals: Boeing, Ericsson
 Small companies in semantics, communications, software
 Research institutes: DERI (Ireland), Fraunhofer (Germany), ETRI (Korea),
MBARI (US), SRI International (US), MITRE (US), US Defense, CTIC
(Spain), CSIRO (Australia), CESI (China)
 W3C SSN Incubator Group

Two main objectives:

 The development of an ontology for describing

sensing resources and data, and
 The extension of the SWE languages to support
semantic annotations.
Sensor Standards Landscape
SSN Ontology

 OWL 2 DL ontology

 Authored by the XG
participants

 Edited by Michael Compton

 Driven by Use Cases

 Terminology carefully tracked

to sources through annotation
properties

 Metrics
 Classes: 117
 Properties: 148
 DL Expressivity: SIQ(D)
SSN Ontology – http://purl.oclc.org/NET/ssnx/ssn
SSN Use Cases
SSN Use Cases
SSN Ontology
 Stimulus-Sensor-Observation

 The SSO Ontology Design Pattern is developed following the principle of minimal
ontological commitments to make it reusable for a variety of application areas.
 Introduces a minimal set of classes and relations centered around the notions of stimuli,
sensor, and observations. Defines stimuli as the (only) link to the physical environment.
 Empirical science observes these stimuli using sensors to infer information about
environmental properties and construct features of interest.
SSN Ontology Modules
SSN Ontology Modules
 SSN Sensor

 A sensor can do (implements) sensing: that is, a sensor is any entity that can follow a
sensing method and thus observe some Property of a FeatureOfInterest.
 Sensors may be physical devices, computational methods, a laboratory setup with a
person following a method, or any other thing that can follow a Sensing Method to
observe a Property.
 SSN Measurement Capability

 Collects together measurement properties (accuracy, range, precision, etc) and the
environmental conditions in which those properties hold, representing a specification of a
sensor's capability in those conditions.
 SSN Observation

 An Observation is a Situation in which a Sensing method has been used to estimate or calculate a
value of a Property.
 Links to Sensing and Sensor describe what made the Observation and how; links to Property and
Feature detail what was sensed; the result is the output of a Sensor; other metadata gives the
time(s) and the quality.
 Different from OGC‘s O&M, in which an ―observation‖ is an act or event, although it also provides
the record of the event.
Alignment with DOLCE
What SSN does not model

 Sensor types and models

 Networks: communication, topology

 Representation of data and units of measurement

 Location, mobility or other dynamic behaviours

 Animate sensors

 Control and actuation

 ….
 Semantic Annotation of SWE

Recommended technique
via Xlink attributes requires
no change to SWE

 xlink:href - link to
ontology individual

 xlink:role - link to
ontology class

 xlink:arcrole - link to
ontology object
property
How do we design the Sensor Web?

 Integration through shared semantics

 OGC Sensor Web Enablement
 W3C SSN ontology and Semantic Annotation

 Interpretation through integration of heterogeneous

data and reasoning with prior knowledge
 Semantic Perception/Abstraction
 Linked Open Data as prior knowledge

 Scale through distributed local interpretation

 ―intelligence at the edge‖
Abstraction

Abstraction provides the ability to interpret and synthesize information in a way

that affords effective understanding and communication of ideas, feelings,
perceptions, etc. between machines and people.
Abstraction

 People are excellent at abstraction; of

sensing and interpreting stimuli to
understand and interact with the world.

 The process of interpreting stimuli is

called perception; and studying this
extraordinary human capability can
lead to insights for developing effective
machine perception.
Abstraction

conceptualization
of “real-world”

observe perceive

“real-world”
 Semantic Perception/Abstraction

Fundamental Questions

What is perception, and how can we

design machines to perceive?

What can we learn from cognitive

models of perception?

Is the Semantic Web up to the task of

modeling perception?
What is Perception?

Perception is the act of

 Abstracting

 Explaining

 Discriminating

 Choosing
What can we learn from Cognitive
Models of Perception?

Ulric Neisser (1976) Richard Gregory (1997)

 A-priori background knowledge is a key enabler

 Perception is a cyclical, active process
Is Semantic Web up to the task of
modeling perception?

Representation
 Heterogeneous sensors, sensing, and observation records
 Background knowledge (observable properties,
objects/events, etc.)

Inference
 Explain observations (hypothesis building)
 Focus attention by seeking additional stimuli (that
discriminate between explanations)

Difficult Issues to Overcome

 Perception is an inference to the best explanation
 Handle streaming data
 Real-time processing (or nearly)
Both people and machines are capable of observing qualities,
such as redness.

observes
Observer Quality

* Formally described in a sensor/ontology (SSN ontology)

The ability to perceive is afforded through the use of
background knowledge, relating observable qualities to entities
in the world.

Quality

* Formally described in
inheres in domain ontologies
(and knowledge bases)

Entity
With the help of sophisticated inference, both people and
machines are also capable of perceiving entities, such as apples.

perceives
Perceiver Entity

 the ability to degrade gracefully with incomplete information

 the ability to minimize explanations based on new information
 the ability to reason over data on the Web
 fast (tractable)
Perceptual Inference

Abductive Logic (e.g., PCT) Deductive Logic (e.g., OWL)

high complexity (relatively) low complexity

minimize
explanations tractable

Web reasoning
degrade gracefully

Perceptual Inference
(i.e., abstraction)
The ability to perceive efficiently is afforded through the cyclical
exchange of information between observers and perceivers.

Observer

sends sends Traditionally called the

observation focus Perceptual Cycle
(or Active Perception)

Perceiver
Neisser‘s Perceptual Cycle
Cognitive Theories of Perception

 1970’s – Perception is an active, cyclical process of

exploration and interpretation.
- Nessier’s Perception Cycle

 1980’s – The perception cycle is driven by background

knowledge in order to generate and test hypotheses.
- Richard Gregory (optical illusions)

 1990’s – In order to effectively test hypotheses, some

observations are more informative than others.
- Norwich’s Entropy Theory of Perception
Key Insights
 Background knowledge plays a crucial role in perception; what we know
(or think we know/believe) influences our perception of the world.
 Semantics will allow us to realize computational models of perception
based on background knowledge.

Contemporary Issues
 Internet/Web expands our background knowledge to a global scope;
thus our perception is global in scope
 Social networks influence our knowledge and beliefs, thus influencing our
perception
Integrated together, we have an general model – capable of
abstraction – relating observers, perceivers, and background
knowledge.

observes
Observer Quality

sends sends
observation inheres in
focus

perceives
Perceiver Entity
 Ontology of Perception – as an extension of SSN

 Provides abstraction of sensor data through perceptual

inference of semantically annotated data
 Prior Knowledge

W3C SSN Ontology Bi-partite Graph

 Prior knowledge conformant to SSN ontology (left),

structured as a bipartite graph (right)
Semantics of Explanation

Explanation is the act of accounting for sensory observations (i.e.,

abstraction); often referred to as hypothesis building.

Observed Property: A property that has been observed.

ObservedProperty ≡ ∃ssn:observedProperty—.{o1} ⊔ … ⊔
∃ssn:observedProperty—.{on}

Explanatory Feature: A feature that explains the set of observed

properties.

ExplanatoryFeature ≡ ∃ssn:isPropertyOf—.{p1} ⊓ … ⊓
∃ssn:isPropertyOf—.{pn}
Semantics of Explanation

Example
Assume the properties elevated blood pressure and
palpitations have been observed, and encoded in RDF
(conformant with SSN):

ssn:Observation(o1), ssn:observedProperty(o1, elevated blood pressure)

ssn:Observation(o2), ssn:observedProperty(o2, palpitations)

Given these observations, the following ExplanatoryFeature

class is constructed:

ExplanatoryFeature ≡ ∃ssn:isPropertyOf—.{elevated blood pressure} ⊓

∃ssn:isPropertyOf—.{palpitations}

Given the KB, executing the query ExplanatoryFeature(?y) can

infer the features, Hypertension and Hyperthyroidism, as
explanations:

ExplanatoryFeature(Hypertension)
ExplanatoryFeature(Hyperthyroidism)
 Semantics of Discrimination

Discrimination is the act of deciding how to narrow down the multitude of

explanatory features through further observation.

Expected Property: A property is expected with respect to (w.r.t.) a set of

features if it is a property of every feature in the set.
ExpectedProperty ≡ ∃ssn:isPropertyOf.{f1} ⊓ … ⊓ ∃ssn:isPropertyOf.{fn}

NotApplicable Property: A property is not-applicable w.r.t. a set of features if it

is not a property of any feature in the set.
NotApplicableProperty ≡ ¬∃ssn:isPropertyOf.{f1} ⊓ … ⊓
¬∃ssn:isPropertyOf.{fn}

Discriminating Property: A property is discriminating w.r.t. a set of features if it

is neither expected nor not-applicable.
DiscriminatingProperty ≡ ¬ExpectedProperty ⊓ ¬NotApplicableProperty
Semantics of Discrimination

Example
Given the explanatory features from the previous example,
Hypertension and Hyperthyroidism, the following classes are
constructed:

ExpectedProperty ≡ ∃ssn:isPropertyOf.{Hypertension} ⊓
∃ssn:isPropertyOf.{Hyperthyroidism}

NotApplicableProperty ≡ ¬∃ssn:isPropertyOf.{Hypertension} ⊓
¬∃ssn:isPropertyOf.{Hyperthyroidism}

Given the KB, executing the query DiscriminatingProperty(?x)

can infer the property clammy skin as discriminating:

DiscriminatingProperty(clammy skin)
How do we design the Sensor Web?

 Integration through shared semantics

 OGC Sensor Web Enablement
 W3C SSN ontology and Semantic Annotation

 Interpretation through integration of heterogeneous

data and reasoning with prior knowledge
 Semantic Perception/Abstraction
 Linked Open Data as prior knowledge

 Scale through distributed local interpretation

 ―intelligence at the edge‖
Efficient Algorithms for IntellegO

 Use of OWL-DL reasoner too resource-intensive for use in resource

constrained devices (such as sensor nodes, mobile phones, IoT devices)
 Runs out of resources for problem size (prior knowledge) > 20 concepts
 Asymptotic complexity: O(n3) [Experimentally determined]

 To enable their use on resource-constrained devices, we now describe

algorithms for efficient inference of explanation and discrimination.

 These algorithms use bit vector encodings and operations, leveraging a-

priori knowledge of the environment.
Efficient Algorithms for IntellegO

Semantic (RDF) Encoding Bit Vector Encoding

Lower

Lift

 First, developed lifting and lowering

algorithms to translate between RDF
and bit vector encodings of
observations.
Efficient Algorithms for IntellegO

Explanation Algorithm Utilize bit vector operators to efficiently

compute explanation and discrimination

 Explanation: Use of the bit vector AND

operation to discover and dismiss those features
that cannot explain the set of observed
properties
Discrimination Algorithm
 Discrimination: Use of the bit vector AND
operation to discover and indirectly assemble
those properties that discriminate between a set
of explanatory features. The discriminating
properties are those that are determined to be
neither expected nor not-applicable
Efficient Algorithms for IntellegO

Evaluation: The bit vector encodings and algorithms yield significant and necessary
computational enhancements – including asymptotic order of magnitude improvement, with
running times reduced from minutes to milliseconds, and problem size increased from 10‘s
to 1000‘s.
Adoption of SSN
SSN Applications
Linked Sensor Data

Linked Sensor Data

(~2 Billion Statements)
Sensor Discovery Application

Query w/ location name to find nearby sensors

SSN Applications

Applications of SSN
Weather Rescue Healthcare
 SSN Application: Weather

50% savings in sensing resource

requirements during the detection of a
blizzard

 Order of magnitude resource

savings between storing observations vs.
relevant abstractions
SSN Application: Fire Detection

Weather Application
SECURE: Semantics-empowered Rescue Environment
(detect different types of fires)

DEMO: http://www.youtube.com/watch?v=in2KMkD_uqg
SSN Application: Health Care

MOBILEMD: Mobile app to help reduce re-admission

of patients with Chronic Heart Failure
 SSN Application: Health Care
Passive Monitoring Phase

Observed Symptoms Possible Explanations

• Abnormal heart rate • Panic Disorder

• Clammy skin • Hypoglycemia
• Hyperthyroidism
• Heart Attack
• Septic Shock

Passive Sensors – heart rate, galvanic skin response

 SSN Application: Health Care
Active Monitoring Phase
Are you feeling lightheaded?

yes

Are you have trouble taking deep breaths? Observed Symptoms Possible Explanations

• Abnormal heart rate • Panic Disorder

yes •
• Clammy skin Hypoglycemia
• Lightheaded • Hyperthyroidism
Do you have low blood pressure? • Trouble breathing • Heart Attack
• Low blood pressure • Septic Shock

yes

Have you taken your Methimazole

medication?

no

Active Sensors – blood pressure, weight scale, pulse oxymeter

Future work

 Creating ontologies and defining data models are not enough

 tools to create and annotate data
 Tools for publishing linked IoT data
 Designing lightweight versions for constrained environments
 think of practical issues
 make it as much as possible compatible and/or link it to the other
existing ontologies
 Linking to domain knowledge and other resources
 Location, unit of measurement, type, theme, …
 Linked-data
 URIs and naming
Some of the open issues

 Efficient real-time IoT resource/service

query/discovery
 Directory
 Indexing

 Abstraction of IoT data

 Pattern extraction
 Perception creation

 IoT service composition and compensation

 Integration with existing Web services
 Service adaptation
Selected references
 Payam Barnaghi, Wei Wang, Cory Henson, Kerry Taylor, "Semantics for the Internet of Things: early progress and back to the future", (to appear)
International Journal on Semantic Web and Information Systems (special issue on sensor networks, Internet of Things and smart devices), 2012.
 Atzori, L., Iera, A. & Morabito, G. , ―The Internet of Things: A survey‖, Computer Networks, Volume 54, Issue 15, 28 October 2010, 2787-2805.
 Suparna De, Tarek Elsaleh, Payam Barnaghi , Stefan Meissner, "An Internet of Things Platform for Real-World and Digital Objects", Journal of
Scalable Computing: Practice and Experience, vol 13, no.1, 2012.
 Suparna De, Payam Barnaghi, Martin Bauer, Stefan Meissner, "Service modelling for the Internet of Things", in Proceedings of the Conference on
Computer Science and Information Systems (FedCSIS), pp.949-955, Sept. 2011.
 Cory Henson, Krishnaprasad Thirunarayan, and Amit Sheth, ―An Efficient Bit Vector Approach to Semantics-based Machine Perception in Resource-
Constrained Devices,‗‖ In: Proceedings of 11th International Semantic Web Conference (ISWC 2012), Boston, Massachusetts, USA, November 11-25,
2012.
 Cory Henson, Amit Sheth, and Krishnaprasad Thirunarayan, ―Semantic Perception: Converting Sensory Observations to Abstractions‖, IEEE Internet
Computing, Special Issue on Context-Aware Computing, March/April 2012.
 Cory Henson, Krishnaprasad Thirunarayan, Amit Sheth, ―An Ontological Approach to Focusing Attention and Enhancing Machine Perception on the
Web.,‖ Applied Ontology, vol. 6(4), pp.345-376, 2011.
 Payam Barnaghi, Frieder Ganz, Cory Henson, Amit Sheth, ―Computing Perception from Sensor Data‖, In proceedings of the 2012 IEEE Sensors
Conference, Taipei, Taiwan, October 28-31, 2012.
 Michael Compton et al, ―The SSN Ontology of the W3C Semantic Sensor Network Incubator Group‖, Journal of Web Semantics, 2012.
 Harshal Patni, Cory Henson, and Amit Sheth , “Linked Sensor Data”, in Proceedings of 2010 International Symposium on Collaborative Technologies
and Systems (CTS 2010), Chicago, IL, May 17-21, 2010.
 Amit Sheth, Cory Henson, and Satya Sahoo , ―Semantic Sensor Web IEEE Internet Computing‖, vol. 12, no. 4, July/August 2008, pp. 78-83.
 Wei Wang, Payam Barnaghi, Gilbert Cassar, Frieder Ganz, Pirabakaran Navaratnam, "Semantic Sensor Service Networks", (to appear) in
Proceedings of the IEEE Sensors 2012 Conference, Taipei, Taiwan, October 2012.
 Wang W, De S, Toenjes R, Reetz E, Moessner K, "A Comprehensive Ontology for Knowledge Representation in the Internet of Things", International
Workshop on Knowledge Acquisition and Management in the Internet of Things (KAMIoT 2012) in conjunction with IEE IUCC-2012, Liverpool, UK.
Liverpool. 25-27 June, 2012.
Some useful links related to IoT
 Internet of Things, ITU
 http://www.itu.int/osg/spu/publications/internetofthings/InternetofThings_summary.pdf
 IoT Comic Book
 http://www.theinternetofthings.eu/content/mirko-presser-iot-comic-book
 Internet of Things Europe, http://www.internet-of-things.eu/
 Internet of Things Architecture (IOT-A)
 http://www.iot-a.eu/public/public-documents
 W3C Semantic Sensor Networks
 http://www.w3.org/2005/Incubator/ssn/XGR-ssn-20110628/
 Kno.e.sis Semantic Sensor Web Group
 http://knoesis.org/projects/ssw