Beruflich Dokumente
Kultur Dokumente
3G Overview
1
Company Confidential
What’s New in WCDMA?
Characteristic to WCDMA
• RAKE receiver takes advantage of multipath propagation
• Fast power control keeps system stable by using minimum power necessary for
links
• Soft handover ensures smooth handovers
GMSC GGSN
AuC
Gc
C H
HLR Gn
PSTN PSTN
Gr
D EIR
F Gf
G
VLR VLR
Gs
B B SGSN
MSC MSC
E
CN
A Gb IuCS IuPS
BSS RNS
Iur
BSC RNC RNC
Abis Iubis
Um Uu
ME
Document
SIM USIM
Ref. 3GPP TS23.002 MS 3
Company Confidential
3G Spectrum Allocation
4
Company Confidential
IMT2000 Frequency Allocation for
UMTS
1900 1920 1980 2010 2025 2110 2170 2200
FDL
FDL/UL
FUL
7
Company Confidential
UMTS System Characteristics
• W-CDMA : 5 MHz
• Carrier Spacing : multiples of 200 kHz
• W-CDMA spreading rate = 3.84 Mchip/s
• Chip Rate = 3.84 MHz
• Raised cosine filtering with roll-off 0.22
• Information bit rate: between 8 kbit/s and 2 Mbit/s (currently up to 384 Kbit/s)
• Spreading Factor (SF): 4 -256
• Multiple Access Scheme : Wideband DS-CDMA
• Duplex Scheme : FDD and TDD modes
• Carrier Spacing : 4.4 – 5.4 MHz
• 10 ms frame with 15 time slots
• NodeB synchronization: asynchronous
• Highly variable data rates, data rate constant within 10 ms frame
• Bandwidth on demand, efficient resource usage
• Multiple services with different variable data rates over one physical channel
• DL Transmission diversity
8
Company Confidential
Key features of WCDMA
•Soft handoff: user equipment (UE) and base stations use special rake receivers that
allow each UE to simultaneously communicate with multiple base stations. The diversity
gain associated with soft handoff is known as the "soft handoff gain factor".
– Soft handover (links from different NodeB) is performed at RNC level using Selection
Combination (RNC choose the best link)
– Softer handover (links from same NodeB) is performed at NodeB using Maximum Ratio
Combination (NodeB combines the signals). Softer handover usually is higher then soft
•Multi-path reception: the rake receivers also allow the UE to decode multiple signals
that have traveled over different physical paths from the base station. For example, one
signal may travel directly from the base station to the UE, and another may reflect off a
large building and then travel to the UE. This phenomenon, "multi-path propagation", also
provides a diversity gain. The same effect occurs on the uplink from the UE to the base
station.
9
Company Confidential
Key features of WCDMA
Frequency reuse of 1: every base station in the CDMA system operates on the same
frequency for a given carrier, so no frequency planning is required. As every site causes
interference to every other site, careful attention must be paid to each site's radio
propagation.
Soft capacity: capacity and coverage are intertwined in CDMA, depending on the
number of users in the system and the amount of interference allowed before access is
blocked for new users. By setting the allowed interference threshold lower, coverage will
improve at the expense of capacity. By setting the threshold higher, capacity will increase
at the expense of coverage. Because of the fundamental link between coverage and
capacity, cells with light traffic loads inherently share some of their latent capacity with
more highly loaded surrounding cells.
10
Company Confidential
WCDMA Compared to GSM and CDMA IS-95
WCDMA vs. GSM
WCDMA has some similarities with GSM technology, however, it is a fundamentally
different technique for allowing multiple users to share the same spectrum and as a
result it has many differences.
11
Company Confidential
WCDMA Compared to GSM and IS-95 CDMA
12
Company Confidential
TD-SCDMA System Characteristics
Time Division Synchronous CDMA (TD-SCDMA) was proposed by China Wireless Telecommunication Standards
group (CWTS) and approved by the ITU in 1999 and technology is being developed by the Chinese Academy of
Telecommunications Technology and Siemens. TD-SCDMA uses the Time Division Duplex (TDD) mode, which
transmits uplink traffic (traffic from the mobile terminal to the base station) and downlink traffic (traffic from the
base station to the terminal) in the same frame in different time slots. That means that the uplink and downlink
spectrum is assigned flexibly, dependent on the type of information being transmitted. When asymmetrical data
like e-mail and internet are transmitted from the base station, more time slots are used for downlink than for
uplink. A symmetrical split in the uplink and downlink takes place with symmetrical services like telephony.
Real time
Non
Real time
14
Company Confidential
RLC Transfer Mode
• AC determines the UL & DL RLC transfer mode for RNC & UE
• RLC transfer mode indicates, whether the RLC entity of a RB shall
use an Acknowledged (AM) (delivery guaranteed), an
Unacknowledged (UM) (delivery not guaranteed ) or a Transparent
(Tr) Mode (no protocol overhead added to higher layers ) data
transfer.
• Equal transfer mode is used in UL and DL for a RB.
• The transfer mode is determined from the RAB parameters and CN
domain information.
CS domain PS domain
TM
TM TM
TM UM
UM AM
AM AM
AM AM
AM
RLC
RLC RLC
RLC RLC
RLC RLC
RLC RLC
RLC RLC
RLC
DCH
DCH DCH
DCH DCH
DCH DCH
DCH RACH/FACH
RACH/FACHDCH/HS-DSCH
DCH/HS-DSCH DCH/DCH
DCH/DCH
15
RT traffic NRT traffic
Company Confidential
Conversational services
• Speech service:
– Real time conversational service require the low time delay from
end to end , and the uplink and the downlink service bandwidth
is symmetrical
– Adopt AMR ( adaptive multi rate ) technique (WCDMA).
• 12.2 (GSM), 10.2, 7.95, 7.40(IS-41), 6.70(PDC), 5.90, 5.15 and
4.75kbps
• The bit rate of AMR voice can be controlled by the RAN according
to the payload of air interface and the quality of voice service
• According to the requirement of the operator ,AMR technique can
balance the relationship among the network capacity, coverage and
the service quality
16
Company Confidential
Streaming services
• Multimedia data streaming:
– Preserve time relation between information entities of the data
streaming
– Data is processed into stable and continuous streaming
– Non-symmetry service
• Services example :
– Telemetry ( monitoring )
17
Company Confidential
Interactive services
• Services example :
– Location based services
– Online game
– Web browsing
18
Company Confidential
Background services
• Service example:
– E-mail ( server to server )
– SMS
– Download of database
– Reception of measurement records
19
Company Confidential
Transmission diversity - STTD
• Space Time Transmit Diversity
20
Company Confidential
Transmission diversity - TSTD
Time Switch Transmit Diversity, used in Synchronization physical channels
P-SCH and S-SCH
P-SCH
S-SCH
P-SCH
S-SCH
21
Company Confidential
Part II
WCDMA Fundamentals
22
Company Confidential
WCDMA = DS-CDMA
•WCDMA is a code-division multiple access technology which separates each
user’s voice or data information by multiplying the information by pseudo-random
bits called "chips".
•Direct sequence (DS): with DS, a binary modulated signal is ‘directly’ multiplied
by a code. The code is a pseudo-random sequence of ±1, where the bit rate of
the code (Chip Rate) is higher than the rate of the signal, usually considerably
higher. This has the effect of spreading the signal to a wideband. At the receiver,
the same code is used to extract the original signal from the incoming wideband
signal. A bit of the code is referred to as a chip, and the defining parameter for
such a system is the chip rate.
•The pseudo-random bit sequences have a rate of 3.84 Mcps (millions of chips
per second), resulting in the narrowband information bits of the user being spread
across a much wider bandwidth of approximately 5 MHz.
• For this reason, CDMA technology is sometimes referred to as “spread
spectrum.”
•Spread spectrum describes any system in which a signal is modulated so that its
energy is spread across a frequency range that is greater than that of the original
signal
•The user data (signal) is first spread by the channelisation code (based on
Hadamard matrix) called Orthogonal Variable Spreading Factor (OVSF) Code.
•OVSF code has the property that two different codes from the family are
perfectly orthogonal if in phase
23
Company Confidential
TDMA based System
24
Company Confidential
W-CDMA based System
25
Company Confidential
Processing Gain and Spreading
According to information theory, as the frequency spectrum a signal occupies is expanded, the overall power level
decreases. In CDMA, the user signals are spread up to a wideband by multiplication by a code.
Consider a narrowband signal, say, for example, a voice call.
When viewed in the frequency spectrum, it occupies some frequency and has some power level. Once the frequency
is spread across a wideband, the total power of this signal is substantially reduced.
Now consider that another user has the same procedure performed on it and is also spread to the same wideband.
The total system power is increased by a small amount as the two users are transmitted at the same time. Therefore,
each new user entering the system will cause the power of the wideband to increase. 26
Company Confidential
Spreading and Despreading
• At the receiver, the process of extracting one user is performed (DESPREADING)
• The regenerated signal needs to be retrieved with enough power that it can be
perceived above the level of the remaining spread signals. That is, it needs to be
of a sufficient strength, or margin, above the rest of the signals so that the signal
can be accurately interpreted.
• Considering this as a signal to interference ratio (SIR), or carrier to interference
(C/I) ratio, the noise affecting one signal is the remaining spread signals that are
transmitting at that frequency. This SIR is classified in CDMA as Eb/No.
• For mobile device measurements of the quality of the signals from the network, it
uses a pilot channel (CPICH), which is broadcast by each cell. The mobile device
measures Ec/Io, the energy level of this pilot channel, Ec, compared to the total
energy received, Io
• Another important characteristic is the rejection of unwanted narrowband noise
signals. If a wideband signal is affected by a narrowband noise signal, then since
the spreading function is commutative, the despreading operation while extracting
the wanted signal will in turn spread the narrowband noise to the wideband, and
reduce its power level
•The lower the power that the original signals are transmitted with, the lower the
noise in the system. It is therefore essential that each user in the system transmits
with an optimum power level to reach the receiver with its required power level. If
the power level is too high, then that user will generate noise, which in turn affects
the performance of all the other users. If there is too little power, then the signal
which reaches the receiver is of too low quality, and it cannot be accurately ‘heard’
27
Company Confidential
Spreading and Despreading
• There are two solutions to the problem of noise levels:
– First, an Admission Control policy is required that monitors the number of users
and the noise level, and once it reaches some maximum tolerable level, refuses
admission of further users. In a cellular system, such admission control needs to
be considered not only for one cell, but also for the effects that noise levels within
that cell have on neighboring cells.
– The second solution is to implement Power Control. Each user needs to transmit
with just enough power to provide a clear signal at the receiver above the noise
floor. This should be maintained regardless of where the users are located with
respect to the receiver, and how fast they are moving. Power control needs to be
performed frequently to ensure that each user is transmitting at an optimum
level.
• The ratio of the original signal to the spread signal is referred to as the
spreading factor and is defined as:
Spreading factor (SF) = chip rate/symbol rate
28
Company Confidential
Spreading and Despreading
• In the next slides, the SF is 4. Hence, variable data rates can be supported by using
variable length codes and variable SF to spread the data to a common chip rate
• When considering CDMA systems, it is useful to define how the different signals
interact with each other. Correlation is defined as the relationship or similarity
between signals. For pulse-type waveforms, such as CDMA codes, the cross-
correlation between two signals is defined as:
where R12 is the correlation between two signals v1 and v2, and τ is their relative
time offset.
• For the code to be effective, the receiver must know the specific code (in this case
1,-1,1,-1, see next slides) which is being used for transmission and it must also be
synchronized with this transmission. On reception the receiver can then simply
reintroduce the correct code which is multiplied with the incoming signal and
reproduce the actual symbol sent by the transmitter.
• The receiver also needs to know the actual number of chips that represent a
symbol (spreading factor) so that the chips can be regenerated to the sent symbol
through averaging the value of the chips over the symbol time. This is achieved
through integration, where the chips are summed over the total time period of the
symbol they represent.
• The principle of correlation is used at the receiver to retrieve the original signal out
of the noise generated by all the other users’ wideband signal.
29
Company Confidential
Spreading and Despreading
• At the receiver, the received signal is multiplied by the code and the result is
integrated over the period of each baseband bit to extract the original data.
Since the receiver has four chips over which to integrate, the procedure yields
a strong result at the output.
• However, consider now that the receiver does not know the correct code.
Then the integration process will result in a signal which averages to around
zero.
• For both of these, the relative strength of the desired signal and the rejection
of other signals is proportionate to the number of chips over which the
receiver has to integrate, which is the SF. Large SFs result in more
processing gain and hence the original signals do not need so much
transmission power to achieve a target quality level.
• As can be seen, the longer the symbol time (i.e. lower data rate and higher
chip rate), the longer the integration process, thus the higher the amplitude of
the summed signal. This is referred to as processing gain (Gp) and is directly
proportional to the SF used.
• For example, if the symbols were spread over 8 chips then the Gp will be 8; if
spread over 16 chips, Gp would be 16. This means that the processing gain is
higher for lower data rates than for higher data rates, i.e. lower data rates can
be sent with reduced power since it is easier to detect them at the receiver 30
Company Confidential
Spreading and Despreading
SF= chip rate/symbol rate = 4
The spreading sequences must have good correlation properties to facilitate the
separation of the wanted signal from all others:
•One sharp and dominant peak of the autocorrelation function for zero phase shift
•As small as possible values of the autocorrelation function for all out-of-phase shift
•As small as possible values of the cross-correlation (different signals) function for
31
all phase shift Company Confidential
Spreading and Despreading
32
Company Confidential
CDMA Multiple Access Advantages :
Multiple Access Features
1. All Users’ Signals overlap in TIME and FREQUENCY
2. Correlating the Received Signal despreads ONLY the WANTED SIGNAL
p p
S1
RECEIVER of USER 1
S1xC1
f f p
p S1 = S1 X C1 X C1
p p S2 X C2 X C1
S2
f
f
S2xC2
f f
33
Company Confidential
CDMA Multiple Access Advantages :
Interference Rejection
p p
S1
S1xC1
f f
p
S1
p p
I I IxC1
f f
f
1/Tb
Tb 2Tb 3Tb
c1(t)
C1(f)
f
1/Tb 1/Tc
Tc 4Tc
m1(t).c1(t)
C1(f)* M1(f)
f
1/Tb 1/Tc
35
Company Confidential
Processing gain (Gp)
• Gp = Wc/Wi
• Where
– Wc: chip rate
– Wi: user data rate
Wc f
Wi
• The more processing gain the system has, the more the power of uncorrelated
interfering signals is suppressed in the despreading process
• Thus, processing gain can be seen as an improvement factor in the SIR
(Signal to Interference Ratio) of the signal after despreading
• Example: Voice AMR 12.2 Kbps Gp = 10*log(3840000/12200)= 25 dB
• After despreading the signal power has to be typically few dB above the
interference and noise: Eb/No = 5dB; therefore the required wideband signal-
to-interference ratio is 5dB – Gp = -20 dB.
• In other words, the signal power can be 20 dB under the interference and the
WCDMA receiver can still detect the signal
• Wideband signal-to-interference ratio is also called carrier-to-interference ratio:
C/I
• Thanks to spreading and desporeading, C/I can be much lower in WCDMA
than GSM (C/I = 9-12 dB)
36
Company Confidential
Processing gain (Gp)
Processing Gain
G=W/R=25 dB
W
(W/Hz)
Spread wideband
signal
"narrowband"
W signal
(W/Hz)
Spread wideband
signal
Frequency (Hz)
37
Company Confidential
Spreading in WCDMA
Consists of 2 operations:
1. Channelisation (OVSF: Orthogonal Variable SF)
• Transforms each symbol (data bit) to the number of chips (increases
bandwidth)
• Number of chips per symbol = Spreading Factor (SF)
2. Scrambling (does not affect the signal bandwidth)
• Scrambling code is applied (PN codes)
TX
Scrambling Code
38
Company Confidential
Spreading in WCDMA
• Channel coding
– CRC attachment
• Check for error during transmission
• Voice: if CRC check returns error discard information
• Data: if CRC check returns error ask for retransmission
– Convolutional or Turbo Coding
• Convolution coding for voice and low speed signalling
• Turbo coding for large data transmission, better performances than
convolutional coding
– Interleaving
• Distribute error (burstly error) over data transmitted
– Rate Matching
• Match Symbol Rate to that accepted by spreading
• Rate matching techniques: Repeat or Puncturing (remove)
Example: voice 12.2 (SF 128)
Rs=3.84/128= 30Ksps
Rs 3.84 Mcps If the output of Channel cod is
Rb Channel
coding < 30Ksps, Rate matching
Symbol Rate Chip Rate repeats the info else uses
puncturing (remove) 39
(OVSF) Company Confidential
Spreading in WCDMA
42
Company Confidential
OVSF properties
• To verify if two codes have a zero cross-correlation, they are tested in the below equation, first
multiplied together and then integrated, as shown in Figure below. The result is zero, indicating that
indeed they are orthogonal.
• The number of chips which represent a symbol is known as the SF or the processing gain. To support
different data rates within the system, codes are taken from an appropriate point in the tree. These
types of orthogonal codes are known as orthogonal variable spreading factors (OVSF).
• In the 3G WCDMA system the chip rate is constant at 3.84 Mchips/s. However, the number of chips
that represent a symbol can vary. Within this system as laid down by the specifications, the minimum
number of chips per symbol is 4 which would give a data rate of 3 840 000/4 = 960 000 symbols per
second. The maximum SF or number of chips per symbol is 256,1 which would give a data rate of 3
840 000/256 = 15 000 symbols per second. Thus it can be seen that the fewer chips used to represent
a symbol, the higher the user data rate. The actual user data rate must be rate matched to align with
one of these SF symbol rates.
• Although orthogonal codes demonstrate perfect signal separation, they MUST be perfectly
synchronized to achieve this. Another drawback of orthogonal codes is that they do not evenly
spread signals across the wide frequency band, but rather concentrate the signal at certain
discrete frequencies. As an example, consider that the code ‘1 1 1 1’ will have no spreading
effect on a symbol. To overcome these drawbacks the PN codes are introduced
43
Company Confidential
Scrambling code properties
• Another code type used in CDMA systems is the pseudo-random noise (PN)
sequence. This is a binary sequence of ±1 that exhibits characteristics of a
purely random sequence, but is deterministic. Like a random sequence, a PN
sequence has an equal number of +1s and −1s, with only ever a difference of
1. PN sequences are extremely useful as they fulfill two key roles in data
transmission:
1. Even spreading of data: when multiplied by a PN sequence, the resultant signal is
spread evenly across the wideband. To other users who do not know the code, this
appears as white noise.
2. Signal separation: while PN sequences do not display perfect orthogonality
properties, nevertheless they can be used to separate signals. At the receiver, the
desired signal will show strong correlation, with the other user signals exhibiting
weak correlation
• Another property of PN sequences is that they exhibit what is known as
autocorrelation. This is defined as the level of correlation between a signal and
a time-shifted version of the same signal, measured for a given time shift. For
a PN sequence, the autocorrelation is at a maximum value, N, when perfectly
time aligned, i.e. τ = 0. N is the length in numbers of bits of the PN sequence.
This single peak drops off quickly at ±Tc, where Tc is the width of a chip of the
code. This allows a receiver to focus in on where the signal is, without a
requirement for the transmitter and receiver to be synchronized. In
comparison, the autocorrelation of time-shifted orthogonal codes results in
several peaks, which means that this signal locking is much more problematic.44
Company Confidential
Scrambling code properties
• The OVSF codes are effective only when the channels are perfectly
synchronized at symbol level
• The loss in cross-correlation, e.g. due to multipaths, is compensated
by the additional scrambling operation
• Scrambling codes are used to separate different cells in the downlink
and different terminals in the uplink
• They have good correlation properties (interference averaging) and
are always used on top of the spreading codes, thus not affecting the
transmission bandwidth
• Gold sequence is used to generate scrambling codes
• For downlink physical channels, a total of 218 = 262,143 scrambling
codes can be generated
• Only scrambling codes k = 0, 1, ..., 8191 are used
• 8192 scrambling codes are divided into 512 groups each of which
contains 16 scrambling codes
• The first scrambling code of each group is called Primary Scrambling
Code (PSC) and the other 15 are Secondary Scrambling Codes
(SSC)
45
Company Confidential
Usage of the codes
Channelization Code Scrambling Code
Usage Uplink: separation of physical data Uplink: Separation of terminals
(DPDCH) and control channels (DPCCH)
for the same terminal
Downlink: separation of downlink Downlink: Separation of sectors
connections to different users within on cell (cells)
Length 4-256 chips Uplink: 10ms = 38400 chips
In downlink also 512 chips
Downlink: 10ms = 38400 chips
Number of codes Spreading Factor indicates the number of Uplink: over 16 millions
codes under one scrambling code Downlink: 512
Code Family Orthogonal Variable Spreading Factor 10ms code: Gold Code
(OVSF) 66.7µs code: Extended code
family
Spreading Yes, indicates bandwidth No, does not affect bandwidth
46
Company Confidential
Receivers
• Both NodeB and Terminals use the same type of correlation
receivers
• Due to multipath propagation it’s necessary to use multiple
correlation receivers (fingers) in order to recover (combine) the
energy from all paths coherently and obtain multipath diversity
47
Company Confidential
Multipath propagation and diversity
• A transmission from a mobile device is more or less omni directional, and this is also the
case for base stations when they have only one cell. Base stations which are sectorized will
have directional antennas, which will transmit only over a certain range.
• For example, a three-sectored site will have three antennas which each transmit over the
range of 120 degrees. From the point of view of the mobile device, it would be ideal if a
transmission were unidirectional; however, this is impractical since it would mean that the
antenna of the mobile device would need to point towards the base station at all times.
• In this ideal situation the device could transmit with reduced power, thus causing less
interference to other users and increasing the device’s battery life. In the cellular
environment, much of the power transmitted is actually in the wrong direction. In urban
areas there is considerable reflection of the signal from surrounding buildings. This is
actually a reason why cellular systems work, since the mobile device can thus be out of
direct line of sight of the BTS and its signal will still be received.
• The reflected signals travel further distances than the direct line of sight transmission and
therefore arrive slightly later,
later with greater attenuation and possible phase difference.
• It would be advantageous if these time-shifted versions in the multipath signal could be
combined at the receiver with the effect that a much stronger signal is received.
• Because this combined signal is stronger, it is possible that the BTS may tell the mobile
device to reduce its transmitting power. Any process of combining multiple versions of the
same signal to provide a more powerful, better quality signal is known as diversity.
• The autocorrelation property of the PN sequence is again used. Since the received signal
resolves into a single peak around the chip width, then as long as the multipath profile is of
a duration longer than the chip width, a number of peaks will be observed, each one
representing a particular multipath
48
Company Confidential
Wide Band Channel
• Definition:
• A channel is defined wide when its bandwidth (Bw) is greater than the
Coherence Bandwidth: Bw >> ∆fc
1
∆f c =
2πSτ
49
Company Confidential
Wide Band Channel – Delay Spread
Channel impulse response (power delay profile) and delay spread
Dominant Path
τ1
50
Company Confidential
Wide Band Channel – Narrow/Wide Band System
Microsoft Word
Document
51
Company Confidential
WCDMA and GSM in TU3 Channel
52
Company Confidential
Optimal Receiver for WCDMA signal
• For a channel with only one signal path optimal receiver is one correlator
(code de-spreading and integration
Basic unit of
Rake Receiver
53
Company Confidential
Optimal Receiver for WCDMA signal
• In a multipath environment optimal receiver utilizes several correlators (Rake
Fingers) tuned for dominant delays = Rake receiver Adobe Acrobat
Document
54
Company Confidential
Rake Receiver
• Rake finger delays tuned based on channel impulse
response estimation
• Code Matched Filter, Search Finger
• Fingers combined with Maximal Ratio combining
• Performance of Rake Receiver depends on the channel
powers delay profile
• Max path delay difference vs. chip time amount of
multipath diversity
55
Company Confidential
Rake Receiver - Combining
• Combined signal without and with phase estimation and correction (example
6 path channel)
56
Company Confidential
Maximal Ratio Combining of Symbols
Finger n.2
Finger n.3
UE
WBTS
57
Company Confidential
Maximal Ratio Combining of Symbols
Transmitted Received Modified with channel Combined
symbol symbol+noise estimate and relative symbol +
delay compensation residual
(for combining) noise
Finger n.1
Finger n.2
Finger n.3
UE
WBTS
58
Company Confidential
WCDMA in TU Channel
time
Corr 1 Corr 3
Corr 2 Corr 4 Corr 5 Corr 6 Corr 7
59
Company Confidential
WCDMA in Indoor Channel
• No multipath diversity
60
Company Confidential
Part III
Scrambling Code Planning
61
Company Confidential
Scrambling Code Planning
62
Company Confidential
Scrambling Code Planning
63
Company Confidential
Scrambling Code Planning
64
Company Confidential
Scrambling Code Planning
65
Company Confidential
Scrambling Code Planning
66
Company Confidential
Scrambling Code Planning
67
Company Confidential
Scrambling Code Planning
68
Company Confidential
Scrambling Code Planning
69
Company Confidential
Scrambling Code Planning
70
Company Confidential
Scrambling Code Planning
71
Company Confidential
Scrambling Code Planning
72
Company Confidential
Part IV
Physical Layer
73
Company Confidential
Channel Mapping
In GSM, we distinguish between logical and physical channels. In UMTS there are three different
types of channels:
• Logical Channels
Logical Channels were created to transmit a specific content. There are for instance logical channel to
transmit the cell system information, paging information, or user data. Logical channels are offered as data
transfer service by the Medium Access Control (MAC) layer to the next higher layer. Consequently, logical
channels are in use between the mobile phone and the RNC.
• Transport Channels (TrCH)
The MAC layer is using the transport service of the lower, the Physical layer. The MAC layer is responsible to
organise the logical channel data on transport channels. This process is called mapping. In this context, the
MAC layer is also responsible to determine the used transport format. The transport of logical channel data
takes place between the UE and the RNC.
• Physical Channels (PhyCH)
The physical layer offers the transport of data to the higher layer. The characteristics of the physical
transport have to be described. When we transmit information between the RNC and the UE, the physical
medium is changing. Between the RNC and the Node B, where we talk about the interface Iub, the
transport of information is physically organised in so-called Frames.
Between the Node B and the UE, where we find the WCDMA radio interface Uu, the physical transmission
is described by physical channels. A physical channel is defined by the UARFCN and the a spreading
code in the FDD mode.
74
Company Confidential
Radio Interface Channel Organisation
L2
Transport Channels
define how and with which type of characteristics the MAC
data is transferred by the physical layer Layer
75
Company Confidential
Radio Interface Protocol Architecture GC Nt DC
Duplication avoidance
control
control
control
On the following figures. you can see the mapping of logical channels onto transport channels,
as well as the mapping of transport channels onto physical channels.
79
Company Confidential
Physical Channels (PhyCH)
Physical Channels are characterised by
•UARFCN,
•scrambling code,
•channelisation code (optional),
•start and stop time, and
•relative phase (in the uplink only, with relative phase being 0 or π/2)
Transport channels can be mapped to physical channels. But there exist physical channels, which
are generated at the Node B only, as can be seen on the next figures.
In addition to the physical channels mapped from the transport channels, there exist
physical channels for signaling purposes (blue color) to carry only information
between network and the terminals.
80
Company Confidential
Channel Mapping DL (Network Point of View)
Logical Transport Physical
Channels Channels Channels
P-CPICH
81
Company Confidential
Channel Mapping UL (Network Point of View)
Logical Transport Physical
Channels Channels Channels
DCCH
HS-DPCCH
DTCH DCH DPDCH I branch
DPCCH Q branch
82
Company Confidential
Example – Channel configuration during call
Data Logical Transport Physical
Channels Channels Channels
RRC
DCCH0-4 DCH1
signalling
DPDCH
Speech
DTCH1 DCH2-4 DPCCH
data
AMR speech
NRT
data
DTCH2 DCH5 +
NRT data
AMR speech connection utilises multiple transport channels
RRC connection utilises multiple logical channels, signalling radio bearers
83
Company Confidential
Transport Channels Physical Channels
84
Company Confidential
Transport Formats
TFCS
TB
DCH 2
TB TB TB
TB
TB
TBS DCH 1
TB TB
TB TB TB
TFS
TTI TTI TTI
TFC
TF
Cell synchronisation is achieved with the Synchronisation Channel (SCH). This channel divides up
into two sub-channels, P-SCH and S-SCH are not under the cell-specific primary scrambling
code (the UE must be able to synchronize to the cell before knowing the downlimk scrambling
code)
Matched 86
(continued on the next text slide) Filter
Company Confidential
Synchronisation Channel (SCH)
CP CPP CP CP
10 ms Frame
Cp = Primary Synchronisation Code (Activity Factor 10%)
Cs = Secondary Synchronisation Code (Activity Factor 10%)
87
Company Confidential
Cell Synchronisation
• Secondary Synchronisation Channel (S-SCH)
• (FRAME SYNCH and Scrambling Code Group DETECTION)
The S-SCH also uses only the first 10% of a timeslot; Secondary Synchronisation Codes (SSC) are
transmitted. There are 16 different SSCs, which are organised in a 10 ms frame (15 timeslots) in
such a way, that
• the beginning of a 10 ms frame can be determined, and
• 64 different SSC combinations within a 10 ms frame are identified. There is a total of
512 primary scrambling codes, which are grouped in 64 scrambling code families, each
family holding 8 scrambling code members. The 15 SSCs in one 10 ms frame identify the
scrambling code family of the cell‘s downlink scrambling code.
• The sequence permits downlink frame synchronization and indicate which of the code
grouping the downlink scrambling code belongs to.
• This is done by correlating the received signal with all possible secondary
synchronization code sequences and identifying the maximum correlation value. Since
the cyclic shifts of the sequences are unique, the code group as well as the frame
synchronization is determined
88
Company Confidential
SSC Allocation for S-SCH
scrambling slot number
code group 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
group 00 1 1 2 8 9 10 15
15 8 10 16 2 7 15 7 16
group 01 1 1 5 16 7 3 14 16 3 10 5 12 14 12 10
group 02 1 2 1 15 5 5 12 16 6 11 2 16 11 15 12
group 03 1 2 3 1 8 6 5 2 5 8 4 4 6 3 7
group 04 1 2 16 6 6 11 15 5 12 1 15 12 16 11 2
group 05 1 3 4 7 4 1 5 5 3 6 2 8 7 6 8
group 62 9 11 12 15
15 12 9 13 13 11 14 10 16 15 14 16
group 63 9 12 10 15 13 14 9 14 15 11
11 11 13 12 16 10
I monitor
the S-SCH
11 15 5
89
Company Confidential
Common Pilot Channel (CPICH)
With the help of the SCH, the UE was capable to perform chip, TS, and frame synchronisation. Even
the cell‘s scrambling code group is known to the UE. But in the initial cell selection process, it does
not yet know the cell‘s primary scrambling code. There is one primary scrambling code in use over the
entire cell, and in neighbouring cells, different scrambling codes are in use. There exists a total of 512
primary scrambling codes.
The CPICH is used to transmit in every TS a pre-defined bit sequence (stream of 256 ‘1’) with a fixed
data rate of 30 kbps, which corresponds to spreading factor 256. The CPICH divides up into a
mandatory Primary Common Pilot Channel (P-CPICH) and optional Secondary CPICHs (S-CPICH).
The P-CPICH is in use over the entire cell. And it is the first physical channel, where a spreading code
is in use. A spreading code is the product of the cell‘s scrambling code and the channelisation code.
The channelisation code is fixed: Cch,256,0. I.e., the UE knows the P-CPICH‘s channelisation code, and it
uses the P-CPICH to determine the cell‘s primary scrambling code by trial and error (UE tries 8 SC
Codes of the group identified).
The P-CPICH is not only used to determine the primary scrambling code. It also acts as
- phase reference for most of the physical channels,
- measurement reference in the FDD mode (and partially in the TDD mode).
There may be zero or several S-CPICHs. Either the cell‘s primary scrambling code or its secondary
scrambling codes can be used. In contrast to the P-CPICH, it can be broadcasted just over a part of the
cell.
CPICH has activity factor of 100% (continuous transmission)
90
Company Confidential
Primary Common Pilot Channel (P-CPICH)
10 ms Frame
2560 Chips 256 Chips
Synchronisation Channel (SCH)
CP
P-CPICH
Cell scrambling
code? I get it with
trial & error! P-CPICH
applied spreading code =
cell‘s primary scrambling code ⊗ Cch,256,0
• Phase reference
• Measurement reference
91
Company Confidential
CPICH as Measurement Reference
The UE has to perform a set of L1 measurements, some of them refer to the CPICH channel:
• CPICH RSCP
RSCP stands for Received Signal Code Power. The UE measures the RSCP on the Primary-
CPICH. The reference point for the measurement is the antenna connector of the UE. The
CPICH RSCP is a power measurement of the CPICH. The received code power may be high, but
it does not yet indicate the quality of the received signal, which depends on the overall noise
level.
• UTRA carrier RSSI.
RSSI stands for Received Signal Strength Indicator. The UE measures the received wide band
power, which includes thermal noise and receiver generated noise. The reference point for the
measurements is the antenna connector of the UE.
• CPICH Ec/No
The CPICH Ec/No is used to determine the „quality“ of the received signal. It gives the received
energy per received chip divided by the band‘s power density. The „quality“ is the primary
CPICH‘s signal strength in relation to the cell noise. (Please note, that transport channel
quality is determined by BLER, BER, etc. )
If the UE supports GSM, then it must be capable to make measurements in the GSM bands,
too. The measurements are based on the
• GSM carrier RSSI
The wideband measurements are conducted on GSM BCCH carriers.
92
Company Confidential
P-CPICH as Measurement Reference
CPICH RSCP Received Signal Code Power (in dBm)
CPICH Ec/No received energy per chip divided by the power density in the band (in dB)
UTRA carrier received wide band power, including thermal noise and noise
RSSI generated in the receiver (in dBm)
94
Company Confidential
Primary Common Control Physical Channel (P-CCPCH)
10 ms Frame
2560 Chips 256 Chips
Synchronisation Channel (SCH)
CP
P-CCPCH
• WCEL: PtxPrimarySCH
Transmission power of the primary synchronization channel, the value is
relative to primary CPICH transmission power.
[-35 dB … 15 dB], step size 0.1 dB, default: -3 dB
• WCEL: PtxSecSCH
Transmission power of the secondary synchronization channel, the value is
relative to primary CPICH transmission power.
[-35 dB… 15 dB], step size 0.1 dB, default: -3 dB
97
Company Confidential
Nokia Parameters for Cell Search
• WCEL: PtxPrimaryCCPCH
This is the transmission power of the primary CCPCH channel, the value is
relative to primary CPICH transmission power.
[-35 dB … 15 dB], step size 0.1 dB, default: -5 dB
• WCEL: PriScrCode
Identifies the downlink scrambling code of the Primary CPICH (Common Pilot
Channel) of the Cell.
[0 ... 511], default: 0 dB
98
Company Confidential
Secondary Common Control Physical Channel (S-CCPCH)
More than one S-CCPCH can be deployed. The FACH and PCH information can multiplexed on
one S-CCPCH – even on the same 10 ms frame -, or they can be carried on different S-CCPCH’s.
When 2 S-CCPCH’s are broadcast, the first S-CCPCH has a spreading factor of 256 and
carries PCH, while the spreading factor of the remaining S-CCPCH can range between 256 (30
Kbps or 15 Ksps) and 4 (1920 Kbps) and carries FACH. UTRAN determines, whether a S-CCPCH
has the TFCI (Transport Format Combination Indicator) included (supports variable rates). Please
note, that the UE must support both S-CCPCHs with and without TFCI.
S-CCPCH is on air ONLY when there is data to transmit (FACH or Paging), however the TFCI bits
are broadcast irrespective of whether or not there is any data to transmit (min activity factor
25%)
Typical value is SF = 64 120 Kbps (60 Ksps)
99
Company Confidential
Secondary CCPCH Power Setting
100
Company Confidential
Secondary CCPCH Power Setting
101
Company Confidential
Secondary CCPCH Power Setting
• The power of SCCPCHs are set relative to CPICH transmission power, but it is
based on the bitrate.
– The Spreading factor for SCCPCH, which is carrying FACH (with or without
PCH), is 64 (60ksps)
– The Spreading factor for SCCPCH, which is carrying PCH only is 256 (15ksps)
– The Spreading factor for SCCPCH, which is carrying FACH-s/FACH-c idle for
SAB, is 128 (30ksps)
102
Company Confidential
Secondary Common Control Physical Channel (S-CCPCH)
10 ms Frame
TFCI
Data Pilot bits
(optional)
104
Company Confidential
S-CCPCH and the Paging Process
UTRAN
BCCH (SIB 5)
common
channel
definition,
UE including Node B RNC
a lists of
Index of S-CCPCHs
106
Company Confidential
S-CCPCH and its associated PICH
S-CCPCH frame,
τS-CCPCH
associated with PICH frame
τPICH
= 7680
PICH frame chips
10 ms
108
Company Confidential
Nokia Parameters for S-CCPCH and Paging
• WCEL: PtxPICH
This is the transmission power of the PICH channel. It carries the paging
indicators
which tell the UE to read the paging message from the associated secondary
CCPCH. This parameter is part of SIB 5.
[-10 dB..5 dB]; step 1 dB; default: -8 dB (with Np =72)
NP
Repetition of PICH bits
[18, 36, 72, 144] with relative power [-10, -10, -8, -5] dB
• RNC: CNDRXLength
The DRX cycle length used for CN domain to count paging occasions for
discontinuous reception. This parameter is given for CS domain and PS domain
separately. This parameter is part of SIB 1.
[640, 1280, 2560, 5120] ms; default = 640 ms.
• WCEL: UTRAN_DRX_length
The DRX cycle length used by UTRAN to count paging occasions for
discontinuous reception.
[80, 160, 320, 640, 1280, 2560, 5120] ms; default = 320 ms
109
Company Confidential
FACH and S-CCPCH
• The transport channel Forward Access Channel (FACH) is used, when relatively small amounts of data have to
be transmitted from the network to the UE. In-band signalling is used to indicate, which UE is the recipient
of the transmitted data (see MAC PDU with UE-ID type). This common downlink channel is used without
(fast) closed loop power control and is available all over the cell. FACH data is transmitted in one or several
S-CCPCHs. FACH and PCH data can be multiplexed on one S-CCPCH, but they can also be be transmitted on
different S-CCPCHs.
• The FACH is only transmitted downlink. The FACH is organised in FACH Data Frames via the Iub-interface.
Each FACH Data Frames holds the Transmission Blocks for one TFS. The used TFS is identified by the TFI. A TFI
is associated with one Transmission Time Interval (TTI), which can be either 10, 20, 40 or 80 ms. The TTI
identifies the interleaving time on the radio interface.
A FACH Data Frame has header fields, which identify the CFN, TFI, and the Transmit Power Level.
• The Transmit Power Level gives the preferred transmission power level for the FACH and for the TTI time. The
values specified here range between 0 and 25.5 dB, with a step size of 0.1 dB. The value is taken as a
negative offset to the maximum power configured for the S-CCPCHs, specified for the FACH.
• The pilot bits and the TFCI-field may have a relative power offset to the power of the data field, which may
vary in time. (The offset is determined by the network.) The power offsets are set by the NBAP message
COMMON TRANSPORT CHANNEL SETUP REQUEST, which is sent from the RNC to the Node B. There are two
power offset information included:
• PO1:defines the power offset for the TFCI bits; it ranges between 0 and 6dB with a 0.25 step size.
• PO3:defines the power offset for the pilot bits; it ranges between 0 and 6dB with a 0.25 step size.
Another important parameter is the maximum allowed power on the FACH: MAX FACH Power.
110
Company Confidential
FACH and S-CCPCH
max. transmit
power for S-CCPCH
UE
0..25.5 dB,
step size 0.1
• WCEL: PowerOffsetSCCPCHPilot
Defines the power offset for the pilot symbols relative to the downlink
transmission power of a Secondary CCPCH. This parameter is part of SIB 5.
P03_15/30/60
15 ksps : [0..6 dB]; step 0.25 dB; default: 2 dB
30 ksps : [0..6 dB]; step 0.25 dB; default: 3 dB
60 ksps : [0..6 dB]; step 0.25 dB; default: 4 dB
112
Company Confidential
Code Tree Capacity
Note: there are not P-SCH and S-SCH !! P-SCH and S- S-SCH
are not under the cell-
cell- specific primary scrambling code
(the UE must be able to synchronize to the cell before
knowing the downlimk scrambling code)
113
Company Confidential
Part V
Power Control
114
Company Confidential
Effect of TX & RX Powers on Interference Levels
Near-Far Problem
Pr,1
Pr,2
P = 21 dBm
P = 21 dBm
PL1 = 100 dB
MS2
MS1
(S/N)1 = Pr,1 - Pr,2 = -10 dB MS2 must be Power Controlled by -10 dB to have
(S/N)2 = Pr,2 - Pr,1 = +10 dB the same S/N for both users MS1 and MS2
116
Company Confidential
Near-Far Effect
117
Company Confidential
Purpose of Power Control in WCDMA
118
Company Confidential
Physical Random Access (Open loop Power Control)
Outer Loop Power Control
Fast Closed Loop (Inner) Power Control
119
Company Confidential
Physical Random Access (Open loop Power Control)
In the random access (based on Slotted ALOHA approach with fast acquisition indication) ,
initiated by the UE (MOC), two physical channels are involved:
UE Node B
No response PRACH (preamble)
by the
Node B
OLA!
AICH
121
Company Confidential
Random Access Timing
The properties of the PRACH are broadcasted (SIB5, SIB6). The candidate PRACH is randomly
selected (if there are several PRACH advertised in the cell) as well as the access slots (= 2 TIME
SLOTS) within the PRACH. 15 access slots are given in a PRACH, each access slot lasting two
timeslots or 5120 chips. In other words, the access slots stretch over two 10 ms frames.
frames A
PRACH preamble, which is transmitted in an access slot, has a length of 4096 chips.
Also the AICH is organised in (AICH) access slots, which stretch over two timeslots. AICH access
slots are time aligned with the P-CCPCH. (Activity factor 80%)
The UE sends one preamble in uplink access slot n. It expects to receive a response from the
Node B in the downlink (AICH) access slot n, τp-a chips later on. If there is no response, the UE
sends the next preamble τp-p chips after the first one. The maximum numbers of preambles in
one preamble access attempt can be set between 1 and 64. The number of PRACH preamble
cycles can be set between 1 and 32.
If the AICH_Transmission_Timing parameter in the SIB is set to BCCH SIB5 & SIB6 to
•0,
then, the minimum preamble-to-preamble distance is 3 access slots, the minimum preamble-to-message
distance is 3 access slots, and the preamble-to-acquisition indication is 3 timeslots.
•1,
then, the minimum preamble-to-preamble distance is 4 access slots, the minimum preamble-to-message
distance is 4 access slots, and the preamble-to-acquisition indication is 5 timeslots.
122
Company Confidential
Random Access Timing
SFN mod 2 = 0 SFN mod 2 = 1 SFN mod 2 = 0
P-CCPCH
AICH access
slots 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 0 1 2 3 4 5 6 7
5120
chips
(distances depend on AICH_Transmission_Timing )
UE point of view
Acquisition
AICH Indication
access slots
preamble-to-AI AS # i
distance τp-a
4096 chips
5120 chips AS # i
preamble-to-preamble preamble-to-message
123
distance τp-p distance τp-m
Company Confidential
PRACH Power Setting
Preamble_Initial_Power =
UL interference
1st preamble: + Primary CPICH TX power – CPICH_RSCP
power setting + Constant Value
Constant Value
attenuation
estimated receive level
in the DL
UL interference
at Node B -5..10
dB
“RRC
Connection
1..8 dB Request”
Pp-p Pp-m
Pp-p
Pre- Message
Pre- part
Pre- amble
amble
amble
125
Company Confidential
Nokia Parameters Related to the PRACH and AICH
• WCEL: RACH_tx_Max
Maximum number of RACH preamble cycles defines how many times the PRACH
pre-amble ramping procedure can be repeated before UE MAC reports a failure on
RACH transmission to higher layers. This message is part of SIB5/6.
[1 ... 32]; default 8. We use 16
WCEL: PRACHScramblingCode
The scrambling code for the preamble part and the message part of a PRACH
Channel, which is part of SIB5/6.
[0 ... 15]; default 0.
• WCEL: AllowedPreambleSignatures
The preamble part in a PRACH channel carries one of 16 different orthogonal
complex signatures. Nokia Node B restrictions: A maximum of four signatures can be
allowed (16 bit field).
[0 ... 61440]; default 15. We use 4
• WCEL: AllowedRACHSubChannels
A RACH sub-channel defines a sub-set of the total set of access slots (12 bit field).
[0 ... 4095]; default 4095.
126
Company Confidential
Nokia Parameters Related to the PRACH and AICH
• WCEL: PtxAICH
This is the transmission power of one Acquisition Indicator (AI) compared to CPICH
power. This parameter is part of SIB 5.
[-22 ... 5] dB, step 1 dB; default: -8 dB.
• WCEL: AICHTraTime
AICH transmission timing defines the delay between the reception of a PRACH
access slot including a correctly detected preamble and the transmission of the
Acquisition Indicator in the AICH.
0 ( Delay is 0 AS), 1 ( Delay is 1 AS) ;default 0.
• WCEL: RACH_Tx_NB01min
In case that a negative acknowledgement has been received by UE on AICH a backoff
timer TBO1 is started to determine when the next RACH transmission attempt will be
started. The backoff timer TBO1 is set to an integer number NBO1 of 10 ms time
intervals, randomly drawn within an Interval 0 ≤ NB01min ≤ NBO1 ≤ NB01max (with
uniform distribution).
[0 ... 50]; default: 0.
• WCEL: RACH_Tx_NB01max
[0 ... 50]; default: 50.
127
Company Confidential
Outer Loop Power Control
“RRC Conn
Setup”
UL DPDCH “RRC
Conn Setup
Complete”
“RRC Conn
Request”
OL PC is needed to keep the quality of the communication at the required level (BLER,
SIR, BER,…) by setting the target (SIR) for the fast power control. It aims at providing the
required quality: no worse, no better. Too high quality would waste capacity. It is needed
in both UL and DL since there is Fast PC (Closed Loop or Inner Loop) in both UL and DL
128
Company Confidential
Outer Loop Power Control
In RADIO BEARER SETUP Message you can find the Target BLER (for the DL)
For AMR and PS 128 = 1% BLER, CS T (VIDEO) = 0.1%, CS NT = 0.2%
129
Company Confidential
UL Outer Loop Power Control Algorithm
130
Company Confidential
UL Outer Loop Power Control Algorithm
131
Company Confidential
UL OL PC: BLER Eb/No (Initial SIR Target, SIR Target Max, SIR Target Min)
132
Company Confidential
DL Outer Loop Power Control
The adjustments of the SIR Target done by the UE is a proprietary algorithm that
provides the same measured quality (BLER) as the quality target set by the RNC
133
Company Confidential
Fast Closed Loop (Inner) Power Control
134
Company Confidential
Fast Closed Loop (Inner) Power Control
• UL (Near-Far Problem): UE1 and UE2 operate within the same frequency,
separable at the base station only by their respective spreading codes. It
may happen that UE1 at the cell edge suffers a path loss, say 70 dB above
that of UE2 which is near to NodeB. If there were no mechanism for UE1
and UE2 to be power-controlled to the SAME level at the NodeB, UE2 could
easily overshoot UE1 and thus a large part of the cell. Power control tries
to equalizes the Rx power per bit of all UE’s at NodeB. Since Fast Fading is
uncorrelated between uplink and downlink (large freq separation between ul
and dl bands in FDD) we can not use only a method based on Open Loop
Power Control. Solution: Closed Loop PC: in UL the NodeB performs
frequent (1.5 KHz) estimates of the received SIR and compares it to the
SIR Target (calculated during Outer Loop PC).
• DL: We do not have Near-Far Problem due to one-to-many scenario: all the
signals within one cell originate from one NodeB to all mobiles. However it is
desirable to provide a marginal amount of additional power to UE at the cell
edge, as they suffer from increased other-cell-interference.
135
Company Confidential
DL Fast Closed (Inner) Loop Power Control
Inner loop power control is also often called (fast) closed loop power control. It takes place between
the UE and the Node B. We talk about UL inner loop power control, when the Node B returns
immediately after the reception of a UE‘s signal a power control command to the UE. By doing so,
the UE‘s SIR ratio is kept at a certain level.
DL inner loop power control control is more complex. When the UE receives the transmission of the
Node B, the UE returns immediately a transmission power control command to the Node B, telling
the Node B either to increase or decrease its output power for the UE‘s DPCH. The Node B‘s
transmission power can be changed by 0.5, 1, 1.5 or 2 dB. 1 dB must be supported by the
equipment. If other step sizes are supported or selected, depends on manufacturer or operator.
The transmission output power for a DPCH has to be balanced for the PICH, which adds to the
power step size.
There are two downlink inner loop power control modes:
• DPC_MODE = 0: Each timeslot, a unique TPC command is sent uplink.
• DPC_MODE = 1: 3 consecutive timeslots (for DL), the same TPC command is transmitted.
One reason for the UE to request a higher output power is given, when the QoS target has not been
met. It requests the Node B to transmit with a higher output power, hoping to increase the quality
of the connection due to an increased SIR at the UE‘s receiver. But this also increases the
interference level for other phones in the cell and neighbouring cells. The operator can decide,
whether to set the parameter Limited Power Increase Used. If used, the operator can limit the
output power raise within a time period.
136
Company Confidential
DL Fast Closed (Inner) Loop PC Algorithm
137
Company Confidential
Downlink Inner Loop Power Control
TPC
two modes
cell
DPC_MODE = 0 DPC_MODE = 1
138
Company Confidential
UL Inner Loop Power Control
SIRest
SIRtarget
time
T TC TCP
TC CP P = = 0
P= = 0 TPC ⇒
1 1
TPC_cmd
in FDD mode:
1500 times per second
139
Company Confidential
UL Fast Closed (Inner) Loop PC Algorithm
140
Company Confidential
UL Inner Loop Power Control
Power Control Algorithm 1 is applied in medium speed environments. Here, the UE is
commanded to modify its transmit power every timeslot. If the received TPC value is 1, the UE
increases the transmission output at the DPCCH by ∆DPCCH, otherwise it decreases it by ∆DPCCH.
The ∆DPCCH is either 1 or 2 dB, as set by the higher layer protocols. TPC values from the same
radio link set represent one TLC_cmd. TPC_cmds from different radio link sets have to be
weighted, if there is no reliable interpretation.
Power Control Algorithm 2 (300 times/s) was specified to allow smaller step sizes in the
power control in comparison to PCA1. This is necessary in very low and high speed
environments. In these environments, PCA1 may result in oscillating around the target SIR.
PCA2 changes only with every 5th timeslot, i.e. the TPC_cmd is set to 0 (do not do anything)
the first 4 timeslots. In timeslot 5, the TPC_cmd is –1, 0, or 1. For each radio set (Radio set is
combined radio links from same NodeB), the TPC_cmd is temporarily determined. This can be
seen in the next figure. The temporary transmission power commands (TPC_temp) are
combined as can be seen in the figure after the next one. Here you can see, how the final
TPC_cmd is determined.
141
Company Confidential
UL Inner Loop Power Control Algorithms (1 and 2)
• The optimum PC step size varies depending on the UE speed. For a given
quality target, the best UL PC step size is the one that results in the lowest
target SIR. With an update rate of 1500 Hz, a PC step size of 1dB can
effectively track a typical Rayleigh fading channel up to Doppler frequency of
about 55 Hz (30 Km/h). At higher speeds, up to about 80 Km/h, a PC step
size of 2dB gives better results.
• For speeds greater than 80 Km/h the inner loop PC can no longer follow the
fades and just introduces noise into the UL transmission. This adverse effect
on the UL performance could be reduced if a PC step size smaller than 1 dB
was employed. Also, for UE speeds lower than about 3 Km/h where the
fading rate of the channel is very small, a smaller step size is more
beneficial.
• Algorithm 1 is used when the UE speed is sufficiently low to compensate for
the fading of the channel (PC step size should be 1 or 2 dB)
• Algorithm 2 was designed for emulating the effect of using a PC step size
smaller than 1 dB and can be used to compensate for the slow fading trend
of the propagation channel rather than rapid fluctuations. It performs better
than Alg 1 when the UE moves faster than 80 Km/h or slower than 3 Km/h.
The UE does not change its transmission power until it has received 5
consecutive TPC commands. 142
Company Confidential
UL Inner Loop Power Control
algorithms for processing power
control commands TPC_cmd
PCA1 PCA2
TPC_cmd for each TS TPC_cmd for 5th TS
TPC_cmd values: +1, -1 TPC_cmd values: +1, 0, -1
step size ∆ TPC: 1dB or 2dB step size ∆ TPC: 1dB
0 ≈3 ≈ 80 km/h
Rayleigh fading can be compensated
143
Company Confidential
Soft Handover Case: UL Inner Loop Power Control Algorithm 1
“UP” TPC3 = 1
⇒ “Down”
TPC_cmd = -1
TPC1 = 1 TPC3 = 0
“UP” “Down”
Cell 1
Cell 2
144
Company Confidential
No Soft Handover Case - UL Inner Loop Power Control Algorithm 2 (Part 1)
TPC = 1 TPC_temp
TPC = 1 0
TPC = 1 0
0 • if all TPC-values = 1
TPC = 1
⇒ TPC_temp = +1
TPC = 1 0
• if all TPC-values = 0
TPC = 1 1
⇒ TPC_temp = -1
TPC = 0 0 • otherwise
TPC = 1 0 ⇒ TPC_temp = 0
TPC = 0 0
TPC = 1 0
TPC = 0 0
TPC = 0 0
TPC = 0 0
TPC = 0 0
TPC = 0 0
-1
145
Company Confidential
Soft Handover Case :UL Inner Loop Power Control Algorithm 2 (Part 2)
Example:
1 N
∑
N i=1
TPC_tempi
-1 -0.5 0 0.5 1
TPC_cmd = - 0 1 146
1 Company Confidential
Part VI
Dedicated Physical
Channels
147
Company Confidential
Downlink Dedicated Physical Channel (DPCH)
The downlink DPCH is used to transmit the DCH data. Control information and user data are time
multiplexed. The control data is associated with the Dedicated Physical Control Channel (DPCCH),
while the user data is associated with the Dedicated Physical Data Channel (DPDCH). Data is not
only User Traffic/Data but could be also High layer signalling (“Measurement control”, “RRC
messages”,L3 Dedicated signalling)
The transmission is organised in 10 ms radio frames, which are divided into 15 timeslots. The
timeslot length is 2560 chips. Within each timeslot, following fields can be found:
• Data field 1 and data field 2, which carry DPDCH information
• Transmission Power Control (TPC) bit field
• Transport Format Combination Indicator (TFCI) field, which is optional
• Pilot bits
The exact length of the fields depends on the slot format, which is determined by higher layers. The
TFCI is optional, because it is not required for services with fixed data rates. Slot format are also
defined for the compressed mode; hereby different slot formats are in used, when compression is
archived by a changed spreading factor or a changed puncturing scheme. The pilot sequence is
used for channel estimation as well as for the SIR ratio determination within the inner loop power
control. The number of the pilot bits can be 2, 4, 8 and 16 – it is adjusted with the spreading
factor. A similar adjustment is done for the TPC value; its bit numbers range between 2, 4 and 8.
The spreading factor for a DPCH can range between 4 and 512. The spreading factor can be
changed every TTI period.
Superframes last 720 ms and were introduced for GSM-UMTS handover support.
148
Company Confidential
Downlink Dedicated Physical Channel (DPCH)
Superframe = 720 ms
10 ms Frame
TFCI
TPC
Data 1 bits bits Data 2 bits Pilot bits
bits
(optional)
Multicode usage:
Several physical channels can be allocated in the downlink to one UE. This can occur, when
several DPCH are combined in one CCTrCH in the PHY layer, and the data rate of the CCTrCH
exceeds the maximum data rates allowed for the physical channels. Then, on all downlink
DPCHs, the same spreading factor is used. Also the downlink transmission of the DPCHs takes
place synchronous. One DPCH carries DPDCH and DPCCH information, while on the remaining
DPCHs, no DPCCH information is transmitted.
But also in the case, when several DPCHs with different spreading factors are in use, the first
DPCH carries the DPCCH information, while in the remaining DPCHs, this information is
omitted (discontinuous transmission).
151
Company Confidential
Downlink Dedicated Physical Channel (DPCH)
maximum bit rate discontinuous transmission with lower bit rate
TS TS TS TS TS
Multicode usage:
DPCH 1
TS TS TS
DPCH 2
TS TS TS
DPCH 3
152
Company Confidential
• Power offsets
Power Offsets for the DPCH • TFCS
• DL DPCH slot
format
• FDD DL TPC step
size
P0x: 0..6 dB
step size: 0.25
dB UE
• RNC: PowerOffsetDLdpcchTpc,
The parameter defines the power offset for the TPC symbols relative to the data
symbols in dedicated downlink physical channel
[0 … 6 dB]; step size 0.25 dB; default: 3 dB for 12.2 kbps
• RNC: PowerOffsetDLdpcchTfci,
The parameter defines the power offset for the TFCI symbols relative to the data
symbols in dedicated downlink physical channel.
[0 … 6 dB], step size 0.25 dB; default: 3 dB for 12.2 kbps
154
Company Confidential
Uplink Dedicated Physical Channels
The uplink dedicated physical channel transmission, we identify two types of physical channels:
Dedicated Physical Control Channel (DPCCH),
which is always transmitted with spreading factor 256 (3840/256=15Ksps=15Kbps). Following
fields are defined on the DPCCH:
- Pilot bits for channel estimation. Their number can be 3, 4, 5, 6, 7 or 8.
- Transmitter Power Control (TPC), with either one or two bits
- Transport Format Combination Indicator (TFCI), which is optional, and a
- Feedback Indicator (FBI). Bits can be set for the closed loop mode transmit diversity and site
selection diversity transmission (SSDT)
6 different slot formats were specified for the DPCCH. Variations exist for the compressed mode.
Dedicated Physical Data Channel (DPDCH),
which is used for user data transfer (Data is not only User Traffic/Data but could be also
High layer signalling (“Measurement Reports”, “RRC messages”,L3 signalling) . Its
spreading factor ranges between 4 and 256. 7 different slot formats are defined, which are set by
the higher layers.
The DPCCH and DPDCH are combined by I/Q code multiplexing with each multiframe.
Multicode usage is possible. If applied, additonal DPDCH are added to the uplink transmission, but
no additional DPCCHs! The maximum number of DPDCH is 6; when more than one DPDCH is used
(Multicodes) they all use SF = 4.
The transmission itself is organised in 10 ms radio frames, which are divided into 15 timeslots. The
timeslot length is 2560 chips.
155
Company Confidential
Uplink Dedicated Physical Channels
Superframe = 720 ms
10 ms Frame
157
Company Confidential
Discontinuous Transmission and Power Offsets
DPDCH
DPDCH
DPDCH
DPCCH DPCCH DPCCH
TTI TTI TTI
159
Company Confidential
Initial Uplink DCH Transmission
When we look to the PRACH, we can see, that a preambles were used to avoid UEs to access
UTRAN with a too high initial transmission power. The same principle is applied for the DPCH.
After PRACH procedure the UE transmits between 0 to 7 radio frames only the DPCCH uplink
(the period is called DPCCH power control Preamble), before the DPDCH is code multiplexed.
The number of radio frames is set by the higher layers (RRC resp. the operator). Also for this
period of time, only DPCCH can be found in the downlink.
The UE can be also informed about a delay regarding RRC signalling – this is called SRB delay,
which can also last 0 to 7 radio frames. The SRB delay follows after the DPCCH preamble.
How to set the the transmission power of the first UL DPCCH preamble? Its power level is
DPCCH_Initial_power = – CPICH_RSCP + DPCCH_Power_offset
The DPCCH Power Offset is retrieved from RRC messages. It’s value ranges between –164 and
–6 dB (step size 2 dB). CPICH_RSCP is the received signal code power on the P-CPICH,
measured by the UE.
160
Company Confidential
Initial Uplink DCH Transmission
DPCCH only DPCCH & DPDCH
reception
at UE
trans-
mission
at UE T0
0 to 7 frames for
power control preamble
DPCCH only, DPCCH & DPDCH
always based on PCA1 PCA based on RRC
DPCCH_Initial_power = – CPICH_RSCP + DPCCH_Power_offset 161
Company Confidential
Radio frame timing and access slot timing of downlink physical channels
Primary
SCH
Secondary
SCH
Any CPICH
τPICH
10 ms 10 ms
162
Company Confidential
Part VII
WCDMA Planning
163
Company Confidential
Radio Network Planning Process
164
Company Confidential
Planning issues
• Planning should meet current standards and demands and also comply with future
requirements.
• Uncertainty of future traffic growth and service needs.
• High bit rate services require knowledge of coverage and capacity enhancements
methods.
• Real constraints
– Coexistence and co-operation of 2G and 3G for old operators.
– Environmental constraints for new operators.
• Network planning depends not only on the coverage but also on load.
165
Company Confidential
Planning methods
• Preparation phase
– Defining coverage and capacity objectives
– Selection of network planning strategies
– Initial design and operation parameters
• Initial dimensioning
– First and most rapid evaluation of the network elements count and capacity of
these elements
– Offered traffic estimation
– Joint capacity coverage estimation
• Detailed planning
– Detailed coverage capacity estimation
– Iterative coverage analysis
– Planning for codes and powers
• Optimization
– Setting the parameters
• Soft handover
• Power control
• Verification of the static simulator with the dynamic simulator
166
Company Confidential
A strategy for dimensioning
• Plan for adequate load and number of sites.
• Enable optimized site selection.
• Avoid adding new sites too soon.
• Allow better utilization of spectrum.
• Recommended load factor 30- 70 %
167
Company Confidential
Dimensioning process
168
Company Confidential
Detailed Planning – Workflow of RNP Tool
169
Company Confidential
Input data preparation
• Digital map.
- for coverage prediction.
- topological data (terrain), morphological data (terrain type), building location
and height.
- Resolution: urban areas 10-20m, rural areas 50-100m.
• Plan.
- logical concept combining various items.
digital map, map properties, target plan area, selected radio access
technology, input parameters, antenna models.
• Antenna editor.
- logical concept containing antenna radiation pattern, antenna gain, frequency
band.
• Propagation model editor.
- Different planning areas with different characteristics.
- For each area type many propagation models can be prepared.
- tuning based on field measurements.
• BTS types and site/cell templates
- Defaults for the network element parameters and ability to change it.
- Example BTS parameter template:
maximum number of wideband signal processors.
maximum number of channel units.
noise figure. 170
Tx/Rx diversity types. Company Confidential
Planning
• Importing sites.
– Utilization of 2G networks.
• Optimising dominance.
– Interference and capacity analysis.
– Locating best servers in each location in the service area.
– Target to have clear dominance areas. 171
Company Confidential
Iterative traffic planning process
• Monte-Carlo simulation.
– Finding average over many snapshots: average, minimum, maximum, std.
– Averages over mobile locations.
– Iterations are described by:
• Number of iterations.
• Maximum calculation time.
• Mobile list generation.
• General calculation settings.
172
Company Confidential
Example of WCDMA analysis
• Reporting:
– Raster plots from the selected area.
– Network element configuration and parameter setting.
– Various graphs and trends.
– Customized operator specific trends.
173
Company Confidential
Uplink iteration step
• Allocate MS transmit powers so that
the interference levels and BS
sensitivities converge.
174
Company Confidential
Downlink iteration step
UL DCH Coverage
– Whether an additional mobile having certain bit rate could be served.
– The transmit power need for the MS is calculated and compared to the
maximum allowed
DL DCH Coverage
– Pixel by pixel is checked whether an additional mobile having certain bit
rate could be served. Concentration on the power limits per radio link.
– The transmit power need for supporting the link is calculated and
compared to the maximum allowed
DL CPICH Coverage
– Pixel by pixel is checked whether the P-CPICH channel can be listened.
176
Company Confidential
Examples of Plots/Arrays
177
Company Confidential
Examples of Plots/Arrays
178
Company Confidential
Examples of Plots/Arrays
179
Company Confidential
Examples of Plots/Arrays
180
Company Confidential
Capacity&Coverage Trade Off
• The coverage for a WCDMA system is generally limited by the uplink. This is
because the maximum output power of the mobile is lower than for the base
station, so the base station can reach longer than the mobile can.
• Due to the increase of traffic, the effective cell area has shrunk. This behavior is
known as cell breathing.
breathing In an FDMA or TDMA-system this problem does not arise,
since coverage and capacity is largely independent.
• To reduce cell breathing interference margins are included when dimensioning the
network, which has the effect of increasing site density. 181
Company Confidential
Coverage Limited Uplink
• Another way to reduce cell breathing would be to add a frequency, which would
mean that the users could be spread over two or more carriers. Since the different
carriers are not interfering with each other, the interference level is reduced, and
an increase in capacity or coverage is achieved
• When making the initial design, the aim is to provide a certain capacity, or service
level, over an area. One design strategy could be to design a very low-density
network, capable of providing low capacity over a wide area.
• This would reduce the number of base stations as compared to building for higher
capacity. Since the cost of base stations are a large part of the cost of building a
network, minimizing the number of base stations are important.
• On the other hand, it is important to be able to provide attractive services to the
customers. This could be difficult if not enough bandwidth is available. Building
less dense means that the maximum distance between the mobile and base
station is increased, which is the same as allowing a higher maximum path loss
between the two.
• A higher path loss between the mobile and the base station can be tolerated if the
interference is decreased. If the interference in a cell were reduced by a certain
amount of dB, the maximum allowed path loss would increase by the same
amount.
182
Company Confidential
Coverage Limited Uplink
• Using a propagation model like for example Okumura-Hata, it is possible to
convert a change of the interference level into a changed site density, compared
to a reference case.
• Table below shows the change in number of sites if the interference margin in
the link budget is changed. A negative dB value means that the link budget is
worse compared to the reference case, and thus more sites are needed.
183
Company Confidential
Uplink Load Factor
Interference degradation margin: describes the amount of increase of interference
due to multiple access . It is reserved in the link budget.
Can be calculated as the Noise Rise: the ratio of the total received power Itotal to
the Noise Power PN
I total 1 1
Noise Rise = = =
1 − ∑ j =1 L j 1 − ηUL
N
PN
Where Load Factor η ULis :
ηUL = ∑ j =1 L j , L j is the load factor of one connection
N
ζ 1+
W
( Eb / N O ) j ⋅ R j ⋅ υ j
Noise Rise (dB) is equal to - 10 ⋅ log10 (1 − ηUL )
Where ηUL ∈ [0,1]
184
Company Confidential
Uplink Load Factor
Definitions Recommended Values
Eb/No Signal energy per bit divided by noise spectral Dependent on service, bit rate, mulitpath,
density that is required to meet a predefined fading channel, receive antenna diversity,
BLER. Noise includes both thermal and mobile speed, etc
interference
NS Number of Sectors
i Other cell to own cell interference ratio seen by Macro Cell with omni antennas: 55%. Macro
the base station receiver Cell with 3 sectors: 65% 185
Company Confidential
Uplink Noise Rise as a Function of
Throughput
12
11
Noise Rise (dB)
10
9
8
7
6
5
4
3
2
1
Throughput (Kbps)
144 Kbps
Voice (12.2 Kbps)
186
Company Confidential
Coverage Limited Uplink
• For voice services a typical noise rise would be between 1-3 dB, which
corresponds to a throughput between 150 kbps and 375 kbps. A network is
designed for a certain throughput.
• After some time that throughput is reached, and as a result the noise rise rises
over the design value.
• The choice is then to either increase site density, or add more frequencies.
Adding a frequency has its own set of problems, most notably that soft handover
does not work between frequencies. This problem is less of an issue if new
frequencies are added to a number of sites over a wider area.
• The mobile can then move freely on the frequency it has been assigned, and the
probability of making a hard inter-frequency handover is reduced
• Assume that traffic increases so that the actual noise rise is 4 dB, 1 dB above the
design level. The noise figure needs to be improved, for example down to 2 dB,
to improve quality and make room for future capacity demands. In other words,
the average throughput per cell needs to be reduced.
• Building more sites, or adding another frequency can do this.
• Adding a second frequency would half the throughput for each cell and carrier.
For a 4 dB noise rise the throughput is 450 kbps according to the graph. A new
throughput of 450/2 kbps=225 kbps per carrier gives a noise rise of 1.5 dB, an
improvement of 2.5 dB. A 2.5 dB lower allowable path loss corresponds roughly
to 40% more sites, that is. The cost of building these sites can then be said to be
the value of having one extra frequency. Adding a second and a third frequency
follows the same pattern, with a slight difference. The relative decrease in noise
rise will be lower. When a third frequency is added the traffic is spread over three
carriers, and reduced with a third for each frequency. 187
Company Confidential
Coverage Limited Uplink
• It is also possible to do the other way around, that is, build sites less dense to
start with. This saves money in the roll out phase, but could cause problems if
high capacity is needed in the future. Using figures from the example above,
assume a design for a maximum throughput of 375 kbps for one carrier, which
corresponds to a noise rise of 3 dB. Using two carriers gives a throughput per
carrier of 375/2 kbps=190 kbps, which corresponds to a noise rise of 1.3 dB.
The saving is 2.7 dB, which converts to roughly 70% of the original number of
sites is needed. This is the same as each site covers approximately 1.4 times
the area of the original one carrier site.
2.5
Range [km]
1.5
0.5
0
32 kbps 64 kbps 144 kbps 384 kbps 1024 kbps 2048 kbps
(Eb No ) j
N rf ⋅W ⋅ L ⋅ ∑ j =1υ j
N
BS _ TxPw =
(W Rj )
1 − η DL
Where N rf is the noise spectral density of the mobile
N rf = k ⋅ T + NF = −174dBm + NF (assuming T = 290K)
k is the Boltzmann constant of 1.381⋅10-23 J / K , NF is mobile Noise Figure (5 - 9 dB)
190
Company Confidential
Downlink Common Channels
• Part of the downlink power has to be allocated for the common channels that
are transmitted independently of the traffic channels
Downlink common Relative to CPICH Activity Average Power allocation
channels with 20W max Power
CPICH 0 dB 100% 2.0 W (33 dBm)
* S-CCPCH control (TFCI) bits transmitted with higher power than data bits
192
Company Confidential
Relation of Uplink and Downlink Load
100
90
• Downlink load is always
80 higher than uplink load due to:
70
– asymmetry in user traffic
DL Load [%]
60
– different Eb/No values in
50 Increasing uplink and downlink
asymmetry
40
– orthogonality in downlink
30
– overhead due to soft-
20 handover
10
0
0 10 20 30 40 50
UL Load [%]
193
Company Confidential
Capacity Limited Downlink
• As the demand for downlink capacity increases, there are several different ways of
increasing capacity. The most common ways are adding more frequencies and
power amplifiers, and introducing transmit diversity
• Upgrading capacity in the ways just mentioned is of course dependant on the base
station equipment being able to handle it. It is reasonable to assume that as the
capacity demand increases, the equipment vendors will produce equipment that
can handle it
• Assume an initial base station configuration of one 20W power amplifier per sector,
one carrier per sector and three sectors per site. This is called the baseline
configuration, and has a baseline capacity
• The first step to upgrade the capacity is to add a second frequency. This gives a
capacity increase of 80%. The reason why the increase is not 100% is that the
power amplifier only can deliver 20W, which has to be split between the two
frequencies, making the output 10W per carrier.
• The second step could be to add a second 20W power amplifier (restoring the
power per frequency to 20W) and introduce transmit diversity (STTD, Closed loop
mode 1, Closed loop mode 2).
• With these two upgrades the capacity now is 180% compared to the baseline.
Adding a third frequency would decrease the output power to 13 W per carrier, but
the extra carrier would still mean a capacity increase of 290% compared to the
baseline
• If there are no more frequencies available, changing the power amplifiers from two
20W to two 40W will give a modest capacity increase, making the increase
compared to baseline 320%. Adding a fourth frequency and at the same time
changing out the two 20W power amplifiers to two 40W amplifiers, if that has not
194
been done before, gives a capacity increase 460% compared to the baseline.
Company Confidential
Capacity Limited Downlink
• Upgrading the power amplifier restores the power per frequency to 20W, the
same as the baseline case. With the stronger PA’s there is power to add a fifth
and a sixth carrier. This would give capacities of 550% respectively 680%
compared to the baseline
• Using two PA’s means that no modification to the antenna system is required.
Adding a third PA means that either a combiner, or an extra antenna needs to be
used. A combiner typically has a 3dB insertion loss, offsetting the gain achieved
• Adding a third antenna is complicated from a site-engineering point of view. An
extra feeder cable is needed, and adding an extra antenna could be difficult since
it means renegotiating the agreement with the house owner.
• With a third PA the 6 frequencies is transmitting at 20W, giving a 740% increase
gain compared to the baseline capacity.
160,00
145,00
Coverage is
uplink limited Capacity is
downlink limited
140,00
0 10 20 30 40 50 60 70 80
UL Load
196
Company Confidential
Part VIII
WCDMA Link Budget
197
Company Confidential
WCDMA Link Budget
198
Company Confidential
WCDMA Link Budget
199
Company Confidential
WCDMA Link Budget
200
Company Confidential
WCDMA Link Budget
201
Company Confidential
WCDMA Link Budget
202
Company Confidential
WCDMA Link Budget
203
Company Confidential
WCDMA Link Budget
204
Company Confidential
WCDMA Link Budget
205
Company Confidential
WCDMA Link Budget
206
Company Confidential
WCDMA Link Budget
207
Company Confidential
WCDMA Link Budget
208
Company Confidential
WCDMA Link Budget
209
Company Confidential
WCDMA Link Budget
210
Company Confidential
WCDMA Link Budget
211
Company Confidential
WCDMA Link Budget – Cell Sizes
•Output of Link Budget is MAPL (Maximum Allowed Path Loss) based on different:
- Clutter types (Dense Urban, Urban, Sub-Urban, Rural)
- Services (AMR, PS64, CS64, PS128, PS384,…)
- Indoor/Outdoor
- Area Location Probability
- Mobile speed: Pedestrian/Vehicular
•Given an area to be covered (Km²) the cell count has to be performed based on
Cell Area ( Cell Radius)
•Cell Radius can be calculated using Propagation models (Cost231, Hokumura-
Hata, Walfish-Ikegami,…)
•The models need as input:
•MAPL
•UE antenna height
•NodeB antenna height
•Frequency
•Clutter correction factors
212
Company Confidential
WCDMA Link Budget – Cell Count
r = Cell Radius
r
3 2
Surface of a tri-sectorial cell : ACell =3 r
2
Number of Sites = Number of Cells /3
213
Company Confidential
Part IX
Coverage-Capacity
enhancement
214
Company Confidential
Coverage Improvement Alternatives
• Mast head amplifier
– basic solution for optimized uplink performance
– compensates feeder cable loss
– supported by Nokia's base stations
– can be used together with Smart Radio Concept
• 6 sectored site
– utilizing narrowbeam antennas
– ~ 2 dB better antenna gain than in 3
sectored site
215
Company Confidential
Capacity Improvement Alternatives
• 6 sectored site
– ~ 80% capacity gain compared to 3
sectors (not 100% due to inter-sector
interference)
216
Company Confidential
Smart Radio Concept
Uplink coverage
– 4-branch diversity reception per sector dB Received signal power
10
– Maximal ratio baseband combining of 4
uplink signals forms a beam 5
RX
RX
RX + TX
WCDMA Combined
Transceiver received
signal
217
Company Confidential
144 kbps Coverage / Capacity in Macro Cells
Max. allowed
path loss [dB]
170
Downlink
load curve
165
Better 160
coverage Coverage is
uplink limited
155 Capacity is
downlink limited
150 Uplink load
curve with RX
diversity for
144 kbps
145
100 200 300 400 500 600 700 800 900 1000110012001300
Load per sector [kbps]
218
Company Confidential
Nokia Smart Radio Concept
Phase 1: Increase Uplink Coverage
Max. allowed
path loss [dB]
170
165
Uplink 2.5-3.0 dB
load curve coverage
160 with SRC improvement
with SRC
155
160 Downlink
20W
no diversity
155
150
70% increase
in capacity
145
100 200 300 400 500 600 700 800 900 1000110012001300
Load per sector [kbps]
220
Company Confidential
Coverage : 30 % less sites with SRC
Sites / km2
0.3
0.25
2.5 - 3.0 dB gain
0.2 corresponds to 30%
less sites with SRC
0.15
0.1
0.05
0
3-sector (rx div) 3-sector (SRC)
221
Company Confidential
Capacity Upgrade with Smart Radio Concept
• No changes to antennas or antenna cables
• All these capacity upgrades within one Ultrasite cabinet
Morphology analysis
Morphology Percentage Area
Suburban 65.2%
Open/Fields 13.7%
Open in Urban 10.1%
Industrial 6.3%
Roads in Urban 2.2%
Forest 2.0%
Urban 0.5%
Water 0.1%
224
Company Confidential
Radio Network Configuration
51 sites (3 sector)
Parameter Assumptions
Parameter Value
Max. transmit power 43dBm
Max. power per link not limited
Min. transmit power per link not limited
CPICH power 30dBm
Common channel power 30dBm
Cable/connector loss 3dB
Soft handover window 5dB
RF carriers available 1
Slow fading standard dev. 8dB
Maximum uplink load 50%
225
Company Confidential
Antenna Configuration
1G and 2G antenna list 3G antenna list
60° antenna x 56 60° antenna x 56
85° antenna x 97 90° antenna x 97
741415 CS72138
226
Company Confidential
Traffic Modeling
Priority placed on modeling Uniform distribution of mobile terminals
traffic services separately
System loaded to maximum capability
12.2kbps speech
fixed uplink load limit
64kbps data fixed BTS power capability
144kbps data
Link level simulations used to define Eb/No
Symetric data services requirements, SHO Gain
Service
Parameter 12.2 kbps 64 kbps data 144 kbps data
voice MS Numbers
Max. transmit power 21 dBm 21 dBm 21 dBm
Min. transmit power -50 dBm -50 dBm -50 dBm
Distrib. Supp.
Antenna height 1.5 m 1.5 m 1.5 m
Antenna gain 0 dBi 0 dBi 2 dBi 12.2kbps speech 15000 ~5400
Body loss 3 dB 0 dB 0 dB
Uplink bit rate 12.2 kbps 64 kbps 144 kbps
64kbps data 5000 ~1100
Downlink bit rate 12.2 kbps 64 kbps 144 kbps
Uplink activity factor 0.67 1 1
144kbps data 1500 ~500
Downlink activity factor 0.67 1 1
Mobile speed 50 km/hr 3 km/hr 3 km/hr
227
Company Confidential
Benchmark Results
MHA, SRC, 6 Sector not included
Percentage of Cells
Number of Cells
upon
polygon
6
Envir. Service of the Mean
Probe Mobile
Outdoor Speech 99.83
64kbps Data 98.54
144bps Data 96.74
Indoor Speech 88.05 0 0
0 30 60
64kbps Data 70.05
144bps Data 59.71 Number of 12.2kbps speech users
228
Company Confidential
Impact of MHA
MHA introduced at all sites
Improves uplink power budget
Improved
Example
indoor speech: 88 93%
indoor 64kbps data: 70 79%
indoor 144kbps data: 60 71%
Improved
speech by 9%
64kbps data by 11% Improved
144kbps data by 30%
Becomes limited by BTS tx power indoor speech: 88 92%
indoor 64kbps data: 70 77%
indoor 144kbps data: 60 68%
Observations:
Remains uplink limited
Once downlink limited, soft handover window has
great impact upon capacity
Further Improved
231
Company Confidential
Impact of 6 Sectors
2 scenarios
6 Sector introduced at all sites
6 Sector introduced at TACS sites only
33º beam width antennas increased gain
Doubled
Usual message for 6S:
Example, outdoor
capacity improves by
speech, 2500 5100 users less than a factor of 2
64kbps data, 440 950 users due to increased SHO
144kbps data, 230 450 users & inter-cell
interference
Improved
Example,
Not true in this case due to antenna
indoor speech: 85 93% pattern selected:
indoor 64kbps data: 65 80 %
reduced SHO and inter-cell
indoor 144kbps data: 50 70 % 232
interference
Company Confidential
Part X
WCDMA/GSM
Co-siting issues
233
Company Confidential
Antenna System Co-siting
234
Company Confidential
Antennas: WCDMA/GSM Co-site
Antenna 1: Antenna 2:
3 sector site:
Shared antenna lines GSM 900 / 1800 WCDMA X-pol
GSM 900/1800 antennas: 3 pcs
Dual Band X-pol
• GSM 900 / GSM 1800 / WCDMA WCDMA antennas:
WCDMA MHAs:
3 pcs
6 pcs
triplexers Triplexers: 6 pcs
Feederlines: 6 pcs
235
Company Confidential
Upgrades to Current GSM Antennas
Upgrade :
Current : space +
space polarization
diversity diversity
Space
Space diversity
diversity
improves
improves performance
performance
0.5..1.0
0.5..1.0 dB
dB compared
compared
to single radome.
to single radome.
The
The gain
gain of
of 2.5
2.5 dB
dB
assumes
assumes single
single radome.
radome. Antennas can
Current : be shared
1300 mm
DPX
TPX
DPX
Triplexer
Abis/Iub
To/From
BSC/RNC Triplexer
Iub
Power
237
Company Confidential
Nokia Base Stations and Co-Siting
Air-interface issues
238
Company Confidential
Co-Siting with other manufacturers
Air-interface issues
• WCDMA Co-Siting with other manufacturers'
equipment
– theoretical worst case requires 50 dB extra
isolation in GSM BTS
– in practice this much will not be needed
– Nokia can provide assistance with co-siting
issues
• Note: 30 dB Minimum Coupling Loss (MCL)
assumed between antennas
239
Company Confidential
WCDMA - GSM Interference Outline
• Spurious emissions
• Nonlinear distortion
• Specifications and isolation requirements
• Interference mitigation methods
• Co-located sites
240
Company Confidential
WCDMA - GSM Interference Outline
• Site and equipment sharing is an important issue to cut costs down
and to guarantee proper function of the networks.
• Common
• base station mechanics
• site support
• transmission
• antennas and feeders
• site construction
• network management
• By proper site design (antenna installation etc.) interference coupling
between systems can be reduced and unreasonable degradation of
service due to co-sited installations avoided.
• Co-siting preferred to avoid high path loss differences between own
and neighbour systems.
241
Company Confidential
Spurious emissions
242
Company Confidential
Nonlinear system
• Nonlinear system transfer function can be expressed as a series expansion
243
Company Confidential
Nonlinear components
• Nonlinearities of active components like amplifiers under normal
operation.
• Nonlinearities of passive components
– Antennas
– Feeders
– Connectors
• Antenna mismatching
– Reflected wave can cause IMD in the power amplifier.
• Damaged feeders => mismatching
• Loose connectors => mismatching, reflections and rectification.
244
Company Confidential
Active nonlinear distortion
• Active nonlinear distortion is generated in nonlinearities of active components
like amplifiers and modulators
• The nonlinearity effect is especially strong in power amplifiers if they
are driven to saturation.
• Intermodulation levels of the amplifiers can be decreased by
backing-off of them.
3rd order intercept
• The amplitude of the 3rd order product point
increases 3 dB compared to the fundamental
frequencies due to x3 term of it.
Desired signal
• Active IMD generated inside an offender BTS slope = 1
can be removed by BTS TX filtering.
3rd order
IMD
slope = 3
245
Company Confidential
Passive nonlinear distortion
246
Company Confidential
Harmonic distortion
2nd harmonics
• 2nd harmonics
can be filtered
out at the
fGSM = 950 - 960 MHz output of
... GSM900 BTS.
1900 -
247
1920 MHz
Company Confidential
IMD3 from GSM1800 DL to WCDMA UL
• GSM1800 IM3 products are hitting
into the WCDMA FDD UL RX band if • For active elements IM
products levels are higher
• 1862.6 ≤ f2 ≤ 1879.8 MHz than IM products produced
• 1805.2 ≤ f1 ≤ 1839.6 MHz by passive components
• Typical IM3 suppression
fIM3 = 2f2 - f1 values for power amplifiers
are -30 … -50 dBc
depending on frequency
spacing and offset
• Typical values for passive
f1 f2 elements are
-100 … -160 dBc
X dBc fIM3
1710 - 1785 MHz 1805 - 1880 MHz 40 MHz 1920 - 1980 MHz 2110 - 2170 MHz
248
Company Confidential
Nonlinear distortion conclusions
• Second harmonics from the GSM900 system may fall into the WCDMA TDD
band.
• Intermodulation can be a problem if an operator has a splitted GSM1800 band or
in multioperator systems.
• The most harmful intermodulation products are 3rd order products which may fall
into the WCDMA RX band:
fIM3 = 2f1,2 — f2,1
• IM products can be avoided by proper frequency planning in GSM.
• fIM3 is hitting into the WCDMA FDD RX band (1920 - 1980 MHz) if GSM1800
channels are from 512 to 684 (f2) and from 799 to 885 (f1).
• Active intermodulation products can be filtered out in GSM1800 BTS TX
– IM products generated inside a WCDMA receiver cannot be filtered out.
• Passive IM products can not be filtered out in BTS TX if they are generated in
feeder lines and connectors after the filtering unit of BTS.
• Some aging problems may be avoided by installation, site administration and
maintenance recommendations.
249
Company Confidential
RF Specifications
• GSM 05.05-8.7.1, WCDMA TS 25.104-3.5.0
• Two main reasons to isolate GSM and WCDMA
– Blocking
– Sensitivity
250
Company Confidential
Interference mitigation methods
251
Company Confidential
Antenna isolation measurements
252
Company Confidential
Isolation measurements
Antennas and configurations
Horizontal
Antenna beamwidth Gain Polarisation Frequency band
Vert. Pol A 65º 18 dBi Vertically linear 1710 – 1880 MHz
B 90º 16 dBi Vertically linear 1710 – 1880 MHz
C 90º 17.5 dBi Vertically linear 1710 – 1880 MHz
Dual. Pol D 90º 16 dBi +/- 45º dual pol. 1710 – 1880 MHz
120°
d d d d
d
I (90°)
II (120°) III (180°) IV (Horizontal) V (Vertical)
1TSG-RAN Working Group 4 (Radio) Meeting #8
TSGR4#8(99)631
Sophia Antipolis, France 253
26-29 October 1999
Source: Allgon Company Confidential
Antenna isolation measurements
Setup Antenna d [mm] / Min d [mm] / Max
isolation [dB] isolation [dB]
I A 250 / 50 850 / 63
d B 250 / 46 975 / 59
d
C 250 / 54 950 / 62
D, Co-polar 200 / 46 1250 / 59
I (90°) D, Cross-polar 200 / 49 1000 / 58
II A Same mast / 49 1050 / 66
d
120° B Same mast / 38 1100 / 66
C Same mast / 53 1150 / 68
D, Co-polar Same mast / 38 1100 / 65
II (120°) D, Cross-polar Same mast / 43 1050 / 63
254
Company Confidential
Antenna isolation measurements
Setup Antenna d [mm] / Min d [mm] / Max
isolation [dB] isolation [dB]
III A Same mast / 52 750 / 71
B Same mast / 49 1300 / 69
d
C Same mast / 52 1150 / 76
D, Co-polar Same mast / 38 1250 / 62
III (180°) D, Cross-polar Same mast / 53 1250 / 62
IV A 250 / 37 6000 / 57
B 250 / 27 6000 / 52
C 250 / 34 6000 / 48
D, Co-polar 250 / 33 4250 / 53
d
D, Cross-polar 250 / 36 6000 / 57
IV (Horizontal)
255
Company Confidential
Antenna isolation measurements
Setup Antenna d [mm] / Min d [mm] / Max
isolation [dB] isolation [dB]
V A 2250 / 50 6000 / 70
d B 2250 / 55 5500 / 69
C 2250 / 61 6000 / 66
D, Co-polar 1500 / 42 6000 / 61
V (Vertical) D, Cross-polar 1500 / 44 5500 / 65
256
Company Confidential
Antenna isolation measurements
• Measurements performed in a more realistic environment by Nokia.
• The used antennas are listed in the table below
257
Company Confidential
Antenna isolation measurements
• Measured frequencies from 1710 to 1980 MHz and results collected from
1900, 1950 and 1980 MHz.
• Measurement corresponds spurious emissions attenuation from the
GSM1800 band into the WCDMA band.
output input
horizontal
separation
distance
Side View
direction of radiation
1000mm
2000mm
400mm 650mm
70.00
65.00
1900MHz
Isolation (dB)
1950MHz
60.00
1980MHz
55.00
50dB marker
50.00
45.00
40.00
00 00 00 00 00 00 00 00 00 00 ..
0. 1. 2. 3. 4. 5. 6. 7. 8. 9. 1.
Distance (m)
260
Company Confidential
Antenna isolation measurements: Horizontal
60.00
55.00
Isolation (dB)
50.00
50dB marker
1900MHz
45.00
1950MHz
1980MHz
40.00
35.00
30.00
0
00
00
00
00
00
00
00
00
00
00
.0
6.
0.
1.
8.
2.
3.
4.
5.
7.
9.
10
Distance (m)
261
Company Confidential
Antenna isolation measurements: Horizontal
60.00
55.00
Isolation (dB)
50dB marker
50.00 1900MHz
45.00 1950MHz
40.00 1980MHz
35.00
30.00
0
00
00
00
00
00
00
00
00
00
00
.0
.0
.0
0.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10
11
12
Distance (m)
262
Company Confidential
Antenna isolation measurements: Face
Antenna A
GSM1800 Front View
1m 5m
Antenna B
UMTS
Side View
direction of radiation
1000mm
2000mm
300mm
400mm 650mm
direction of radiation
85.00
1900MHz
1950MHz
80.00 1980MHz
75.00
70.00
0.00 1.00 2.00 3.00 4.00 5.00
264
Company Confidential
Antenna isolation measurements: Vertical
Antenna B
UMTS
Antenna A
GSM1800
(fixed)
10m
Noise Floor
80.00
75.00
Isolation (dB)
70.00 1900MHz
1950MHz
65.00
1980MHz
60.00
55.00
50.00
0
25
50
25
50
0
0
0.
0.
1.
1.
0.
0.
1.
Distance (m)
266
Company Confidential
Antenna measurement conclusions
267
Company Confidential
Isolation 800/900 - UMTS
Horizontal Separation : XPol 900 65° _ XPol UMTS
(824-960) (1710-2170)
268
Company Confidential
Isolation 800/900 - UMTS
Horizontal Separation : XPol 900 90° _ XPol UMTS
(824-960) (1710-2170)
269
Company Confidential
Isolation 800/900 - UMTS
Vertical Separation : XPol 900 65° _ XPol UMTS
(824-960) (1710-2170)
270
Company Confidential
Isolation 800/900 - UMTS
Vertical Separation : XPol 900 90° _ XPol UMTS
(824-960) (1710-2170)
271
Company Confidential
Isolation 800/900 - UMTS
Separation by 120° : XPol 900 65° _ XPol UMTS
(824-960) (1710-2170)
272
Company Confidential
Isolation 800/900 - UMTS
Separation by 120° : XPol 900 90° _ XPol UMTS
(824-960) (1710-2170)
273
Company Confidential
Isolation 1800/1900 - UMTS
Horizontal Separation : XPol 1800 65° _ XPol UMTS
(1710-1990) (1710-2170)
274
Company Confidential
Isolation 1800/1900 - UMTS
Horizontal Separation : XPol 1800 90° _ XPol UMTS
(1710-1880) (1710-2170)
275
Company Confidential
Isolation 1800/1900 - UMTS
Vertical Separation : XPol 1800 65° _ XPol UMTS
(1710-1990) (1710-2170)
276
Company Confidential
Isolation 1800/1900 - UMTS
Vertical Separation : XPol 1800 90° _ XPol UMTS
(1710-1880) (1710-2170)
277
Company Confidential
Isolation 1800/1900 - UMTS
Separation by 120° : XPol 1800 65° _ XPol UMTS
(1710-1990) (1710-2170)
278
Company Confidential
Isolation 1800/1900 - UMTS
Separation by 120° : XPol 1800 90° _ XPol UMTS
(1710-1880) (1710-2170)
279
Company Confidential
Isolation Dualband GSM 900/1800 - UMTS
Horizontal Separation : XXPol 900/1800 65°/65° _ XPol UMTS
(870-960/1710-1880) (1710-2170)
280
Company Confidential
Isolation Dualband GSM 900/1800 - UMTS
Vertical Separation : XXPol 900/1800 65°/65° _ XPol UMTS
(870-960/1710-1880) (1710-2170)
281
Company Confidential
Isolation UMTS - UMTS
Horizontal Separation : XPol UMTS 65° _ XPol UMTS
(1710-2170) (1710-2170)
282
Company Confidential
Isolation UMTS - UMTS
Vertical Separation : XPol UMTS 65° _ XPol UMTS
(1710-2170) (1710-2170)
283
Company Confidential
Isolation UMTS - UMTS
Separation by 120°: XPol UMTS 65° _ XPol UMTS
(1710-2170) (1710-2170)
284
Company Confidential
Part XI
WCDMA Optimization
285
Company Confidential
Network Optimization Process
Objective: To optimize the outdoor part of the 3G network, this done cluster wise,
as they are being integrated.
286
Company Confidential
Pre Launch Optimization-Overview
287
Company Confidential
Pre Launch Optimization-Process
288
Company Confidential
Optimization-Overview
289
Company Confidential
Optimization-Overview Block A
290
Company Confidential
Optimization-Overview Block A
291
Company Confidential
Optimization-Overview Block B
292
Company Confidential
Optimization-Overview Block C
293
Company Confidential
Optimisation - required performance
294
Company Confidential
Key Performance Indicators, KPI
• KPIs are a set of selected indicators which are used for measuring the
current network performance and trends.
• KPIs highlight the key factors of network monitoring and warn in time of
potential problems. KPIs are also used to prioritise the corrective actions.
• KPIs can be defined for circuit switched and packet switched traffic
separately and be measured by field measurement systems and Nokia
NetActTM network management system.
• An example set of KPIs
– RRC Setup Complete Ratio
– RAB Setup Complete Ratio
– RAB Active Complete Ratio
– Call Setup Success Ratio
– Call Drop Rate
– Softer/Soft Handover Fail Ratio
295
Company Confidential
WCDMA RAN Optimisation
Network Management WCDMA RAN
• Nokia NetActTM for 3G
• Field Tool Server configuration
KPIs, counters
me
as ur e
Configuration KPIs, me air-interface
measurements
nt s
RAN Optimisation
• pre-defined procedures
• semi / full automated Start
NMS: Collect
network
performance data
No
Evaluate KPI
'HO Overhead'.
OK ?
Field Tool
Yes
Yes
End
296
Company Confidential
WCDMA Field Tool
• Measurement data with • Measurement data with
location and timestamp
location and timestamp
Phase 1 • File & remote IP based
interface
Phase 2
299
Company Confidential
Cluster Preparation
• Cluster Identification
– Site locations, major roads, RNC borders, other geographical
aspects, e.g. rivers
– Interference analysis to minimise the amount of external
interference to a cluster from neighbouring clusters
– MapInfo plot illustrating sites within cluster, cluster borders etc.
– Site information spreadsheet (cluster id, priority etc.)
– Site Integration percentage criteria check
• Drive route planning
– A good percentage of main roads, motorways
– Different clutter types, where applicable
– Areas of special interest, e.g. airport routes, corporate routes
• Drive test equipment check
• Site Verification (crossed feeders etc.)
300
Company Confidential
Cluster Preparation
• Collection of Cluster data
– Dominance plot
– CPICH coverage plot
– Best server Ec/Io plot
– RSCP plot
– SHO plot
– Configuration data (scrambling codes, tilts & bearing, data build)
• Network health check
– Identifying faulty Sites with
• Cell availability check
• Alarm check
– Neighbour list check (Planner vs. configuration management)
– Parameter consistency check (configuration management)
• RNW parameter check
• Transmission parameter check 301
Company Confidential
Cluster Tuning
• RF optimisation (physical change of antenna tilt, azimuth, type and
height) based on scanner data
– Target exit criteria for RF optimisation is based on
• CPICH RSCP (Ec) per each SC
• Best Server CPICH Ec/Io
• Pilot Pollution (high RSCP vs. low EcNo)
• SHO overhead (active set count)
• Neighbour list (scrambling code) verification based on scanner data
– incl. GSM/EDGE neighbours
• Call Performance KPI Verification with logging tool
• AMR call setup success rate (CSSR)
• AMR call drop rate (DCR)
• Call setup time (CST)
302
Company Confidential
Cluster Optimisation
• Maximise call performance KPIs by
– Analysis of Drive test data
– Analysis PM data (counters and KPIs)
• Investigation of bad quality
– lack of coverage
– external interference
– wrong or missing neighbours
• Additionally Golden cluster can be used for solving
– UE-specific problems (hanging on the cell, poor cell reselection,
poor power control)
– UE-NW incompatibilities
– Troubleshooting (call tracing)
303
Company Confidential
Network Verification
• Used to accept the network performance based on drive tests along
”reference routes” in the area of multiple clusters
• KPIs are measured in areas of ”acceptable” coverage is achieved
304
Company Confidential
RF & Call Performance Targets
• Minimum RF conditions for field measurements for all KPIs
(examples)
• Best server CPICH RSCP > -100 dBm
• Best server CPICH EcNo > -11 dBm
• CPICH EcNo of the 4th strongest cell > 6dB below best server
• Call Performance target examples for AMR
– Call setup success rate, 98 %
– Call drop rate, 2 %
– Overall call success rate, 96 %
– Call connection time <8 s, 98 %
305
Company Confidential
Field Measurement Tools
• Field Measurement tools
– Scanner could be used for Coverage and Scrambling code
analysis
– Logging tools are available from a many manufacturers:
– Nemo Technologies – Layer 1 & 3 trace capability
– SwissQual - Layer 1 & 3 trace capability + subjective voice
measurement (MOS)
306
Company Confidential
Dominance Verification
307
Company Confidential
RSCP Verification
308
Company Confidential
EcIo Verification
309
Company Confidential
Pilot Pollution Verification- example
310
Company Confidential
Pilot Pollution Analysis I
• Pilot pollution areas can be shown with Scanner Pilot Pollution query with
user defined RSCP and EcNo levels
(CPICH_SCAN_EcNo_Sorted_By_EcNo (0)<-12)AND
(CPICH_SCAN_RSCP_Sorted_By_EcNo (0)>-92)
311
Company Confidential
Pilot Pollution Analysis II
• Pilot polluter workbook shows the worst polluters in the area e.g. SC
344 have most of the samples
SC 344 displayed
over pilot pollution
312
Company Confidential
Pilot Pollution Example
Scanner sees 5
SC’s, all within 5 dB
of each other.
This is clearly an
area suffering from
pilot pollution.
SC 272
314
Company Confidential
UL Coverage Verification
315
Company Confidential
Throughput Verification
316
Company Confidential
RF Optimization based on Scanner data
Coverage Data collection
A Possible Actions/Solutions
target agreed 1. Antenna Tilting
with customer 2. Antenna Panning
3. Change Antenna Type
No 4. Change Antenna Height
CPICH Ec > Threshold 5. Change CPICH Tx Pwr
e.g. –100 dBm
Currently X =
6. Add sites…
Low Max AS size = 3
CPICH Ec Yes
for Nokia RAN
A
No amount of Scrambling Yes
CPICH Ec/Io > Threshold
e.g. –11 dB Codes > X Pilot
Ec/No target agreed Pollution
with customer &
Yes
depends service mix No
No A
Dominance Area OK
Aggregated to Peak
> 3 dB Yes Multi-path
Yes Problem
A Bad No
Ec/Io
A
A
317
Company Confidential
RF Optimization
Tuning methods for Tuning methods for
Coverage Problem Area High
HighPriority
Priority Dominance Problem Area
318
Company Confidential
Physical Optimisation - Antenna changes
• Antenna tilting (or panning) is needed mainly if:
– There is too much interference created by a site covering too much
(overshooting)
– There is lack of coverage or dominance
• One should carefully consider the tilt type
– Electrical or mechanical (both have advantages and disadvantages)
• Antenna tilting should be followed by another round of drive-tests in order
to evaluate the impact
• Multiple antenna tilt or azimuth changes in the same time in the same area
should be avoided
Mechanical
Down tilt kit
Deformation of the
horizontal pattern
321
Company Confidential
Electrical Tilt
• The Adjustable EDT antennas can be adjusted manually or remotely
• Phase shifters provides variable phase distribution which in turn keeps the
pattern shape constant
• Maximum Adjustable EDT range approx. 0-14° (normally 0-8°)
• For a higher downtilt angle a combination of the Mechanical DT and the
Adjustable EDT is recommended
Manual Remote
use use
Horizontal pattern
remains constant
322
Company Confidential
Antenna tilt example (6 degrees downtilt)
Impact
Impacton
onEc/No
Ec/No
323
Company Confidential
Antenna tilt example (4 degrees downtilt)
Impact
Impacton
onEc/No
Ec/No
324
Company Confidential
Antenna tilt example -1 deg E-tilt – Overshooting Cell
No dominant
Server in the area
due the missing
site JS9218 Before After
Before After
326
Company Confidential
New Site Integration
• New sites should be integrated without degrading the performance of existing
sites.
• It is possible that after driving the new and surrounding site, both will require
RF optimisation (tilting, panning)
327
Company Confidential
New Site Integration
Different strategies can be used This option does not give us
• Integrate new sites on different frequency right information about
– Define intra-site neighbours network situation (Ec/No etc)
– Drive new site and analyse
– Apply necessary tilts to new and surrounding sites
– Add all Missing Neighbours
=> Turn Site on with correct frequency
• Integrate new site on SAME frequency during low traffic period
(Nigh Time)
This option require
• Define neighbours
possibilities to use drive test
• Turn site “on” during nigh time only team during night time (cost
• Drive new site and analyse issue)
• Apply necessary tilts to new and surrounding sites
• Add all Missing neighbours
• => Turn Site on
328
Company Confidential
Neighbour list verification
• The neighbour list can be verified by comparing
– Best server Ec/Io results (from Scanner) with existing neighbours
e.q. with UE logging
– Best server Ec/Io results to certain Handover margin (Event 1a
criteria) based on certain number of samples
• Analysis can be done in Actix analyzer
329
Company Confidential
Neighbour List Verification
• Neighbour definitions required by cell re-selection and handover
• Soft handovers are based upon intra-frequency neighbour list
• Hard handovers are based upon either intra-frequency (Between RNCs without Iur
or Iur congestion) , inter-frequency (IFHO) or inter-system (ISHO) neighbour lists
Max. 32
Max. 31
Max. 48
… and “realistic worst case values”, SIB11 length = 3187.5 < 3552 -> OK!!
• Some sites might need additional neighbors and might pose a problem with the
SIB11 limitation
– Avoid setting AdjsQoffset2 values, different CPICH values or other parameters
used to tune cell reselection or handover
Further information Technical Note No. 046 / Restriction on number of
cells in SIB11/12 due to inconsistency problem in 3GPP TS 25.331 331
Company Confidential
Neighbour List Verification – SIB 11/12
• When offsets are added to the neighbours or the CPICH power of the
neighbour cell is different to that of the serving cell the length of the
message for each neighbour in SIB 11/12 is increased
– The length of one ADJS with no offsets is 48 bits
– The length of an ADJS with AdjsQoffset1 or AdjsQoffset2 is 48 or 56 bits
on a case by case basis (average length 55.2 bits)
– The length of an ADJS with AdjsQoffset1 and AdjsQoffset2 is 56 or 64
bits on a case by case basis (average length 62.1 bits)
• When the neighbour cell has a different CPICH Tx Power from the
serving cell
– Without Offset and AdjsPtxCPICH : 54.2bits
– With AdjsQOffset1 or AdjsQOffset2 : 61.1bits
– With both AdjsQOffset1 and AdjsQOffset2 : 68bits
332
Company Confidential
Neighbour List Verification
• There are a number of approaches that can be used to both plan and
verify the neighbour plan
Other
Manual Check
Drive Testing
Measured
Network Stats
333
Company Confidential
Neighbour List Verification - Analytical
334
Company Confidential
Neighbour List Verification - Manual
• 6 missing neighbours
were identified for this
site and added – Manual
Check would have
identified at least 5
335
Company Confidential
Neighbour List Verification – Drive Test
• Process starts directly with the drive test
High Level
Process
Start Record drive test
results with Scanner in
‘TOP N’ mode
336
Company Confidential
Neighbour List Verification
• Part automatic and part manual neighbour list analysis
• Analysis within a tool such as Actix Analyzer is automatic and numerical
• Analysis requires the definition of a neighbour window which is applied to the
CPICH Ec/Io measurements
• Recommended to set the neighbour window as 10 dB (drop window + margin)
Strongest Neighbours
Ec/Io High Level Process
Ec/Io reported when
within this range Record drive test
Start
results with Scanner in
‘TOP N’ mode
Neighbou
r Window
Analysis (within
Actix Analyzer) Finish
Time
SC Site Sector Num Of Samples Suggested NBR Additions
216 Ewer Park Stud 00161126C 97 85 (47.4%); 303 (17.5%)
349 Reading Rubgy FC 00179719B2 97 351 (28.9%)
83 WOODLEY ATE 00001085A 96 350 (34.4%); 84 (25.0%)
436 TILEHURST ATE 00001081B 90 434 (22.2%); 283 (22.2%)
337
Numerical Analysis
Company Confidential
Neighbour List Verification
338
Company Confidential
Neighbour List Verification
Example for Reading Central (SC 280)
% Scrambling Code Samples
25.00
% of Samples
• Bar chart 20.00
33.90
• List 15.00
18.50
10.00
13.80
9.40
5.00
3.90
3.10
2.00
1.20
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
422 423 420 283 282 286 348 177
Neighbour Scram bling Code
Does
Does list
list include
include all
all
NO
cells belonging
cells belonging to to
Cell for which that
that site
site and
and the
the
the neighbour first
first perimeter of
perimeter of
list is being cells
cells
defined
Yes
Add
Add any
any other
other
neighbours
neighbours which
which
Neighbours may
may improve
improve
coverage
coverage
Finish
340
Company Confidential
Neighbour List Verification – Co Scrambling Code
Problem
• An important part of neighbour list verification (against the assigned
scrambling codes) is to ensure that the neighbour lists do not pose
any scrambling code clashes.
• A possible impact is that the network attempts to set up a link on a
different cell (scrambling code) to that measured by the UE, or due to
combining of the neighbour lists by the RNC duplicate scrambling
codes may be introduced, and the call may drop.
341
Company Confidential
Neighbour List Verification – Co Scrambling Code
• Scenario 1 Problem
– In some cases, the neighbour definition is created to a cell (Cell A), which is far from
the source cell, and having the same scrambling code as Cell C, which is very close
and is not defined as a neighbour.
– This problem is caused by errors in mapping the Primary Scrambling Code Picked by
the scanner to a certain cell ID. The target Primary Scrambling code (123 in the
example) was picked by the scanner, but was associated with a wrong CID, Cell A
instead of Cell C.
343
Company Confidential
Neighbour List Verification – Co Scrambling Code
Problem
• Scenario 3 (three-way SHO)
– In this scenario Cell B and Cell D are not defined as neighbours. However, they are
both in the active set due to ‘three-way’ soft handover between Cell B, C and D, and
thus the combination of the neighbour lists will happen in the RNC.
344
Company Confidential
Part XII
Radio Resource Management
345
Company Confidential
Radio Resource Management
346
Company Confidential
Radio Resource Management
347
Company Confidential
Radio Resource Management
348
Company Confidential
Radio Resource Management
349
Company Confidential
RRM Control Processes
350
Company Confidential
WCDMA Radio Resource Management:
Logical Model
• RRM is made up of a number of closely interdependent functions (i.e.
algorithms)
• These functions can be divided into;
LC
• Cell Based PS
– Load control (LC)
RM
– Admission control (AC) AC
– Packet scheduling (PS)
– High Speed Downlink Packet Access Cell based functions
– Resource manager (RM)
PC
• Connection Based
– Handover control (HC) HC
– Power control (PC)
• Admission control
– Performs the admission control for new bearers to enter/leave
the network.
– Predicts the interference caused by the bearer and checks
whether there is room for it.
– Power allocation
• Packet Scheduler
– Scheduling packets to the radio interface (UL/DL)
• Load Control:
– Takes care of radio network stability
– Gathers interference information and produces a load vector
• Resource manager
– Manages the physical resources of RAN and maintains the
code allocation
352
Company Confidential
RRM control processes
• Power Control
– Closed loop PC compares the measured SIR with SIR-target
and accordingly transmits an up/down PC command at 0.667
ms interval
– Open loop PC estimates the needed power based on pathloss
+ interference measurements (RACH).
– Outer loop PC sets the SIR target for the fast closed loop PC
• Handover Control
– Soft (intra-frequency) handovers: softer between cells within
one BS, intra-RNC soft, inter-RNC soft
– Inter-frequency (hard) handovers: Intra-BS, Intra-RNC, Inter-
RNC (-MSC)
– Inter-RAT handovers: WCDMA <-> GSM
353
Company Confidential
Power Control
Power Control loops in WCDMA
MS BTS RNC
354
Company Confidential
Power Control Loops
• Effective power control is essential in WCDMA due to frequency re-
use factor of one (in ideal case)
• Closed loop e.q. Fast power control
– Makes Eb/No requirements lower
– Equalizes received powers at BTS in uplink (avoids near-far effect)
– Introduces interference peaks in the transmission
• Open loop power control for initial power setting of the UE
• Outer PC loop at a slower rate, across the Iub interface in uplink
– At a much slower rate, across the Iub interface in uplink
– Adjusts the SIR target to achieve a target BLER
– Also similar outer loop power control in MS
– There is also similar outer loop power control in UE
355
Company Confidential
Power Control & Diversity
356
Company Confidential
Admission Control & Packet Scheduler
357
Company Confidential
Admission Control & Packet Scheduler
358
Company Confidential
Admission Control
Uplink admission control
• In uplink the total received wideband interference power measured
indicates the traffic load of the radio resources .
• The fundamental criteria of evaluation is based on
359
load
Company Confidential
Admission Control
Uplink admission control
UL interference power
Prx_target_BS
Marginal load area Prx_offset
Prx_target
TRHO_threshold
planned uplink
interference power
Planned load area
Load
Prx_target defines the optimal operating point of the cell interference power, up to
which the AC of the RNC can operate.
360
Company Confidential
Admission Control
Downlink admission control
DL transmission power
Ptx_target_BS
Marginal load area Ptx_offset
Ptx_target
TRHO_threshold
Load
Downlink power increase estimation is done for non-controllable load just like UL
power increase.
361
Company Confidential
Admission Control Functional Overview
362
Company Confidential
Admission Control Functional Overview
RRC Connection
Establishment
0
RAB AC
admitted
RT CS call Radio Access Bearer
Admission Service Request
Decision Quality
1
4 Requirements of
Power Increase
Estimates 2 Radio Bearer RAB attributes (HLR);
3 • SDU error ratio
RAB request • traffic class
denied • max bit rate
Outer
Outer loop
loop PC
PC RB
RB configuration
configuration
configuration
configuration (RNC);
(RNC); (RNC);
(RNC);
•• target
target BLER
BLER •• RLC
RLC mode
mode
•• target
target Eb/No
Eb/No •• TrCH
TrCH parameters
parameters
•• initial
initial SIRtarget
SIRtarget
363
Company Confidential
Admission control for NRT RAB
364
Company Confidential
UL Cell Load Admission Criteria Thresholds
If the equation is NOT satisfied then the RAB can NOT be admitted!
• If there is queuing time left for the RAB (parameter AcRabQueueTime) then
the RAB remains in the queue to wait for the next PrxTotal measurement,
otherwise it is removed from the queue
365
Company Confidential
Parameters
• When the non-controllable part of the cell load (Prx_NC) exceeds the target
limit, AC will reject those RAB establishment requests that would mean
immediate UL load increase (i.e. RT services)
• The packet scheduler (PS) is responsible on handling the NRT bearer
– When load is too high bit rates are decreased etc.
Prx_Target_BS
Prx_Target
Controllable load
Controllable load
∆Prx_NC ∆Prx_NC
Prx_NC Prx_NC
Case I Case II
Prx_NC ++ ∆Prx_NC
Prx_NC ∆Prx_NC ≤≤ Prx_Target
Prx_Target Prx_NC ++ ∆Prx_NC
Prx_NC ∆Prx_NC ≤≤ Prx_Target
Prx_Target
RT RAB admitted (as well as NRT) RT RAB admitted (as well as NRT)
Priority based scheduling by PS 367
Company Confidential
UL Admission Decisions
Prx_Target_BS
Controllable load
∆Prx_NC
Prx_Target
Controllable
load
∆Prx_NC
Prx_NC
Prx_NC
Prx_NC ++ ∆Prx_NC
Prx_NC ∆Prx_NC ≤≤ Prx_Target
Prx_Target Prx_NC ++ ∆Prx_NC
Prx_NC ∆Prx_NC >> Prx_Target
Prx_Target
• If the equation is NOT satisfied then the RAB can NOT be admitted!
369
Company Confidential
DL Admission Decision
Ptx_Target_BS
Ptx_Target
Controllable load
Controllable load
∆Ptx_NC ∆Ptx_NC
Ptx_NC Ptx_NC
Case I Case II
RT RAB admitted (as well as NRT) RT RAB admitted (as well as NRT)
370
Company Confidential
DL Admission Decision
∆Ptx_NC
Ptx_NC
Ptx_NC
RT RAB admitted (as well as NRT) RT RAB denied (NRT RAB permitted)
Overload actions started by PS RT over NRT pre-emption 371
Company Confidential
Packet Scheduler
HSDPA
reservation
(when HSDPA
preferred over
NRT DCH)
Non-Controllable Load
• real time traffic
• interference from other cell users
• noise
time 373
Company Confidential
Why Packet Scheduling?
It is characteristic for RT traffic that it’s load cannot be controlled in efficient way. Load caused
by RT traffic, interference from other cell users and noise together is called non-
controllable load. The available capacity, which is not used by non-controllable load, can
be used for NRT radio bearers on best effort basis. To fill the whole load budget and
achieve the maximum capacity, the allocation of NRT traffic needs to be fast.
The Packet scheduler is a general feature, which takes care of scheduling radio resources for
NRT radio bearers for both uplink and downlink directions; Packet scheduling happens
periodically (with the period of tens of milliseconds) and is implemented for both dedicated
(DCH) and common control transport channels (RACH/FACH).
Scheduled capacity depends on the UE capabilities, Node B capabilities, current load of the
cell as well as the availability of the physical radio resources.
Packet scheduler and MAC layer together make the decision of the used channel type for
downlink direction, data transmission on dedicated channel is initiated when MAC layer
requests transmission capacity
For uplink direction the decision of the used channel type is based on UE measurements
and parameters controlled by network. Data transmission on dedicated channels is initiated
when a capacity request is received from UE.
The selection of the channel type is done fast - taking into account the data amount in the
buffers and the current radio conditions
374
Company Confidential
Packet Scheduler
• The packet scheduler takes care of scheduling radio resources for non-real
time radio bearers for both the uplink and the downlink directions
• The packet scheduler and the medium access control (MAC) layer together
make the decision regarding which type of channel to use in the downlink
direction
375
Company Confidential
Packet Scheduler as part of RRM
• The packet scheduler (PS) co-operates with other radio resource management
functions like handover control (HC), load control (LC), admission control (AC) and
the resource manager (RM)
– HC provides active set information
– LC provides periodical load information to PS on a cell basis
– PS informs AC and LC of the load caused by non-real time radio bearers
– AC informs PS when new non-real time radio bearers are admitted,
reconfigured or released
– RM allocates the RNC internal resources, downlink spreading codes and takes
care of allocating radio links using the base station application protocol (NBAP)
– RM also takes care of transport resource reservation for non-real time radio
bearers using transport resource manager (TRM) services. RM actions are
done when requested by the PS
– The radio resource control (RRC) protocol takes care of L3 signalling between
the RNC and the UE
• The L3 RRC signalling needed by the PS includes uplink capacity
requests and channel allocation procedures in both directions (uplink and
downlink)
– The medium access control (MAC) protocol produces radio bearer-specific
downlink capacity requests to the PS according to the radio link control (RLC)
buffer levels in the RNC
– MAC also sends activity and inactivity indications on an RB basis. 376
Company Confidential
Packet Scheduler functions
2-3
2-3 NRT
NRT PS
PS data
data
MultiRAB: AMR + 1-3 PS data AMR
AMR 12.2
12.2 ++ 1-3
1-3 NRT
NRT PS
PS data
data
MultiRAB: CS data + PS data Transparent
Transparent CS
CS data
data 64
64 ++ NRT
NRT PS data 88 1)
PS data 1)
Non-transparent
Non-transparent CS
CS data
data 57.6
57.6 ++ NRT
NRT PS
PS data
data
Non-transparent
Non-transparent CS
CS data
data 14.4
14.4 ++ NRT
NRT PS
PS data
data
Streaming
Streaming PS PS data
data ++ NRT
NRT PS data 88 1)
PS data 1)
AMR
AMR 12.2
12.2 ++ streaming
streaming PSPS data
data ++ NRT
NRT PS
PS
data 88 1)
data 1)
AMR
AMR ++ NRT
NRT PS
PS
data
data on
on HS-DSCH
HS-DSCH
1) The total aggregate user bit rate of the RRC connection must not exceed 384
kbit/s (AMR+PS384 not possible)
(The restriction does not apply to HSDPA.) 379
Company Confidential
Bit rate upgrading
• The dedicated channel of a non-real time (NRT) radio access bearer (RAB)
can be upgraded due to
1) High amount of data in buffer – Capacity request/ Bit rate upgrade
2) High utilisation/throughput – Flexible upgrade of the NRT DCH data rate
• It is possible to upgrade the NRT DCH data rate from any bit rate below the
maximum allowed bit rate to the maximum allowed bit rate (Flexible upgrade
feature)
380
Company Confidential
Bit rate downgrading
• The dedicated channel of a non-real time (NRT) radio access bearer (RAB)
can be downgraded or released due to multiple causes
1) Excessive downlink power – Dynamic link optimisation (Dylo) for
non-real time traffic coverage feature
2) Different congestion situations – Enhanced priority-based scheduling
(EPBS) and overload control
3) Low utilisation/throughput – Throughput-based optimisation of the
packet scheduler
4) Maximum bit rate limitation – Another RAB is setup for the same UE
5) Inactivity of the radio bearer
381
Company Confidential
Packet Scheduler actions during call – Unloaded cell
AC - AC makes admission decision and allocates a 0 bit rate to the NRT radio bearer
PS1 - “Bit rate allocation”, after receiving a capacity request PS allocates initial bit rate
PS2 - “Flexible upgrade”, After receiving a capacity request PS allocates maximum (high) bit
rate
PS3 - “Throughput-based optimisation“, PS performs radio bearer reconfiguration to lower bit
rate
PS4 - “Flexible upgrade”, PS performs radio bearer reconfiguration to higher bit rate
PS5 - “RRC state transition”, PS initiates state transition to CELL_FACH due to inactivity
PS1
Initial bit rate
Actual throughput
AC
0 bit rate
382
Company Confidential
Packet Scheduler actions during call – Loaded cell
AC - AC makes admission decision and allocates a 0 bit rate to the NRT radio bearer
PS1 - “Priority based scheduling”, after receiving a capacity request PS allocates initial bit rate
after downgrading an existing bearer (load margin)
PS2 - “Flexible upgrade”, After receiving a capacity request PS allocates higher bit rate (normal
load)
PS4 - “Flexible upgrade”, PS performs radio bearer reconfiguration to higher bit rate (normal
load)
PS5 - “RRC state transition”, PS initiates state transition to CELL_FACH due to inactivity
383
Company Confidential
Packet Scheduler actions during call – Loaded cell
Overload
Load Margin
Normal load
PS4 PS5
Allocated bit rate
Max. bit rate
PS2 PS3
PS1
Initial bit rate
AC
Minimum bit rate
Actual throughput
Inactivity Timer
Overload Traffic volume
RACH load
Cell selection
Cell re-selection Idle Mode
385
Company Confidential
Packet Scheduler
Load Decrease Example
386
Company Confidential
Packet Scheduling Principle
387
Company Confidential
Load Control
Capacity
• The traffic can be divided into two groups
– Real Time (RT)
– Non-Real Time (NRT)
• THUS some portion of capacity must be reserved for the RT traffic
for mobility purposes all the time. The proportion between RT and
NRT traffic varies all the time.
Overload area
Overload Margin
Load Target Estimated capacity for
NRT traffic.
Power
Time 388
Company Confidential
Load Control
Definition of Non-controllable traffic
• Since it is not enough to divide the load to RT and NRT one must take
into account the interference coming from surrounding cells.
Traffic is divided into controllable and non-controllable traffic.
389
Company Confidential
Logical description of load control
Load change
info
AC
Load status
PS
LC NRT load
390
Company Confidential
Handover Control
391
Company Confidential
Handover Control (HC)
• HC is responsible for
• Managing the mobility aspects of an RRC connection as UE move around network
• Maintaining high capacity by ensuring UE is always served by strongest cell
Soft(er) Intra-layer
Intra-frequency Inter-layer
Handover
Intra-System
WCDMA to WCDMA
Inter-frequency Intra-layer
Hard Handover Inter-layer
Intra-frequency Intra-layer
Inter-layer
Requires
Compressed Mode
or Dual Receiver
Inter-System UE
(Inter-RAT)
WCDMA to GSM
WCDMA to GPRS
GSM to WCDMA Inter-frequency Intra-layer
GPRS to WCDMA Hard Handover Inter-layer
392
Company Confidential
Softer Handover
• Handover between cells
within a BS
• softer handover is handled
by BS internally
Sector/Antenna
RAKE • softer handover probability
combining about 5 - 15 %
(MRC) • no extra transmissions
across Iub
• basically same RAKE MRC
processing as for
multipath/antenna diversity
(BS / MS). More RAKE
fingers needed.
• provides additional diversity
gain
• softer handover does create
RNC additional interference and
needs BS PA resources
393
Company Confidential
Soft handover
397
Company Confidential
1A: A Primary CPICH Enters the Reporting Range
1 • Reporting range defined by offset (point 1 on next slide) from either
(i) best CPICH Ec/Io in the AS, or
(ii) sum of the AS measurement results
set using ActiveSetWeightingCoefficient (FMCS) (0 = use best CPICH : <>0 = use
sum)
• Offset defined using AdditionWindow (FMCS)
2 • Event 1A triggered when CPCIH3 Ec/Io enters UE reporting range (point 2)
3 • If CPICH3 Ec/Io remains within reporting range for a time period defined by
AdditionTime (FMCS) , and the AS is not full, UE sends measurement report to RNC
which adds Ncell3 to AS if possible (point 3)
• AdditionTime defines the 'time-to-trigger' interval between the Ncell first entering the
reporting range and the UE sending the measurement report to the RNC
4 • If RNC is unable to add Ncell to AS, UE will wait for a period of time, defined by
AdditionReportingInterval (FMCS) (point 4) after the first measurement report,
before sending further reports periodically, with interval AdditionReportingInterval,
until (a) Ncell moves out of reporting range, or (b) RNC adds Ncell to AS.
398
Company Confidential
1A: A Primary CPICH Enters the Reporting Range
Ec/Io ActiveSetWeightingCoefficient
Strongest CPICH in AS P CPICH 1
P CPICH 2 AdditionWindow
1
P CPICH 3
time
3 AdditionTime
Measurement AdditionReportingInterval 4
no
Report
Add to
the AS?
399
RNC
Company Confidential
1B: A Primary CPICH Leaves the Reporting Range
1 • Event 1B triggered when CPICH3 EC/I0 drops out of the UE reporting range for a
defined period of time (point 1 on next slide)
• RNC drops the cell from the AS and UE drops the cell from the AS to the
Neighbour Set
400
Company Confidential
1B: A Primary CPICH leaves the Reporting Range
DropWindow
P CPICH 2 2
P CPICH 3
time
DropTime
Remove the
Measurement reported cell
Report from the AS
401
Company Confidential
1C: A non-active CPICH becomes better than an
active primary CPICH
• UE AS full (MaxActiveSetSize = 3)
1 • Event 1C triggered when CPICH4 EC/IO > CPICH3 in AS by a defined margin,
ReplacementWindow (FMCS). (point 1 in next slide)
• Sending of measurement report to RNC can be delayed by using 'time-to-trigger'
period set by ReplacementTime (FMCS)
2 • If 'time-to-trigger' is used difference between CPICH4 and CPICH3 must be ≥
ReplacementWindow for the period of ReplacementTime (point 2)
3 • If the RNC is not able to replace the Ncells, the UE continues to send
measurement reports periodically, with interval ReplacementReportingInterval,
to the RNC until (a) CPICH4 falls out of ReplacementWindow or (b) RNC
replaces CPICH4 by CPICH3 (point 3)
402
Company Confidential
1C: A non-active CPICH becomes better than an
active primary CPICH
Ec/Io AS has 3 cells
P CPICH 1
P CPICH 2
P CPICH 4
1
ReplacementWindow
P CPICH 3
weakest CPICH3 in AS
time
2 3
ReplacementTime ReplacementReportingInterval
no
Measurement AS
Report update?
403
Company Confidential RNC
Individual Ncell Offset
404
Company Confidential
Individual Ncell Offset
Ec/Io
Reporting
P CPICH 1 Range
P CPICH 2
AdjsEcNoOffset
Enlarging Cell 3 by x dB
P CPICH 3
time
Reporting Reporting
Event 1B Event 1A
405
Company Confidential
IFHO/ISHO Process Overview
HO
HOTriggering
TriggeringThresholds
Thresholdsset
setininRNC
RNC About 25 HO parameters
Event
EventTriggered
TriggeredCoverage/Capacity
Coverage/Capacity 5 Coverage/Capacity HO Reasons
based
based HO fulfilledininRNC
HO fulfilled RNC
(31 Intra-Freq neighbours)
RNC
RNCcommands
commandsselected
selectedUE(s)
UE(s)totostart
start 48 Inter-Freq neighbours
IF/IS measurements
IF/IS measurements 32 Inter-System neighbours can be measured
Measurements
Measurementsare
aredone
doneinin
Compressed
CompressedMode
Mode(CM)
(CM)
UE
UEreports
reportsGSM
GSMcells
cellswith
with Max 32 neighbours could be measured
strongest
strongest RSSI signals toRNC
RSSI signals to RNC
RSSI
RSSImeasurements
measurementsandandBSIC
BSIC Reporting cells are active set cells
verification for GSM cells
verification for GSM cells (max 3) + max 6 IFHO, max 6 ISHO neighb.
RNC
RNCmakes
makesHO
HOdecision
decisionand
and Different decision methods for IF HO
commands UE to target cell
commands UE to target cell Only one decision method for IS HO
406
Company Confidential
IFHO/ISHO measurements- difference
IF-HO
• IF measurements Measurement Target Cell found
Trigger
WCDMA IF -measurements
IF-HO
Decision
• IS measurements
IS-HO
Measurement IS-HO
Trigger Target Cell found
Decision
407
Company Confidential
IF/IS Handover Measurement Triggering Reasons
1. Low measured absolute 2 . Low measured absolute
CPICH Ec/No, event 1E/1F CPICH RSCP, events 1E/1F
FMCI: IFHOcauseCPICHEcNo FMCI: IFHOcauseCPICHrscp,
FMCG: GSMcauseCPICHEcNo FMCG: GSMcauseCPICHrscp
3. UE Tx power approaches
its maximum allowed power, 4. DL DPCH approaches its
Measurement maximum allowed power
event 6A/6D Trigger
FMCI: IFHOcauseTxPwrUL FMCI: IFHOcauseTxPwrDL
FMCG: GSMcauseTxPwrUL FMCG: GSMcauseTxPwrDL
Cell 1 Cell 2
Cell 3
time
1E: HHoEcNoCancelTime 1F: HHoEcNoTimeHysteresis
409
Company Confidential
1. Measurement trigger CPICH Ec/No
• RNC starts IF/IS measurement when event 1F occurs for all cells in the active
set: A Primary CPICH becomes less than an absolute threshold
• RNC stops IF/IS measurement when event 1E occurs for at least one cell of
the active set : A Primary CPICH becomes better than an absolute threshold
• Note:IF/IS measurements can be stopped if event 1Fs are cancelled by
events 1E only when IFHO/ISHO was not successful and only inside the
time between CM measurements, specified by the
time InterFreqMinMeasInterval GsmMinMeasInterval/default 10s,
recommendation 2s .
• Filtering applied before event evaluation in the UE:
– FMCS: EcNoFilterCoefficient/0= 200ms filtering period
410
Company Confidential
2. Measurement trigger CPICH RSCP
411
Company Confidential
Reporting Events 6A, 6B and 6D
6A: The UE Tx power exceeds an absolute threshold
6B: The UE Tx power falls below an absolute threshold
6D: The UE Tx power reaches its maximum value (Not used in Nokia RAN)
UE Tx Power
6D
min(UEtxPowerMaxDPCH, P_MAX)
UE Transmitted Power
Tx Threshold
6A
6B
time
InterfreqUETxPwrTimeHyst time to
GSMUETxPwrTimeHyst trigger
412
Company Confidential
3. Measurement trigger UE Tx Pw (UL Coverage)
• Event 6A parameters:
– Time-to-trigger: 0s
– UE Transmitted Power threshold = min (UEtxPowerMaxDPCH, Pmax) –
UE_TX_POWER_threshold
– UEtxPowerMaxDPCH is the maximum allowed UL tx power on DPCH in the
active cell (range -50..24 dBm, step 1, default 21 dBm), Pmax is the maximum
RF output power
– UE_TX_POWER_Threshold depends on the type of service:
• FMCI: InterfreqUETxPwrThrAMR/-3 dB (range -10..0, step1)
• FMCI: InterfreqUETxPwrThrCS/-3 dB (same range and step)
• FMCI: InterfreqUETxPwrThrNrtPS/-3 dB
• FMCI: InterfreqUETxPwrThrRtPS/-3 dB
• FMCG: GSMUETxPwrThrAMR/CS/NrtPS/RtPS -1/-3/-1/-3/ dB
• Event 6B parameters: time-to-trigger
– FMCI: InterfreqUETxPwrTimeHyst/1280 ms
– FMCG: GSMUETxPwrTimeHyst/1280 ms
• Event 6D parameter: time-to-trigger: 0s 413
Company Confidential
4. Measurement trigger DL DPCH (DL Coverage)
• RNC orders UE to make IF/IS measurements when DL TX power of a single radio link
reaches threshold defined as:
RNC
default 5s time
ULQualDetRepThreshold (recommendation 2 s)
415
Company Confidential
ISHO measurements
416
Company Confidential
ISHO: BSIC Verification
• After the selection of the target GSM cell, the RNC sends to UE the RRC
message "Measurement control” which includes details to measure ”BSIC” .
• UE stops RSSI measurements and updates the transmission gap pattern to
the pattern used for BSIC decoding. The measurement reports are sent
periodically to RNC.
• If the UE is unable to decode the BSIC during the given period, the BSIC
measurement operation is aborted. IS-HO
IS-HO trigger Target Cell found command
• Gaps can be created using single or double frame approach (Nokia supports both)
• Because same data amount is sent in a shorter time more power is needed during
CM (both in UE and BTS)
=> affects WCDMA coverage
• Fast Power control information might be lost during the gap => higher Eb/No
=> affects WCDMA capacity
• Compressed frames may be lost if power control is not set correctly
=> affects WCDMA quality
• CM methods are Spreading Factor Halving and Higher Layer Scheduling
418
Company Confidential
Measurement Gap
Length of gap:
3, 4, 5, 7 timeslots
Single-frame method
TGL
TG
L
• Pathloss Criterion in case trigger was UE Tx Power, CPICH RSCP or UL DPCH Quality:
AVE_EcNo_NCELL (n) > AdjiMinEcNo(n)
EcNo on new cell better than required EcNo in new cell (default –14 dB)
(continued)
420
Company Confidential
ISHO Decision
The following conditions have to be satisfied, before an inter-system Handover can be conducted.
The best neighboring cell must fulfil following criterion („best“ according to Ec/No):
AVE_RXLEV_NCell(n) > AdjgRxLevMinHO(n) + max(0, AdjgTxPwrMaxTCH(n) - P_max)
where
• AVE_RXLEV _Ncell(n) is the averaged GSM carrier RSSI value of the GSM neighbor
cell(n), averaging done directly from dBm values (no linear averaging);
GSMMeasAveWindow / 6, 1…32 meas report.
• AdjgRxLevMinHO(n) determines the minimum Required RSSI (dBm) level of the
neighbor cell(n).
• AdjgTxPwrMaxTCH(n) indicated the maximum Tx power level (dBm) an UE may use in
GSM neighbor cell(n).
• P_MAX is the maximum UE power capability.
• If several GSM cells fulfil the equation above, cells are ranked based on HOPG:
AdjgPriorityCoverage/0, range 0…7
• A cell is ranked higher than another cell if it has a higher priority level even though
its signal strength condition was worse
• Inter-system HO could be forbidden during the first measurements reports from the UE, to let
the UE report all the candidate inter-system cells in its neighbour hood.
• FMCG NcellSearchPeriod / 0
421
Company Confidential
ISHO 3G 2G - AMR Signalling Flow
UE Node B RNC CN
RRC: Measurement Control
ISHO triggering (3
RRC: Measurement Report reasons are possible)
NBAP: Radio Link Reconfiguration Prepare
NBAP: Radio Link Reconfiguration Ready
Initial Compressed
NBAP: Radio Link Reconfiguration Commit Mode Configuration
RRC: Physical Channel Reconfiguration
RRC: Physical Channel Reconfiguration Complete
NBAP: Compressed Mode Command
RRC: Measurement GSM RSSI
RRC: Measurement Report Control Measurement
NBAP: Compressed Mode Command
RRC: Measurement Control GSM BSIC
RRC: Measurement Report Identification
RANAP: Relocation Required
• BSIC verification always performed for AMR calls – no interrupt in voice call
422
Company Confidential
ISHO 3G 2G - PS Signalling Flow
UE Node B RNC CN
RRC: Measurement Control
RRC: Measurement Report ISHO triggering (5 reasons are possible)
• In most cases BSIC verification is not required (data interrupt as UE moves to 2G)
• PS makes use of RRC: CELL CHANGE ORDER FROM UTRAN message
423
Company Confidential
Load based Handover GSM WCDMA
Packet
PacketData
DatatotoWCDMA
WCDMA Higher
Higherbit
bitrates
rates
regardless
regardless of GSMload
of GSM load for data users
for data users
Packet
PacketBCCH
BCCHparameters
parameterscan
canbe
beused
used
to push GPRS mobiles to WCDMA
to push GPRS mobiles to WCDMA
0%
424
Company Confidential
Load and Coverage Reason Handover
GSM
GSM GSM
GSM GSM
GSM GSM
GSM
WCDMA
WCDMA WCDMA
WCDMA
GSM
GSM WCDMA
WCDMA GSM
GSM
Handover
Parameters are sent from the BSC to the
HandoverTriggering
TriggeringThresholds
Thresholdsset
setininBSC
BSC mobile in the Measurement Information
message
In the S10.5 MS is measuring the UTRAN cells
Inter-RAT
Inter-RATmeasurements
measurementsstarts
startsinincase
case continuously if they are defined as a neighbor ,
The RXLEV
The RXLEV of the serving cell is aboveoror
of the serving cell is above
below
belowthe
thegiven thresholdQsearch_C
giventhreshold Qsearch_C The number of GSM neighbor reported could be 3-6
depending on the set value of parameter
FDD_MULTIRAT_REPORT (0,1,2=def,3).
Handover
HandoverDecision
Decisionisisdone
doneinincase
caseofof The load is measured in the cell level by
Load
Loadofofthe
theserving cell>>Load_Threshold
servingcell Load_Threshold comparing the occupied TCHs to the available
and CPICH Ec/No> min Ec/NoThreshold
and CPICH Ec/No> min Ec/No Threshold TCHs (20 s interval used)
Fdd_Qoffset: select
If, for suitable UMTS cell allways (value is -infinity)
428
Company Confidential
Measurements for 2G 3G inter-system cell reselection
GSM 3G 3G 3G
GSM
- Configuration 1 is not acceptable because camping on 3G in zones with good GSM coverage is not possible.
- Configuration 2: camping on 3G is not possible in high medium&poor GSM coverage.
- Configuration 3: camping on GSM/3G is possible and not constraint to other system’s coverage.
Therefore, configuration 3 is the best choice at expenses of decreasing the battery lifetime of the terminal
429
Company Confidential
Cell Re-selection Parameters
dBm -98 -94 … -74 Always -78 -74 -70 … -54 Never
Value 0 1 2 3 … 8 … 14 15
Always select
Reselect in case RSCP > GSM
irrespective of RSCP 430
value RXLev (RLA_C) +28dB
Company Confidential
Cell Re-selection Parameters
• FDD_Qmin, defines minimum Ec/No threshold that a 3G cell must exceed, in
order the UE makes a cell reselection from 2G to 3G. This is the new mapping
table
Fdd_Qmin mapping
Aif parameter 0 1 2 3 4 5 6 7
Fdd_Qmin (old) [dB] -20 -19 -18 -17 -16 -15 -14 -13
Fdd_Qmin (new) [dB] -20 -6 -18 -8 -16 -10 -14 -12
FDD_(GPRS) FDD_REP_
FDD_REP_ 3G_search_
Parameter
Parameter Qsearch_I
Qsearch_I Qsearch_P
Qsearch_P Qsearch_P
Qsearch_P FDD_QMin
FDD_QMin FDD_(GPRS)
Qoffset QUANT
3G_search_
PRIO
Qoffset QUANT PRIO
Name 2(3)quarter
Nameforfor 2quater 2quater 3quater
2quater 2quater 3quater 2ter
2ter 2quater
2quater 2ter
2ter 2(3)quarter
Sys Info (SI)
Sys Info (SI)
(PBCCH)
(PBCCH)
431
Company Confidential
Cell Reselection 3G 2G
• Whilst camping in a 3G cell the UE performs intra-frequency, inter-frequency, and inter-system
measurements based on the measured CPICH EcNo of the serving cell according to the
following rules:
– Serving cell parameters Sintrasearch, Sintersearch and SsearchRAT are compared
with Squal (CPICH Ec/No - Qqualmin) in S-criteria for cell re-selection
– UE will measure neighbor cells depending on how parameters are set & sent (if
parameters are not sent UE shall measure all cells)
• 1 – None (Squal > Sintrasearch )
• 2 - WCDMA intra-frequency (Sintersearch < Squal ≤ Sintrasearch)
• 3 - WCDMA intra- and inter- frequency, no inter-RAT cells (SsearchRAT < Squal ≤ Sintersearch)
• 4 - WCDMA intra- and inter-frequency and inter-RAT cells (Squal ≤ SsearchRAT )
4 3 2 1
WCDMA
CELL 432
Company Confidential
Cell Reselection 3G 2G
chanellization scrambling
code code 441
Company Confidential
Spreading Code Allocation
• Code Allocation Algorithm chooses the correct spreading code depending on the
TFC type
C 3 (0)=(…)
C 2 (0)=(1,1,1,1)
C 3 (1)=(…)
C1 (0)=(1,1)
C 3 (2)=(…)
C 2 (1)=(1,1,-1,-1)
C 3 (3)=(…)
C 0 (0)=(1)
C 3 (4)=(…)
C 2 (2)=(1,-1,1,-1)
C 3 (5)=(…)
C1 (1)=(1,-1)
C 3 (6)=(…)
C 2 (3)=(1,-1,-1,1)
C 3 (7)=(…)
Code Order Code Order 1 Code Order 2 Code Order 3
0 (SF 1) (SF 2) (SF 4) (SF 8)
The codes are layered from 0 to 11 according to the Spreading Factor (SF)
• Cm(n) : The code order, m, and the code number, n, designates each and every
code in the layered orthogonal code sequences
• In DL code order 2 to 8 (SF 4 to 512) are available (Nokia RAN does not support
SF = 512)
• In UL code order 2 to 7 (SF 4 to 256) are available 442
Company Confidential
Spreading Code Allocation – Example
Ordinary downlink speech channel, requires 30 ksps physical capacity (AMR
12.2 – 4.75 kbit/s). The code order is 8, which means there are 128 chips to
illustrate one symbol (2n, n=7). If the requested channel is 120kbit/s (including
DPCCH and channel coding) data channel, then the type of code is 6.
Spreading code cycle depends on the symbol rate.
9 15 7.5 512 40
8 30 15 256 160
443
Company Confidential
Downlink Code Allocation
•HSDPA with 5 codes allocated at cell MAC-hs start-up when HSDPA is enabled
•Code allocation is dynamic in future releases when more than 5 codes are allocated
SF = 1
SF = 2
SF = 4
SF = 8
SF = 16
SF = 32 Codes for 5
SF = 64 HS-PDSCH's
SF = 128
SF = 256
Code for one
HS-SCCH
Codes for the cell common channels
•166 codes @ SF=256 available for the associated DCHs and non-HSDPA uses
444
Company Confidential
Downlink Code Allocation – Common Channels
• Pilot (P-CPICH) and BCCH (P-CCPCH) need a fixed code allocation
– P-CPICH: CH256,0
– P-CCPCH: CH256,1
• AICH and PICH codes in Nokia RAN
– AICH: CH256,2
– PICH: CH256,3
• S-CCPCH code allocation depends on number of active S-CCPCH in Nokia
RAN
– With 1 S-CCPCH: CH64,1
SF = 32
SF = 64 S-CCPCH
SF = 128 X
SF = 256 X X
P-CCPCH
P-CPICH
AICH
PICH
445
Company Confidential
Spreading Code Allocation
• A code is always allocated from the optimum location in the code tree. It
makes the allocated code and the codes in the branches below and above
the allocated code unavailable
• Code tree will fragment quickly if releases is not re-arranged
• Re-arrangements in the code tree is done by reallocating the codes in better
locations
• The above code tree has 4 codes of equal order. The best locations are in
the same branch and very close to one another. The badly located codes are
released and optimally reallocated allowing the use of upper layer codes
• Codes are only reallocated when there is a benefit at two code tree layers
above the code being reallocated
446
Company Confidential
Part XIII
Drive Test Analysis
447
Company Confidential
Drive Test Process
Land Unit
NQMP
FTP
Server
449
Company Confidential
Drive Test Analysis – Test Equipment, Scanner
450
Company Confidential
Drive Test Analysis – Test Equipment, UE
• The Scanner measures all SCs, whereas the UE only measures SC
signals from the cells that the system has informed/ordered the UE through
the BCH (neighbour list) or via the “measurement control” message.
– RSCP Active/Monitored Set
– Carrier RSSI
– Ec/No Active/Monitored Set
– UL/DL Data Throughput
– The BLER downlink
– Pilot BER
– Random Access Initial Tx Power
– Random Access Preamble Count
– Random Access Preamble Step
– Random Access Tx Power
– SIR target (UE dependent)
– UE Tx Power
– Call Statistic: AMR, CS and PS data calls
451
Company Confidential
Drive Test Analysis – Call Patterns
• Enough call samples have to be made to make the measurement statistically valid.
• In a 50 call sample one dropped call will cause a change in performance of -2%
• In a 500 call sample one dropped call will cause a change in performance of -0.2%
• Call length should be defined at the beginning
• We can use different call testing patterns for different optimisation techniques
• Short Calls (for Calls setup performance and delay)
• Long calls (for Drop call performance and SHO performance)
Threshold KPIs
MOC Setup time 5 sec Signalling Statistics Count Success rate
MOC CCR 99 % Attach Attempt 155
Attach Success 140 90.32%
Attach Time more than threshold 5 3.57%
RAW END USER Attach Time less than threshold 135 96.43%
Event Count Ratio Count Ratio Attach Failed 15 9.68%
Call Attempts 132 122 Average Attach Setup Time 1.36
Activate PDP Context Attempt 124
Call Setup Success Rate 108 81.8% 108 88.5%
Activate PDP Context Success 124 100.00%
Call Setup Failure Rate 24 18.2% 14 11.5% Activation Time more than threshold 2 1.60%
Failures due to Tool (TSF) 10 41.7% Activation Time less than threshold 123 98.40%
Failures due to Core Problem 10 41.7% 10 71.4% Activate PDP Context Failed 0 0.00%
Failure ASU (Sync) Problem 2 8.3% 2 14.3% Average PDP Context Activation Time 0.96
FTP Download Attempts 51
Failure due to Low Coverage Levels 2 8.3% 2 14.3%
FTP Download Success 48 94.12%
Call Setup Success and Setup Time > 5s 7 6.5% 7 6.5% FTP Download throughput more than threshold 25 52.08%
Long Setup due to slow cell reselection 0.0% 0 0.0% FTP Download throughput less than threshold 23 47.92%
Long Setup due to clash with InterRAT reselection 0.0% 0 0.0% Average FTP Download Throughput 107.02
Long Setup due to Unknown (suspect UE) 0.0% 0 0.0% FTP Upload Attempts 32
FTP Upload Success 30 93.75%
Long Setup due to Unknown 0.0% 0 0.0% FTP Upload throughput more than threshold 20 66.67%
Average Call Setup Time 3.66 3.66 FTP Upload throughput less than threshold 10 33.33%
Call Completetion Rate 105 97.2% 105 97.2% Average FTP Upload Throughput 55.53
Call Drop Rate 3 2.8% 3 2.8% Data Transfer Cut-off Ratio 6.02%
PDP Context Dropped 4 3.23%
Call Drop Poor 3G Coverage 1 33.3% 1 33.3% Deactivate PDP Context Request 121
Call Drop on GSM due to Interference 2 66.7% 2 66.7% Deactivate PDP Context Accept 121 100.00%
Overall Call Completion Rate 105 79.5% 105 86.1% Deactivate PDP Context Failure 0 0.00%
ISHO Attempt 14 14 Detach Request 281
Detach Accept 129 45.91%
ISHO Success 14 100.0% 14 100.0%
Overall Data Session Completion Rate 78 62.90%
ISHO Failed 0 0.0% 0 0.0%
ISHO Failed cause physical channel failure 0 N/A 0 N/A RAU Statistics Count Success rate
RAU Attempt 22
•What ‘End User’ will contain Cell Reselection from UMTS to GSM Delay E2E
Cell Reselection from UMTS to GSM Delay Signalling
15.27
N/A
Cell Reselection from GSM to UMTS Attempts 1
Cell Reselection from GSM to UMTS Success 0 0.00%
•Other cuts of Data Cell Reselection from GSM to UMTS Delay E2E N/A
KPI-08 PS PDP •Same as 10 •PS PDP activation trigger point: •Cluster_PDP_Context_Activation_SR_% = PDP 98%
context •350 occurrences •UE not PS attached Context Activation Completion / PDP Context
activation •UE sends 1st ‘RRC Connection Request’ Activation Attempts *100
•PS PDP activation completion trigger point:
•UE receives ‘RRC: Downlink Direct Transfer (SM: Activate PDP
Context Accept)’
•Successful completion if:
•UE receives ‘RRC: Downlink Direct Transfer (SM: Activate PDP
Context Accept)’
KPI-09 PS •Repeated FTP calls, 1 Mbyte file •Data Session Set-up trigger point: •Cluster_PS_DL_Throughput = Average 200 Kbps
throughput download, 1 Ue •UE receives 1st DL packet Throughput Over Data Sessions
downlink •384 kbps •Data Session Completion trigger point:
•Average throughput •UE receives last packet
•350 calls •Session output: 454
•Average downlink throughput
Company Confidential
Need to Define the KPI measurement (from Drive test)
• Call Setup Success - CSSR (voice, circuit switched data)
– Successful call setup means that “DL/UL Direct Transfer (CC: Alerting)”
message is received by UE.
• Call Setup Time (voice, circuit switched)
– Call setup delay is measured from L3 messages, starting from “RRC
Connection Setup” message to “DL Direct Transfer (CC: Alerting)” message.
• Call Drop (voice, circuit switched)
– A dropped call occurs. The call will be dropped in case RRC connection release
(not normal release) message has been send from RNC to UE.
Call Established
Call Drop
Direct Transfer (Disconnect)
Rate
Direct Transfer (Release)
Direct Transfer (Release Complete)
Iu Release Command Call
Iu Release Complete
Duration
RRC Connection Release
RRC Connection Release Complete
RRC Connection Release Complete
RRC Connection Release Complete
Radio Link Deletion Request
Radio Link Deletion Response
ALCAP: Release Request
ALCAP: Release Response
ALCAP: Release Request
ALCAP: Release Response
Call Released
457
Company Confidential
Definition of Session Setup Time - PS
UE WBTS RNC
CN
UE already has an RRC connection
INITIAL DIRECT TRANSFER (Attach Request)
INITIAL UE MESSAGE (GPRS Attach)
SCCP: Connection Request
Uplink Direct Transfer (Authentication & Ciphering Response) RANAP: Authentication & Ciphering Response
Security Mode Command RANAP: Security Mode Command
Downlink Direct Transfer: Attach Accept Direct Transfer: Attach Accept Session
Set-up
Uplink Direct Transfer: Attach Complete Direct Transfer: Attach Complete Time
INITIAL DIRECT TRANSFER (Active PDP Context Request) DIRECT TRANSFER (Active PDP Context Request)
RANAP: RAB ASSIGNMENT REQUEST
RRC: Radio Bearer Set-up
RRC: Radio Bearer Set-up Complete
RANAP: RAB ASSIGNMENT
RRC: Measurement Control RESPONSE
DIRECT TRANSFER (Active PDP context Accept)
458
Company Confidential
Failure Breakdown
• Non-genuine failures
– Measurement system fault
It is beneficial to (Collection Tool or Analysis)
categorise call failures • Genuine failures
during the analysis and – RF issue (Coverage /
reporting Interference / Poor
dominance)
– Missing neighbour
– System issue WBTS
– System issue RNC
– Core network issue
UE
– System
– (Unknown )
Iub Iu
WBTS RNC Core NW
459
Company Confidential
Failure Breakdown
Time
460
Company Confidential
Failure Breakdown
AMR Call Setup Failures
Call Set-up Failures Breakdown - Benchmark Drive
Call Setup Failure Breakdown - Final Drive
5% 5% Equipment 0%
0%
0%
Registration Clash
0% Equipment
15%
Core Network 19% 19% Registration Clash
20%
Fail in 2G Core Network
Fail in 2G
RAN (fixed in next release) 0%
RAN (fixed in next release)
Site maintenance (TMUK) 0% 0%
Site maintenance (TMUK)
0% RAN (Other)
RAN (Other)
10%
5% Missing neighbours
Missing neighbours
0% 3G RF 3G RF
0%
0%
slow cell Reselection 0% Slow cell reselection
462
Company Confidential
Non Genuine Call Setup Failure Scenarios
• Measurement systems are often not perfect and may introduce errors
in data collection or analysis
463
Company Confidential
Non Genuine Call Setup Failures
464
Company Confidential
Non Genuine Call Setup Failures
• Measurement system failures by
drive test tool
• Call attempt during Location
Area update (LA update clash)
LA Update
Request
Call attempt
RRC
Release
465
Company Confidential
Non Genuine Call Setup Failures
Measurement system failures by Actix workbook
– Successful or failed Location Area update interpreted as call setup failure
LA Update
Request
RRC
Connection for
Inter-RAT cell
reselection
466
Company Confidential
Non Genuine Drop Call Scenarios
467
Company Confidential
Genuine Call Setup Failure Scenarios
– RF issue
• Interference / Dominance / Coverage
• Missing neighbour
– System Issue - BTS
• No response to “RRC Connection Request”
• “RRC Connection Reject” to “RRC Connection Request”
– System issue - RNC
• “CC Disconnect” after “Call Proceeding” due to “DL RRC Connection
Release”
– Core NW
• “CM Service Abort” after “CM Service Request”
– System issue (test number)
• “CC Disconnect” after “CC Progress”
468
Company Confidential
Genuine Drop Call scenarios
• RF issue
– Interference / Dominance / Coverage
– Missing Neighbours
• System issue BTS
– Sudden “CC Disconnect” due to “DL RRC Connection Release”
– Sudden drop to idle, no disconnect messaging
• System issue RNC
– Sudden “CC Disconnect” due to “DL RRC Connection Release”
469
Company Confidential
Failure Location
• Analyse the
signalling
flow to find
the location
of failure and
potential
cause
• UE log may
only capture
some of the
messages
470
Company Confidential
AMR CS Call Phases
UE WBTS DRNC SRNC CN
1. Cell search, BCCH
decoding & RACH access
4. RAB set-up
6. Service Established
8. Service Released
471
Company Confidential
Call Setup Failure Start
Missing No
Neighbour ?
Dominance Optimization
Yes
Best server’s No
Ec/No > -12dB Yes Neighbour list Optimization
Yes
UL coverage & RACH
B parameter. Optimization
(changing serving cell)
No AICH(ACK) received? Report & Finish
(Check failure cause)
Yes No
“RRC Connection
Setup” received?
No “RRC Setup
Reject” received?
C
Yes Yes
AC optimization (check PrxNoise
Ye
s
E
Report & Finish
Check failure cause
(Not radio problem/cell update) 472
Company Confidential
Call setup failures – RF issue A
• RF issue? Coverage / Interference / Dominance
474
Company Confidential
Call Setup Failure Analysis- Block B - B
• The purpose of this activity is to check the Random Access Process
is working adequately by investigating whether AI (Acquisition
Indicator) has been received through DL AICH
• If AICH was not received by UE, the cause of the problem can be
classified into:
– Inadequate RAN parameter related to Random Access: RAN parameter
settings for pre-amble transmission or open loop power control
information is not correct.
– UL Coverage limit: UL coverage of UE is smaller compared to serving
cells DL coverage so that UE’s Tx power cannot reach serving cell.
• The Basic theory for RACH setup procedure and planning parameters
can be found in Module 6 – Parameter Optimisation
475
Company Confidential
Call Setup Failure Analysis- Block B - B
UE WBTS RNC
Preamble/RACH
Acquisition Indicator/AICH
L1 Synchronisation
UE in CELL_DCH state
476
Company Confidential
RACH Process
B
L1 ACK / AICH
Downlink Not detected
BS
UEtxPowerMaxPRACH
PowerRampStepPRACHpreamble
…… ……
Uplink
MS Preamble Preamble
1 2
Message part
PowerOffsetLastPreamblePRACHmessage
PRACH_preamble_retrans
# PRACH preambles transmitted during one PRACH cycle without receiving AICH response
RACH_tx_Max
# preamble power ramping cycles that can be done before RACH transmission failure is reported
477
Company Confidential
Call Setup Failure Analysis- Block B-
• Solutions for RACH optimisation
B
No
Max UE Tx power hit To
Toincrease
increasePRACH_Preamble_retrans
PRACH_Preamble_retrans
the UE_P_MAX(24dBm)? Or PowerRampStepPRACHPreamble
Or PowerRampStepPRACHPreamble
Yes
Yes
Is UL Interference Report
Reportthere
theremight
mightbebean
aninterfering
interfering
abnormally HIGH? source
source Nearby the servingcell
Nearby the serving cell
No
Change
Changethe
theServing
Servingcell
cellto
tocover
coverthe
theproblem
problemArea
Area
=> UE is too far to reach the serving cell
=> UE is too far to reach the serving cell
478
Company Confidential
Call Setup Failure Analysis- Block B B
479
Company Confidential
Call setup failures – System issue BTS C
• No response to “RRC Connection Request”
– Good RF conditions
– Wrong MHA settings or cable loss settings can cause the site not to “hear” the
UE
– PrxNoise statistics, receive link parameters and HW units to be checked (faulty
MHA, wrong MHA parameters, wrong cable / feeder loss parameters, faulty
units)
480
Company Confidential
Call setup failures – System issue BTS C
• “RRC Connection Reject” after “RRC Connection Request”
– Good RF conditions
– Admission Control can reject too many (or admit too many) connection
requests due to wrong PrxNoise measurements.
– PrxNoise statistics, receive link parameters and HW units to be
checked
481
Company Confidential
Call Setup Failure Analysis
• UE has the appropriate DL/UL coverage but if RNC does not allow to
C
set up the RRC connection of the requested RAB (Radio Access
Bearer), Call setup will fail.
• Admission Control (AC) is involved in RRC connection setup. AC can
reject RRC reject RRC connection Setup due the DL Load, UL load or
DL Spreading codes
– Overload Area:
• If measured UL (PrxTotal) or DL (PtxTotal) load exceeds overload
thresholds (PrxTarget + PrxOffset and PtxTarget + PtxOffset) then
AC can't admit more RABs to the cell 482
Company Confidential
Call Setup Failure Analysis C
• During the pre-optimization phase it is unlikely that AC will stop an RRC
connection setup during the drive testing because there are normally very
few UEs in the network. (Traffic loading is trivial)
• However, it should be checked that measured PtxTotal and PrxTotal are
less than PtxTarget (e.g. 40dBm) and PrxTarget (e.g. 4dB, 60% loading)
respectively.
• If DL AC does not allow RRC setup check the Tx power of WBTS, # of
channels transmitted, Signaling messages.
• If UL AC does not allow RRC setup: Check out if there is an interfering
source nearby the serving cell.
483
Company Confidential
Call Setup Failure Analysis D
To check if Layer 1 Synchronization (slot/frame sync) has failed
• If “RRC Connection Setup” was received by UE but UE does not send “RRC
Connection Setup Completed”, we will report “L1 synchronization failure” and
have to check L1 system messages.
484
Company Confidential
Call setup failures – System issue RNC E
• “CC Disconnect” after
“Call Proceeding”
• Good RF conditions
• Failures in RAB setup
occur between the “RAB
Assignment Request”
being received from Core
Network and the RAN
sending out Radio
Bearer Setup. Therefore
the failure is between
BTS and Core Network.
485
Company Confidential
Call setup failures – System issue RNC E
• “CC Disconnect” after “Call
Proceeding” (cont.)
• An example (site shows high
values on counter
“RAB_STP_FAIL_CS_VOICE
_BTS” during the drive test
• In the recent check the
counter showed no failures.
486
Company Confidential
Call setup failures – Core NW E
• “CM Service Abort” after
“CM Service Request”
• Good RF conditions
• “Security Mode
Command”-message not
received by UE, thus the
failure is believed to be at
Core Network. UE Node B RNC MGW
• RRC: Initial Direct Transfer message is sent using acknowledged mode RLC to the CS core
domain. Routing is to be based upon the local P-TMSI
• The NAS message is not read by the RNC but is forwarded to the multimedia gateway. The NAS
message includes the IMSI as a UE identity
• The SCCP: Connection Request message establishes the connection orientated signalling link in
the same way as it was for the RRC connection phase.This does not reserve any resources for the
AMR call itself.
• The Connection Confirm message identifies the RNC with a destination local reference which is the
same as the source reference within the Connection Request message
• The Connection Confirm message identifies the CS core with a source local reference
• The CS core sends a RANAP: Location Reporting Control message to the RNC requesting
information regarding the location of a particular UE
• The RANAP: Common ID message specifies the IMSI belonging to the UE
487
• The Security Mode Command message triggers the start or stop of ciphering and integrity
protection.
Company Confidential
Call setup failures – System Issue (test number)
E
• “CC Disconnect” after “CC Progress”
• Cause: recovery on timer expiry
• The call goes via IN SCP to a
recording.
• A static test was done by Nokia
Customer Care and in few instances
the call dropped after 30 seconds of
recording passed. Hence the problem
is associated with the test number not
the RAN
30
sec
Cause: recovery
on timer expiry
488
Company Confidential
Call Drop Failure Analysis Process Start
No
Best server’s No
Coverage Optimization
RSCP > -102dBm
Investigate
B possible BTS
or RNC
problem 489
Company Confidential
Call Drop Failure Analysis Process (SHO Analysis)
Start
Yes
Check Iur Inter RNC HO
Yes
Check neighbour definition parameters
No
Yes No DL ASU
Fix SC Clash SC Clash
received
C
Yes
No
Yes Congestion on
Load Optimisation
target cell
UE Tx Power Max Check RF Levels
No
DL Tx Power Max No
Yes
D
Yes Load Optimisation/
Uplink Interference
External Interferer
Yes
No
Yes
Link Unbalanced CPICH Optimisation
490
Company Confidential
Drop call failures – RF issue A
• RF drops mostly due to poor
dominance or interference
• Poor coverage could lead to ISHO,
although poor dominance or
interference can cause ISHO to
fail.
• Rapid field drop can cause drop
due to coverage
• Poor dominance or interference
can cause Compressed Mode
(CM) to start even if RSCP is still
good.
• In CM UE transmits with higher
power (more interference) and
spends less time on 3G (less
Poor dominance
accurate measurement reporting) causes Active Set
• Poor dominance or interference update failures
can lead to Active Set update
failures and eventually to drop call. 491
Company Confidential
Drop call failures – RF issue A
TrChAgg and DL
DPCCH BER
high
492
Company Confidential
Drop call failures – RF issue A
Fairly good
CPICH Pilot
EcNo
Transport Channel
BER. Btw UE<-
>RNC (MAC layer)
493
Company Confidential
Drop call failures – System issue BTS B
• Sudden drop to idle, no disconnect messaging
– Site malfunctions to be checked
– In the example below site had faulty unit (WTR)
Drop to IDLE
494
Company Confidential
Drop call failures – System issue RNC
B
• “CC Disconnect” due to DPCCH BER
“DL RRC Connection
Release”
• No response to UL
Measurement Reports
• In the example site had
no alarms, good RF &
BER
• Not able to add SC265
to Active Set, next call
on the same cell => no
failure.
• Difficult to troubleshoot if
the failure does not
happen systematically Sudden “RRC
=> follow up in the next Connection Release”
weeks drive / do a
separate drive test in the
area 495
Company Confidential
Drop call failures (SC conflict) C
• Sudden drop to idle Transport
mode (no disconnect channel
messaging) BLER 100%
• Cause of the failure:
overshooting site and
SC reuse
• Short term solution to
add overshooting
neighbour in ADJS
definitions
496
Company Confidential
Drop Call - Uplink Interference D
UL interference from the
SIB7 message
497
Company Confidential
Drop Call – Link Balance D
• UL & DL Power Control commands can UE RX power control
message: DL reception
help indicating problems in link balance. weak -> UE is ordering
WBTS to increase power.
• PC frequency is 1500 Hz, thus ideally
the sum of PC commands to increase or
decrease power is 1500
• E.g. if the sum of UL PC commands is <
1500, this would indicate UE is starting
to loose synchronization
• in Compressed Mode there is less PC
commands, UE spends time on 2G
Sum of UL PC commands
< 1500, UE not receiving all
the PC commands.
498
Company Confidential
Drop call failures – System issue RNC or BTS ?
• “CC Disconnect” due to “DL RRC Connection Release” is just a
consequence of failure which can be due to different reasons
– From UE point of view L3-messaging does not identify the point of failure
distinctly
– BTS or RNC failure? => Suspect BTS first, then RNC
500
Company Confidential
Content
• HSDPA technology and RAS05.0 features and main parameters
– HSDPA Introduction
– BTS/MAC-hs features
• Mac-hs Protocol
• Physical channel structure
• HSDPA Power
• Link adaptation
• Flow control
• Scheduling
• L1 error correction
– RRM
• Channel type selection
• Power allocation
• HS-DSCH release due to low utilisation or low throughput
• Associated UL DPCH
• Mobility
• HSDPA RAN dimensioning
– HSDPA Dimensioning process
– Air interface capacity dimensioning
– DL Link budget, coverage
– UL link budget, coverage
– WBTS Channel Element dimensioning
– Example dimensioning case
• Planning Process
– Pre-planning with Nokia NetAct planner
– HSDPA design and upgrade options
– Indoor HSDPA solutions
501
Company Confidential
Why HSDPA?
Improved
– cell throughput
– spectral efficiency
• Meant for NRT services (background and interactive classes supported first).
502
Company Confidential
Faster Content Download Times with HSDPA
100
GPRS 3+2
90 EDGE 3+2
80 WCDMA 384 kbps
70 HSDPA (700 kbps)
WLAN .11b
60
Seconds
50
40
30
20
10
0
30 kB digital 100 kB video 300 kB Symbian 4 MB MP3 Fast
image clip application download of
very large
content
503
Company Confidential
HSDPA – General Principle
• Channel quality
information
• Error correction
Ack/Nack •Shared DL data channel
L1 Feedback •Fast link adaptation,
Data scheduling and L-1 error
correction done in BTS
•1-5 codes in RAN05
Terminal 1 (UE) (max.15 codes RAN06)
L1 Feedback
Data •QPSK or 16QAM
modulation
•User may be time
and/or code multiplexed.
Terminal 2 504
Company Confidential
Link Adaptation in HSDPA
l
• Adaptive modulation and coding : ig na
in ks
modulation, number of codes and code l k
o wn dbac
rate are optimised dynamically according D e
1) I Fe
to the CQI received from the UE CQ l
2 )
igna
s 3) Adjust modulation
n link scheme according
w
) Do to feedback received
4
from the UE
14 by UE with fading
12
10
8
6
4
2
0
-2
Modulation
0 20 40 60 80 100 120 140 160
Time [number of TTIs]
16QAM3/4 Link BTS adjusts link adaptation mode
adaptation (Modulation & Coding Scheme)
16QAM2/4 mode with a few ms delay based on
channel quality (CQI) reports
QPSK3/4 from the UE
QPSK2/4
QPSK1/4
506
Company Confidential
HSDPA Peak Bit Rates
Coding
Codingrate
rate Coding
Codingrate
rate 55codes
codes 10
10codes
codes 15
15codes
codes
1/4
1/4 600
600kbps
kbps 1.2
1.2Mbps
Mbps 1.8
1.8Mbps
Mbps
QPSK
QPSK 2/4
2/4 1.2
1.2Mbps
Mbps 2.4
2.4Mbps
Mbps 3.6
3.6Mbps
Mbps
3/4
3/4 1.8
1.8Mbps
Mbps 3.6
3.6Mbps
Mbps 5.4
5.4Mbps
Mbps
2/4
2/4 2.4
2.4Mbps
Mbps 4.8
4.8Mbps
Mbps 7.2
7.2Mbps
Mbps
16QAM
16QAM 3/4
3/4 3.6
3.6Mbps
Mbps 7.2
7.2Mbps
Mbps 10.7
10.7Mbps
Mbps
4/4
4/4 4.8
4.8Mbps
Mbps 9.6
9.6Mbps
Mbps 14.4
14.4Mbps
Mbps
240Ksps=480Kbps
480*5*3/4= 1800 Kbps (LAYER 1 THR’PUT !!)
RAN05 (Inter-TTI = 1)
• User throughput depends upon the number of active HSDPA users in the
cell/BTS.
2nd TTI
Three HSDPA UEs
receiving data
1st
TTI
3rd TTI
508
Company Confidential
Release’99 vs HSDPA Retransmissions
RNC
RLC
ACK/NACK
BTS
Packet
L1Retransmission
RLC
ACK/NACK L1 ACK/NACK
Terminal
509
Company Confidential
New Node B Functionality for HSDPA
Packets
Scheduler ARQ &
& Buffer Coding
Flow Control
ACK/NACK &
Feedback
Decoding
WBTS processing capability 1 WSPC per BTS 1 WSPC per WCEL (3WSPC
per BTS)
Max number of users per BTS 16 users (QPSK); 12 48 users (16 per cell)
users (16 QAM)
Maximum throughput 3.6 Mbit/s per BTS 3.6 Mbit/s per WCEL
512
Company Confidential
Nokia HSDPA Roadmap
513
Company Confidential
Nokia HSDPA Roadmap
514
Company Confidential
MAC-hs
MAC-hs functions in BTS For each subframe = 3 slots = 2 ms
• Iub Flow Control • Interpret incoming Ack/Nack and CQI
• Data Buffering information.
• Packet Scheduling • Perform Packet Scheduling, i.e. choose
• Link Adaptation one priority queue for transmission.
• Hybrid ARQ – Determine if retransmission is
needed.
• New transmission.
MAC-hs Protocol Services
– Perform Link Adaptation.
• Data transfer on HS-DSCH
– Build MAC-hs PDU.
• In-order delivery
within priority queue • Retransmission
– No need for LA.
– Use same MAC-hs PDU.
• Perform Power Control for HS-SCCH.
SCCH
515
Company Confidential
HSDPA Protocol Model
• HS-SCCH, HS-DPCCH
UE Uu BTS Iub RNC
– Physical Channles
516
Company Confidential
Physical Channels for One HSDPA UE
DL CHANNELS
• HS-PDSCH: High-Speed Physical
BTS Downlink Shared Channel
• HS-SCCH: High-Speed Shared
Control Channel
Rel99 • Associated DPCH, Dedicated
Associated DPCH
Associated DPCH
1-4 x HS-SCCH
UL CHANNELS
• Associated DPCH, Dedicated
Physical Channel
• HS-DPCCH: High-Speed Dedicated
Physical Control Channel
UE
517
Company Confidential
HSDPA DL Physical Channels
– More than one HS-SCCH required when New data indicator 1 bit
code multiplexing is used UE identity 16 bits
518
Company Confidential
HSDPA DL Physical Channels
• Associated DPCH, Dedicated Physical Channel (as many as HSDPA users)
– Transfers signalling (Signalling Radio Bearer (SRB)) information e.g.
RRC measurement control messages
– Fixed SF256
– Power control commands for associated UL DCH
– DPCH needed for each HSDPA UE.
– No SHO support in RAN05 (coming in RAN05.1)
519
Company Confidential
HSDPA UL Physical Channels
• HS-DPCCH: High-Speed Dedicated Physical Control Channel
– MAC-hs Ack/Nack information (send when data received).
– Channel Quality Information, CQI reports (send in every 4ms)
– SF 256
523
Company Confidential
HS-SCCH Inner Loop Power Control
Node B look-up table
CQI Value λCQI
• In RAN05, the CQI report based HS-SCCH power 1 -5.05
2 -5.68
control shall be used based on the following equation: 3 -6.67
4 -7.73
5 -8.62
PSCCH = PCPICH + Γ + λCQI + P0 6
7
-9.66
-10.5
8 -11.56
where: 9 -12.5
10 -13.57
-Γ is power offset between CPICH and HS-PDSCH. 11 -14.54
Signaled to Node B and UE by RNC. (In Nokia 12 -15.62
13 -16.64
implementation: Γ=PtxMaxHSDPA – PCPICH = PHS- 14 -17.65
15 -18.79
PDSCH+PHS-SCCH – PCPICH) 16 -19.66
17 -20.72
-λCQI is the CQI dependent power offset 18 -21.55
19 -22.83
-P0 is the output of the HS-SCCH outer loop power 20 -23.79
control 21
22
-24.72
-25.65
23 -26.65
24 -27.65
λCQI is defined using a Node B look-up table from CQI 25 -28.65
26 -29.65
27 -30.65
28 -31.65
29 -32.65 524
30 -33.65
Company Confidential
HS-SCCH Outer Loop Power Control
• The HS-SCCH outer loop power control is used to fine-tune the power
offset between the HS-PDSCH and HS-SCCH for a given UE.
• When a TB has been transmitted on HS-PDSCH to a UE, the HARQ manager
shall check ACK/NACK feedback from the UE. After the HARQ manager gets
the ACK/NACK feedback or detects the DTX (=no feedback in predefined time)
in uplink HS-DPCCH, it shall inform LA.
• Based on reception of the ACK/NACK the output of the outer loop power control
P0 shall be updated by the following rule:
UL Ack/Nack CQI
detection
527
Company Confidential
HS-DSCH Link Adaptation Inner loop
528
Company Confidential
HS-DSCH Link Adaptation Outer loop
• The outer loop LA algorithm is used to compensate for the bias introduced in the
inner loop LA algorithm. The output from the outer loop algorithm is the scaling
factor A, which is used in the definition of X for CQI offset compensation
• For each UE, the scaling factor A is set as the initial value Ainit when the UE is
assigned to use HS-DSCH for packet data transmission. After each transmission,
the scaling factor A shall be adjusted based on the ACK/NACK feedback by the
following rules:
529
Company Confidential
CQI Reporting from UE CQI TBS codes M ∆
(dB)
• For CQI reporting the UE measures channel 1 137 1 QPSK 0
2 173 1 QPSK 0
quality of CPICH and assumes that HS- 3 233 1 QPSK 0
PDSCH codes will be transmitted with 4 317 1 QPSK 0
5 377 1 QPSK 0
combined power 6 461 1 QPSK 0
PHS-PDSCH = PCPICH + G + ∆ 7 650 2 QPSK 0
8 792 2 QPSK 0
9 931 2 QPSK 0
10 1262 3 QPSK 0
• Measurement power offset G signalled to 11 1483 3 QPSK 0
Node B and UE by NBAP and RRC (In Nokia 12 1742 3 QPSK 0
13 2279 4 QPSK 0
implementation; Γ=PtxMaxHSDPA - 14 2583 4 QPSK 0
PCPICH) 15 3319 5 QPSK 0
16 3565 5 16QAM 0
• Reference power adjustment D is used to 17 4189 5 16QAM 0
signal how much quality is better than 18 4664 5 16QAM 0
19 5287 5 16QAM 0
maximum TBS supported by UE (given in UE 20 5887 5 16QAM 0
category specific tables). 21 6554 5 16QAM 0
22 7168 5 16QAM 0
• Table to the right (from 25.214) defines CQI 23 7168 5 16QAM -1
for UE categories 1 to 6. Based on 10% 24 7168 5 16QAM -2
25 7168 5 16QAM -3
BLER. 26 7168 5 16QAM -4
• Algorithm to calculate CQI value is not 27 7168 5 16QAM -5
28 7168 5 16QAM -6
specified by 3GPP so CQI reporting can 29 7168 5 16QAM -7
have UE specific differences. 30 7168 5 16QAM 530
-8
Company Confidential
CQI Compensations
• The UE reports the CQI measured assuming a transmit power PCPICH + Γ + ∆.
• CQI offset compensation is used to take into account:
• The offset X is used to convert the measured CQI value into a compensated
one using the formula:
– CQIcompensated = CQImeasured + X (X expressed in dB)
where
– X = PHS-PDSCH – (PCPICH + Γ) – A – Y (powers expressed in dB)
where
– A is the output of the Outer Loop LA Algorithm
– Y is the Offset compensation based on TPC. Not supported in WN3.0
(Y=0)
531
Company Confidential
CQI Compensations - Example
CQImeasured
X = 3 dB
CQIcompensated
532
Company Confidential
Missing CQIs
• If BTS hasn’t receive any CQI report before scheduling data for UE, CQI
value #5 is used to define TFRC (transport format and resource combination).
In this case, the CQI offset compensation will not be applied.
533
Company Confidential
Link Adaptantion SUMMARY
• HS-DSCH link adaptation algorithm description:
– Nokia’s HS-DSCH link adaptation algorithm is based on an inner loop and
outer loop control mechanism.
– Inner loop algorithm: Adjusts the HS-DSCH transmission format every TTI
based on CQI (Channel quality indicator) measurement reports from the
HSDPA-user terminal.
– Outer loop algorithm: Controls the BLER operating point of the inner loop
algorithm by monitoring the Layer-1 Ack/Nack’s from past transmissions
(and retransmissions).
• Comments on HS-DSCH link adaptation performance:
– The BLER on the first transmission is typically 10%-15%, which results in
a residual BLER of approx. 1% after the first retransmissions when using
HARQ with Chase Combining.
– The probability of having an RLC retransmission triggered after 3 Layer-1
retransmissions is less than 0.5%, and often less than 0.1%.
534
Company Confidential
HSDPA Flow Control
BTS Buffers
UE1 UE1
Scheduling
UE2 UE2
UE3 UE3
low
Low coverage of high data rates 539
Company Confidential
Scheduling: Round Robin (FR)
• The round robin scheduling method is used in RAN’05 to select the
transmission turn for each priority queue. The scheduling is made at the cycle
of TTI (2ms) and based on the scheduling counter R of each priority queue.
• The R shall be updated in every TTI for all the priority queues as the following
rule:
– If one priority queue was scheduled in any TTI, R of the priority queue
shall be reset as 0.
– If one priority queue cannot be scheduled in any TTI due to the minimum
inter-TTI interval requirement, R of the priority queue shall remain the same.
– If there is no data in BTS buffer of one priority queue and there is no
pending retransmission in HARQ processes for the priority queue, R of the
priority queue shall remain the same.
– Otherwise, R of the priority queue shall be updated as R=R+1
• At each scheduling interval (2ms), PS shall select the priority queue, which has
the highest R among those priority queues that can be scheduled in next TTI.
• If several priority queues have the same R value, the priority queue, which has
the highest CQI report value, shall be selected.
• The initial value of R for the new priority queue is 0. 540
Company Confidential
RAN05.1
PROPORTIONAL FAIR IDEA
The
Thepacket
packetscheduler
schedulerisislocated
located
ininthe
the Node-B. This opensup
Node-B. This opens upfor
for
Channel quality “fast scheduling” optimized
“fast scheduling” optimized
according
accordingtotothethecurrent
currentchannel
channel
(CQI, Ack/Nack, TPC) conditions
conditionswhich
whichcancanyield
yieldlarge
large
gain at cell and user levels.
gain at cell and user levels.
Basic
Basic diversity
diversity mechanism
mechanism
Data utilizedisismulti-user
utilized multi-userdiversity
diversity==
selection
selectiondiversity
diversityamong
amongthe the
active users.
active users.
UE1
TTI 1 TTI 2 TTI 3 TTI 4
Scheduled user
Multi-user selection diversity
(give shared channel to “best” user)
Channel quality
(CQI, Ack/Nack, TPC) Data
UE2
USER 2 Es/N0 USER 1 Es/N0
541
Company Confidential
RAN05.1
Proportional Fair Resource (PF)
HARQ process
544
Company Confidential
HSDPA L1 Retransmissions
545
Company Confidential
HSDPA L1 Retransmissions: Chase Combining
Turbo Encoder
Systematic
Parity 1
Parity 2
Systematic
Parity 1
Parity 2
546
Company Confidential
HSDPA L1 Retransmissions : Incremental Redundancy
Turbo Encoder
Systematic
Parity 1
Parity 2
Systematic
Parity 1
Parity 2
547
Company Confidential
CHANNEL TYPE SELECTION
548
Company Confidential
CHANNEL TYPE SELECTION
HS-DSCH is selected if ALL of the following conditions are met:
1. Traffic class and traffic handling priority are allowed on HS-DSCH
– The operator can configure which traffic classes and handling priorities are allowed to be used with HSDPA with
HSDSCHQoSclasses (RNC) parameter.
– RAN05 only interactive and background traffic classes are supported
2. UE capability supports HS-DSCH
3. The cell supports HSDPA and HS-DSCH is enabled in the cell
4. No multi-RAB (RAN05) or supported multi-RAB combination (RAN05.1)
– No multiRAB in RAN05
– AMR + HSDPA possible in RAN05.1
5. The number of simultaneous HS-DSCH allocations in the BTS/cell is below the
maximum number.
6. HsdschGuardTimerHO and HsdschGuardTimerLowThroughput guard timers are not
runnig
– Both guard timers are operator-configurable parameters
7. UE is not performing inter-frequency or inter-system measurements
8. Active set size = 1 (RAN05); RAS5.1 if HSDPAMobility enable Active set size can be > 1
9. UE does not have DCHs scheduled with bit rates higher than 0kbps.
10. HS-DSCH physical layer category is supported
11. If there is no existing MAC-d flow in the cell, condition (A or B, depending on the
HSDPApriority parameter) has to be valid.
A) PtxNC<=
PtxNC<=PtxtargetHSDPA
PtxtargetHSDPA (HSDPApriority
(HSDPApriority = 1)
B) Ptxtotal<=
Ptxtotal<=PtxtargetHSDPA
PtxtargetHSDPA (HSDPApriority
(HSDPApriority = 0)
549
Company Confidential
Connection setup
• MAC-d flow can be allocated for UE when UE getting capacity request in cell
FACH or in cell DCH 0kbps
550
Company Confidential
MAC-d Flow Setup – UE NOT in CELL DCH
UE BTS RNC PS-CN
Connection between UE and CN. UE is in CELL_FACH state, NRT RAB established, NRT RB and SRBs mapped to CCH.
551
Company Confidential
MAC-d Flow Setup – UE in CELL DCH
552
Company Confidential
HSDPA Power Allocation
Range 0…50 dBm
Step 0.1 dB
Default 37.8 dBm (6 W)
Link adaptation
Power control
• PtxMaxHSDPA determines the achievable throughput
• PtxMaxHSDPA = Ptx hs-pdsch + Ptx hs-scch
• PtxMaxHSDPA is sent to BTS in the NBAP Shared Control Channel Procedure at
BTS start
• PtxMaxHSDPA is allocated when 1st Mac-d flow enters the cell
• PtxMaxHSDPA is time shared between HS-DSCH users. If there is already a
Mac-d flow then a new Mac-d flow is always admitted until the limit of 16 users per
BTS
553
• If there are no Mac-d flows PtxMaxHSDPA power is not kept reserved
Company Confidential
HSDPA Priority
Node-B Tx power
HSDPAPriority =1
Max power
A B C 554
Company Confidential
HSDPA Priority
HSDPAPriority =2 Node-B Tx power
Max power
• Event A: First MAC-d flow
entering the cell
Ptxtotal<=PtxtargetHSDPA
-> HS-DSCH is selected, otherwise
DCH scheduling.
Ptxtotal
• When HSDPA power is in use:
PtxNC target is Ptxtarget. NRT Ptx_offset
DCH scheduling upto Ptx_target
PtxtargetHSDPA. PtxMaxHSDPA
• Event B:
Ptx_offset_HSDPA
PtxnonHSDPA>=PtxtargetHSDPA
+PtxoffsetHSDPA Ptx_target_HSDPA
A B C 555
Company Confidential
PtxtargetHSDPA AND PtxoffsetHSDPA
556
Company Confidential
HS-DSCH & MAC-d release
557
Company Confidential
HS-DSCH & MAC-d release
PDUs in buffer
Throughput
Low utilization indication
MACdflowUtilRelThr
MACdflowthrougputRelThr
558
Company Confidential
HS-DSCH & MAC-d release
PDUs in buffer
Throughput
Low throughput indication
MACdflowUtilRelThr
MACdflowthrougputRelThr
559
Company Confidential
HS-DSCH & MAC-d release
L3 starts procedure to release UL NRT DCH and MAC-d flow if:
– MAC-d flow has low utilization and UL NRT DCH can be released
Or
– MAC-d flow has low throughput and UL NRT DCH can be released. In
this case UE specific timer HsdschGuardTimerLowThroughput is started.
If MAC-d flow has both low utilization and low throughput conditions valid at the
same time, the functionality described in case of low utilization is followed
560
Company Confidential
Associated UL DPCH bitrate
561
Company Confidential
Associated UL DPCH bitrate – use case 1/3
562
Company Confidential
Associated UL DPCH bitrate – use case 2/3
563
Company Confidential
Associated uplink DPCH bitrate – use case 3/3
At t1 due to congestion
HSDPA initial bit rate
cannot be allocated so
DCH is scheduled
instead.
In the example also DCH
scheduling is
unsuccessful so at time
t2 another capacity
request is sent.
564
Company Confidential
Mobility: Nokia 3G Parameter Database Structure
100
1 RNC
FMCG 32*ADJG
HSDPA
NRT
RT WBTS 100
1
100 48*ADJI HOPG
1
32*ADJS 100 NRT
FMCI RT
WCELL
HSDPA
NRT 100
RT 1
100
HOPI
100
1
1
HOPS NRT
FMCS RT
HSDPA
HSDPA NRT
NRT RT
RT
565
Company Confidential
Mobility: FMCx-HOPS-RNC parameter planning
• Events 1F and 6A trigger HS-DSCH release and channel switching to a DCH 0/0
RAB
• Event 1A triggers HS-DSCH release and channel switching to FACH
• Therefore HSDPA HO parameters can be tightened, to maximize HSDPA territory:
– FMCS: AdditionWindow
– FMCS: Addition Time
– FMCS: HHoEcNoThreshold
– FMCS: HHoRscpThreshold
– FMCI: InterfreqUETxPwrThrNrtPS
– FMCG: GSMUETxPwrThrNrtPS
– HOPS: Enable RRC Release = if it is enabled the channel switch to FACH
doesn’t happen when the event 1A is received but when the neighbor cell
CPICH EcNo is above the serving cell CPICH EcNo by negative margins
(HSDPA HOPS parameters ReleaseMarginAverageEcNo and
ReleaseMarginPeakEcNo)
– RNC: SHO HSDPA Capable UE = if it is enabled the SHO for the R5 UEs
when there is only the SRB (DCH0/0) will happen according to the addition
window defined in the HSDPA FMCS set and according to HSDPA HOPS
parameters ReleaseMarginAverageEcNo and ReleaseMarginPeakEcNo 566
Company Confidential
Serving Cell Change via cell_FACH RAS05
• Serving Cell Change switches
The conditions preventing HSDPA allocation in the user from HS-DSCH to
HSPDA coverage area to HSDPA capable UE Cell_FACH
are: • Cell Reselection is done in FACH
• More than one serving cell (Soft Handover)
• Multi-RAB combination (not supported)
through Cell Update
• Lack of HSDPA capacity in the cell • HS-DSCH is selected in the new
cell
Cell B
Cell A Switching to UE moves to
the target cell Switching
Service Cell_FACH
by making to HSDPA
in when Cell B is
cellUpdate Cell B
HSDPA “x”dB stronger
Throughput
HSDPA
Throughput
Download
complete
HSDPA
0
568
Company Confidential
Resumption timer (RAS05+CD1)
The conditions preventing HSDPA allocation in
• HSDPA resumption timer HSPDA coverage area to HSDPA capable UE
switches the user from DCH to are:
• More than one serving cell (Soft Handover)
DSCH when UE enters
HS-DSCH, • Multi-RAB combination (not supported)
HSDPA area • Lack of HSDPA capacity in the cell
Throughput
Download
complete
HSDPA
0
569
Company Confidential
Mobility: Channel type switching
• Event 1A
– If EnableRRCrelease =0 (disable) [HSDPAHopsIdentifier]
• Radio Bearer Reconfiguration to FACH
– If EnableRRCrelease =1 (enable)
• The transition from HS-DSCH to FACH is not triggered directly. Instead,
the UE is instructed to start periodic CPICH Ec/Io measurements. The
RNC parameters ReleaseMarginAverageEcNo (default 2.5dB) and
ReleaseMarginPeakEcNo (default 3.5dB) [HSDPAHopsIdentifier] are
used to trigger the channel type switching to FACH when
– CPICHEcNoneigh>ReleaseMarginAverageEcNo + CPICHEcNoserving or
– CPICHEcNoneigh>ReleaseMarginPeakEcNo + CPICHEcNoserving
• Events 1F and 6A
– Radio Bearer Reconfiguration to DCH 0/0 kbps is allocated but the ISHO/IFHO
measurements are not started immediately. The UE is configured with the
DCH FMCS parameters and measurements are started accordingly.
570
Company Confidential
Addition Window (HSDPA FMCS) = 0dB
Addition Time (HSDPA FMCS) = 1280ms
Ec/No HSDPA AdditionWindow (0 dB) Release Margin Peak EcNo (HSDPA HOPS) = 3.5 dB
EcNo Averaging Window (HSDPA HOPS) = 8
DCH ReleaseMarginAverageEcNo
AdditionWindow ReleaseMarginPeakEcNo
P CPICH 2
P CPICH 1
DPCH
• Since MAC-hs in BTS, HS-DSCH does HSDPA
not support Soft Handover. Softer not
supported either.
572
Company Confidential
RAN05.1
Mobility: RAN05.1 Serving Cell Change
• Following reasons can trigger serving cell change:
– Periodical CPICH Ec/No measurement
– Periodical UL SIRerror measurement
• UL SIR error serving cell < ULSIRerrorServCell (-3dB default)
– Measurement event 1B
• Serving HS-DSCH cell is removed from the active set.
– Measurement event 1C
• Serving HS-DSCH cell is replaced with another cell.
– RL failure of the serving HS-DSCH radio link (loss of UL synchronisation)
• Serving HS-DSCH radio link is deleted.
– RL removal due to Rx-Tx time difference
• Serving HS-DSCH radio link is deleted.
573
Company Confidential
Mobility: RAN05.1 Serving Cell Change triggered by
CPICH EcNo
CPICH EcNo RAN05.1
cell2
cell3
HS- DSCH Serving Cell Change initiates
after events A and B
HSDPASevCellWindow (2dB)
A. CPICHEcNoservingcell<=CPICHEcNo best cell
HSDPA CPICH EcNo Threshold (-5dB)
– HSDPAServCellWindow
B. CPICHEcNoservingcell<HSDPACPICHEcNothr
eshold
cell1
HSDPA cell change min interval (3s)
A B t
575
Company Confidential
HSDPA dimensioning overview
Throughput depends on
the signal to interference
ratio.
UL coverage: additional
margin due to CQI and Ack/Nack
Min. throughput at
the cell edge
UL traffic on DPCH
Can be used for simple RFQ Can be used for more complex
dimensioning cases, where analysis and for pre-planning.
the main goal is equipment
configuration based pricing.
577
Company Confidential
Method 1
578
Company Confidential
Air interface – shared carrier (1/2)
Assumptions:
• Macro cell
• Veh-A @ 3kmph
• RR scheduling
• 5codes, QPSK/16QAM at
both Node-B/UE.
7-8 W power for HSDPA • 1 WSPC/cell for HSDPA
provides good
• Cell throughput without
compromise between HSDPA is 780Kbps (DCH
HSDPA bit rates and R99 traffic).
capacity • Using 7-8 W for HSDPA, the
total cell throughput is
increased by a factor
1150/780=1.47
• Note: 16QAM brings only
about 4% gain to avg. cell
throughput compared to
QPSK in macro cell
environment. For this reason
these curves can be used for
both QPSK and 16QAM
cases.
579
Company Confidential
Air interface – shared carrier (2/2)
Assumptions:
• Macro cell
• Veh-A @ 3kmph
• PF scheduling
• 5 codes, QPSK/16QAM at
both Node-B/UE.
• 1 WSPC/cell for HSDPA
• Cell throughput without
HSDPA is 780Kbps (DCH
traffic).
• Using 7-8 W for HSDPA,
the total cell throughput is
increased by a factor
1320/780=1.69
• Note: Average HSDPA
traffic must be considerable
(min. 3 simultaneous users)
to achieve gain due to PF
scheduling. If the avg.
number of simultaneous
HSDPA users is 1-2 per cell
then results in previous
slide should be used even
the PF feature would be
enabled in the network.
580
Company Confidential
Results with HSDPA-only Traffic
5 HS-PDSCH codes Vehicular-A Pedestrian-A
10W HSDPA Power
Due to using PF Due to using PF
36%
instead of RR instead of RR Gain
Due to 16QAM
Due to 16QAM 27% Gain
14% Gain
4% Gain
RR RR PF RR RR PF
QPSK only QPSK/16QAM QPSK/16QAM QPSK only QPSK/16QAM QPSK/16QAM
581
Company Confidential
Method 2
582
Company Confidential
SINR and throughput mapping – Simulation Result
HS-DSCH POWER 7W (OF 15W), 5 CODES,
• The single-user 1RX-1TX, 6MS/1DB LA DELAY/ERROR
4.0
HSDPA throughput
Service HSDPA384
Transmitter - Node B
Max Tx Power (HSDPA) 12 W
Max Tx Power (HSDPA) 40.8 dBm
Tx Antenna Gain 18 dBi
Cable Loss 3 dB
EIRP 55.8 dBm
Receiver - Handset
Handset Noise Figure 8 dB
Thermal Noise -108 dBm
Background RSSI -100 dBm
Downlink Load 80 %
Interference Margin 7.0 dB
Interference Floor -93.0 dBm
Min SINR 4.5 dB PHS−PDSCH_ tx = PtxMaxHSDPA − PHS−SCCH_ tx
Service PG 12.0 dB
Rx Antenna Gain 0 dBi
Body Loss 0 dB SINR TO MATCH THE EDGE BITRATE
Receiver Sensitivity -100.6 dB
DL Fast Fade Margin 0 dB
DL Soft Handover Gain 0 dB Service PG = 10 ∗ log10 (16) ; SF = 16
Max. Path Loss 156.3 dB
Service HSDPA384
Transmitter - Node B
Max Tx Power (HSDPA) 12 W
Max Tx Power (HSDPA) 40.8 dBm
Tx Antenna Gain 18 dBi
Cable Loss 3 dB
EIRP 55.8 dBm
Receiver - Handset
Handset Noise Figure 8 dB
Thermal Noise -108 dBm
Background RSSI -100 dBm
Downlink Load 80 %
Interference Margin 7.0 dB
Interference Floor -93.0 dBm
Min SINR 0 dB PHS−PDSCH_ tx = PtxMaxHSDPA − PHS−SCCH_ tx
Service PG 12.0 dB
Rx Antenna Gain 0 dBi
Body Loss 0 dB
Receiver Sensitivity -105.1 dB Minimum requirement for SINR
DL Fast Fade Margin 0 dB
DL Soft Handover Gain 0 dB
Max. Path Loss dB
160.8
C = SINR− SF16 + I = SINR− SF16 + InterferenceFloor
DPCCH
⎛ ⎛⎜ ∆HS−20DPCCH ⎞⎟ ⎞ ∆
∆ HS −DPCCH = 6 10⋅ log10 ⎜10⎝ ⎠⎟
= HS−DPCCH = 10⋅ log10 (30 / 15) = 3
⎜ ⎟ 2
⎝ ⎠
• The power offset parameters [DACK; DNACK; DCQI] are controlled by the RNC and
reported to the UE using higher layer signalling. The power offset can be up to 6
dB within 9 possible quantization steps, see the tables in 25.213.
• The power offsets [DACK; DNACK; DCQI] are radio network planning parameters that
should be selected to fulfill the required UL HS-PDCCH detection probabilities.
586
Company Confidential
UL link budget: HS-DPCCH overhead Summary
587
Company Confidential
HSDPA BTS Configuration in RAN05
• WSPC with HSDPA SW card required
• All BTS models/configurations, that can
accommodate at least one WSPC,
support HSDPA.
4 users
• One WSPC unit supports max. 3 HSDPA
cells.
• One WSPC unit supports max. 16
8 users 4 users HSDPA users. 1)
• Those 16 HSDPA users can be divided
freely between three cells.
• Packet Scheduler in MAC-hs schedules
all users equally regardless of their cell.
-> Cells get resources depending on the
amount of users and traffic in each.
• Max. one HSDPA WSPC in BTS
-> Max. three HSDPA cells in BTS.
Note 1: The actual amount of supported HSDPA users per
WSPC depends heavily on used Uplink DCH channel speed
associated with HSDPA user, see dimensioning rules
588
Company Confidential
HSDPA activation : Channel Allocation in WSPC
WSP type CEs
• Channel Elements (CEs) are used as a
medium to explain resources (processing WSPA 32
power) on WSP cards and Node B.
• Only WSPC supports HSDPA WSPC 64
= R5 HSDPA
Traffic channels
= R99 DCH
590
Company Confidential
HSDPA activation : Channel Allocation in WSPC
• The CE reserved in the WSPC for
HSDPA are 32
WSPC 1 WSPC 2
• The WSPC for the associated DCH is DL UL DL UL
selected as for any DCH (Node B sees 1 1
2 2
no difference). 3 3
4 4
• MAC-d flow and DCH of the same UE 5 8 9 10 11 9 9 9 9 5
can be under different WSPC and/or 6 12 13 14 15 10 10 10 10 6
WAM. 7 16 11 11 11 11 7
8 1 1 1 1 12 12 12 12 8
9 1 1 1 1 16 16 16 16 9 1 2 3 4 1 1 1 1
10 1 1 1 1 14 14 14 14 10 5 6 7 2 2 2 2
11 1 1 1 1 15 15 15 15 11 3 3 3 3
12 2 2 2 2 16 16 16 16 12 4 4 4 4
Associated dedicated channels are 13 2 2 2 2 1 1 1 1 13 5 5 5 5
required for HSDPA users: 16 14 2 2 2 2 2 2 2 2 14 6 6 6 6
15 2 2 2 2 15 7 7 7 7
kbps DL for uplink power control 16 16 8 8 8 8
and 64…384 kbps UL channel for
Common channels (for 3 carriers)
TCP ACK’s and uplink user data.
HSDPA 5 codes
592
Company Confidential
WSPC capacity, example
INPUTS:
If the Rel99 traffic requires 30 CE in DL and 20 in UL (SHO OH included).
The average number of HSDPA users is assumed to be 5 and associated UL
traffic is carried on 64kbps bearer.
One WSPC card is allocated for HSDPA. BTS configuration is 1+1+1.
RESULT:
• DL CEs = 30 +5 and UL CEs= 20+4*5=40
• For this capacity two WSPC cards is required (or WSPC + WSPA/D).
593
Company Confidential
Pre-planning with NetAct Planner
• Previous version of NetAct Planner (V4.2) can be used for simple HSDPA
coverage planning by using Pilot Ec/Io –> SINR -> HSDPA throughput
mapping tables.
594
Company Confidential
Pre-planning with NetAct Planner V4.2
• Static Pilot Ec/Io can be mapped to SINR when we know:
– Pilot power
– HSDPA power
– Total BTS power
– DL orthogonality
• SINR vs. throughput lookup table gives the HSDPA throughput
• Draw back is that requires that all BTSs in the analysis area have similar
power assumptions/parameters.
596
Company Confidential
Pre-planning with NetAct Planner V4.2
• Example for 5 codes with 16QAM:
– BTS total power 14W
– Pilot power 2W
4 W HSDPA power 6 W HSDPA power
Pilot HSDPA HSDPA, 5codes HSDPA HSDPA, 5codes
Ec/Io SINR Throughput SINR Throughput
-15 0.56 167 2.32 249
-14.5 1.13 193 2.89 279
-14 1.70 220 3.46 316
-13.5 2.29 248 4.05 354
-13 2.89 279 4.65 393
-12.5 3.50 319 5.26 432
-12 4.14 359 5.90 479
-11.5 4.79 401 6.55 540
-11 5.47 445 7.23 604
-10.5 6.18 506 7.94 671
-10 6.92 576 8.68 745
-9.5 7.72 650 9.48 865
-9 8.57 731 10.34 995
-8.5 9.51 871 11.27 1137
-8 10.57 1030 12.33 1278
-7.5 11.78 1208 13.54 1437
-7 13.26 1400 15.02 1688
-6.5 15.20 1725 16.96 2092
-6 18.23 2415 19.99 2871
-5.5 28.23 3619 29.99 3619 597
Company Confidential
HSDPA planning process
1. Monitoring and analyzing existing R99 network Monitoring and analyzing
performance existing R’99 network
performance
• Average used BTS power
• Average BTS WSP card utilization
• SHO overhead
2. Dimensioning, pre planning and deployment strategy HSDPA performance targets
– BTS power resources for HSDPA and deployment strategy
– HSDPA strategy: hot spot or whole area coverage
– Dedicated or shared carrier
3. Parameter planning
Detailed BTS andIub
– Priority parameter configurations and RNC
– Power thresholds planning
– Shared carrier
• Power allocation for HSDPA
– Dedicated carrier
HSDPA parameter planning
• Directed RRC in cell/cluster level
– Mobility
• SHO parameters
4. Performance monitoring
– Counters
Performance monitoring
– Online monitoring
598
Company Confidential
HSDPA Deployment Strategy - Method 1
Given DCH traffic
requirements assigns
to HSDPA the power
that is not used for
DCH and obtains the
HSDPA cell
throughput and
WBTS configuration
(number of WSPC,
number of carriers)
599
Company Confidential
HSDPA Deployment Strategy - Method 2
600
Company Confidential
Site/cell design possibilities
• Hotspot configuration
– HSDPA is set up in an isolated cell where HSDPA users are more likely to
be present
– In most cases a WSPC can be allocated to handle the (unique) HSDPA cell.
– Dual carrier configuration can be enabled, the second carrier mostly
dedicated to HSDPA users
601
Company Confidential
General HSDPA deployment
Phase 1: Low DCH and HSDPA traffic
– A single carrier shared between DCH and HSDPA.
– Nokia RAN´05 HSDPA solution allows to choose whether prioritize Non-Real-Time DCH users over HSDPA users. In
any case Real-Time CS users (Voice, Video) are prioritized over PS Non-Real-Time users.
– Exact resource (BTS Output Power,Channel elements, RNC resources) allocation between HSDPA and DCH users
can be derived from HSDPA throughput requirements.
– Additionally HSDPA hotspots can be created (Indoor, Microcells) to handle specific, high traffic areas (Company
buildings).
Phase 2: Medium to high RT DCH traffic, medium NRT DCH and HSDPA traffic.
– Second carrier added on high traffic sites.
– Based on traffic predictions and network KPIs measurements, additional carriers can be set up on specific sites or at
cluster level.
– In case of multiple carriers, HSDPA users can be directed to a dedicated or semi dedicated carrier.
– Additionally HSDPA hotspots can be created (Indoor, Microcells) to handle specific, high traffic areas (Company
buildings).
602
Company Confidential
Scenario 1: Carrier shared between HSDPA and R99
603
Company Confidential
Scenario 2 : Dedicated carrier to R99 UEs + Shared
carrier between R99 and HSDPA UEs
• UE moved by RRC connection setup according to
UE release R5 in RRC Connection Setup
message
• Recommended for hotspot, indoor, major
customer office and also in mature network
☺Absorb more traffic for rel99
f2 ☺Less impact to R99 than first solution
f2 = R5 HSDPA+R5 DCH
☺More capacity and more cell throughput
f1
f1
= R99 DCH ☺Additional WSPC card could be required, but
this will avoid the requirement for a
subsequent visit for a subsequent capacity
Main Planning parameters: upgrade
Cost
• HSDPA priority over NRT
DCH More WSP&Iub&DMPG? overhead from 2nd
carrier activation
• PtxHSDPAmax
Additional HW required
• Directed RRC Setup for R5
More complex parameter planning (DRRC,
UE capability, traffic balancing)
604
Company Confidential
Directed RRC setup for HSDPA layer
• All idle mobiles are forced to camp on f1
• Any UE reporting Rel’5 is directed to HSDPA layer, others to Rel’99 layer
Rel’5
RRC setup on f2
f2, HSDPA
All idle
R’99
UEs camp
RRC setup onto f1
on f1
f1, Rel’99
605
Company Confidential
Directed RRC setup for HSDPA layer
DirectedRRCEnabled
Enables / disables the use of Directed RRC connection setup or Directed RRC
connection setup for HSDPA layer in the cell.
If this parameter value is 2 then Directed RRC Connection Setup from a cell that
doesn’t support HSDPA to a cell belonging to the same sector but on different
frequency that supports HSDPA is activated for all UE accessing the service that
are Release 5 (Access Stratum Release Indicator = R-5 in the RRC Connection
Request Message).
606
Company Confidential
Directed RRC setup for HSDPA layer
• There must not be more than two layers. This is checked from SectorID (no
more than 2 same SectorIDs per BTS exist in the cells where
DirectedRRCEnabled=2).
608
Company Confidential
HSDPA within the Indoor Environment
Differences between DAS and Nokia AIR Solution
• DAS - Currently most commonly used indoor solution is coaxial cable based
Distributed Antenna System. Any Nokia BTS can be used to provide DAS
solution.
• AIR - Nokia Advanced Indoor Radio - is fibre optical based active distributed
antenna system, which can be integrated within Nokia Ultrasite WCDMA BTS.
DAS AIR
Cable distribution Coaxial cable based Optical cable based
Supported, but not recommended- need to
Diversity duplicate all DAS component (expensive) Supported
System Sharing Can be used to share different technologies Support only WCDMA
All DAS components, except BTS, are passive
Maintenance & Power so no maintenance are required Active components
Base Station Any Nokia BTS can be used with DAS Currently only Ultrasite
HSDPA Supported Supported
609
Company Confidential
RF Carrier Assignment for Indoor Solutions
• The design of an indoor solution should start with measurement of the existing
macrocell coverage
• Strong macrocell coverage will degrade the indoor solution SINR
• If macrocell coverage is strong then there may be a requirement to assign a
dedicated RF carrier to the indoor solution
Benefits Drawbacks
RF Carrier shared Soft handover may be used Soft handover may trigger channel type
with macrocells to avoid hard handovers switching from HS-PDSCH to DPCH
SINR is likely to be Hard handover required
Dedicated RF Carrier
greater
• The indoor solution can be assigned 2 RF carriers: 1st RF carrier for DPCH
connections and 2nd RF carrier for HSDPA connections
• In the case of active DAS, assigning 2 RF carriers will reduce the maximum
downlink transmit power per carrier
610
Company Confidential
Indoor Solution Link Budget
• The indoor solution HSDPA link budget should account for the
– low downlink transmit power radiated by the DAS
– high orthogonality reducing the level of own cell interference
– lack of soft handover gain
– no fast fade margin due to no power control
– reduced slow fade margin
– potentially high level of intercell interference from the macrocell layer
• The HSDPA link budget should be completed in parallel to the DPCH link
budget
• The average HSDPA throughput should be evaluated based upon link budget
results and the level of intercell interference from the macrocell layer
• It may be necessary to increase the number of indoor solution antennas to
improve the HSDPA SINR conditions
PHSDPA
• SINR conditions should SF16 ×
be evaluated for a range C SINR
SINR = PathLoss
of distances from the =
I SF16 ⎛ PBS ⎞
indoor solution antennas ⎜ ⎟ × α + I Other + PNoise
⎝ PathLoss⎠
611
Company Confidential
HSDPA Parameters for the Indoor Environment
• The HSDPA parameter set for the indoor environment can be the same as that
for the macrocell environment
• The transmit power assigned to HSDPA must be sufficient to achieve adequate
SINR and downlink throughput
• The HSDPA priority should be configured to maintain the HSDPA connections
rather than the DPCH connections
612
Company Confidential
3GPP TSG RAN references
• WG1 specifications
TS 25.201 Physical layer - General description (Release 5)
TS 25.211 Physical channels and mapping of transport channels onto physical channels (Release 5)
TS 25.212 Multiplexing and channel coding (Release 5)
TS 25.213 Spreading and modulation (Release 5)
TS 25.214 Physical layer procedures (Release 5)
• WG2 specifications
TS 25.306 UE Radio Access capabilities (Release 5)
TS 25.308 High Speed Downlink Packet Access (HSDPA) Overall description; stage 2 (Release 5)
TS 25.321 Medium Access Control (MAC) protocol specification (Release 5)
TS 25.331 Radio Resource Control (RRC); protocol specification (Release 5)
• WG3 specifications
TS 25.423 UTRAN Iur Interface RNSAP Signalling (Release 5)
TS25.425 UTRAN Iur Interface user plane protocols (Release 5)
TS 25.433 UTRAN Iub Interface NBAP Signalling (Release 5)
TS 25.435 UTRAN Iub Interface User Plane Protocols for Common Transport Channel Data Streams
(Release 5)
• WG4 specifications
TS 25.101 UE Radio Transmission and Reception (Release 5)
• Technical Reports
TR 25.848 Physical layer aspects of UTRA High Speed Downlink Packet Access (Release 4)
TR 25.855 High Speed Downlink Packet Access; Overall UTRAN Description (Release 5)
TR 25.950 UTRA High Speed Downlink Packet Access (Release 4)
613
Company Confidential