7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 1/359
UMTS Architecture Overview
by Dr Paul Raby
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 2/359
Public Land Mobile Network (PLMN)
▪ A PLMN can be regarded as anindependenttelecommunications entity.
▪ A PLMN is defined as:
▫ One or more switches with:
▪ a common numbering plan
▪a common routing plan▫ Switches act as the interface to
external networks
▪ The PLMN can be separatedinto
▫ Core Network
▫ Access Network
Core Network
Access Network
PLMN
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 3/359
User Equipment
UMTSTerrestrial
Radio AccessNetwork
Core Network
UU IU UE UTRAN CN
UMTS High Level Architecture
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 4/359
Major Network Elements in UMTS
PLMN,PSTN,ISDN
Internet,X25
PacketNetwork
UU
UE
CU
USIM
ME
MobileEquipment
UMTSSIM
CN
MSC/VLR
SGSN GGSN
GMSC
HLR
Serving GSN Gateway
GSN
GatewayMSC
Mobile SwitchingCentre
Home LocationRegister
IU
UTRAN
IUb
IUr
Node B
Node B
Node B
Node B
RNC
RNC
Radio NetworkController
Radio NetworkController
Iu-ps
Iu-cs
IUb
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 5/359
General UTRAN Architecture
UU IU
UE
UTRAN
IUb
IUr
Node B
Node B
Node B
Node B
RNC
RNC
Radio NetworkController
Radio NetworkController
Iu-ps
Iu-cs
IUb
CN (MSC)
CN (SGSN)
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 6/359
Elements of UTRAN
▪ Radio Network Controller
▫ Owns and controls radio resources in its domain (BSC in GSM)
▫ Service Access Point for all services that UTRAN provides for the CN
▫ Note: Service RNC (SRNC) and Drift RNC (DRNC) are subsets
▫ Note: Control RNC (CRNC ) whichever RNC is talking to the UE
▪ Node B
▫ Acts as the radio base station (BTS in GSM)
▫ Converts the data flow between the Iub and Uu interfaces
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 7/359
Major Interfaces in UMTS
▪ There are four major newinterfaces defined in UMTS
▫ Iu
▪The interface between UTRANand the CN
▫ Iur
▪The Interface between differentRNCs
▫ Iub
▪The interface between theNode B and the RNC
▫ Uu
▪The air interface
RNC
Node- B
RNC
UE
CN
Uu
Iu
Iub
Iur
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 8/359
UMTS Interface Implementation
ATM/IP Network
RNC Node B
Node B
Node B
MSC
RNC
SGSN
Node B
Iub Iu_cs
Iu_ps Iur
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 9/359
Handover in UMTS
▪ There are 3 basic types of handover▫ Intra frequency handovers
▪ Handovers between 2 UMTS carriers at thesame frequency
▪ These can be soft handovers
▫ Inter frequency handovers ▪ Handovers between 2 UMTS carriers at different
frequencies
▪ These are hard handovers
▫ Inter system handovers
▪ Handovers between UMTS and GSM carriers
▪ These are hard handovers
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 10/359
Macrodiversity between Node B’s
▪ If Active Set consists of
▫ two connections to cellsparented to different Node Bs
▫ then the combining of the twochannels occurs at the RNC
▪ This is known as a soft handover
▪ This doubles thetransmission „cost‟ of thecall
RNC
Node B
Cell
Cell
Cell
Node B
Cell
Cell
Cell
Iur
Iu
Uu
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 11/359
Maximal Combining between Cells on theSame Node B
RNC
Node B
Cell
Cell
Cell
Node B
Cell
Cell
Cell
Iur
Iu
Uu
▪ If Active Set consists of
▫ two connections to cellsparented to the same Node B
▫ then the combining of the twochannels occurs at the NodeB
▪ This is known as a softer handover
▪ This has no transmissionimplication (but does havecapacity implications) ifcells are collocated.
▪ Uses maximal combining
▫ Adds electrically the
signals making one betterthan each individual
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 12/359
Architecture of a UMTS bearer service
TE TE UE UTRAN CN
edge node
CNgateway
End-to-End
TE/UE LocalBearer UMTS Bearer External Bearer
Radio Access Bearer CN Bearer
Radio Bearer Iu Bearer Backbone Network
UTRA FDD/TDD Physical Bearer
Each bearer service on a specific layer provides services using layers below.
System Architecture Overview
S A hi O i
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 13/359
UMTS Protocol Stratums
▪ Non-access StratumEncompasses layers 4 to
7 of the OSI 7 layermodel, and the upper partof layer 3
N
onA c c e s s S t r a t um
A c c e s s S t r a t um
L1 L1 L1L1
L2L2L2L2
L5L5
L4L4
L6 L6
L7 L7
L3 lower L3 lower L3 lower L3 lower
L3 upper L3 upper
Uu Iu UE UTRAN CN
System Architecture Overview
• Access Stratum
→ Encompasses layers 1and 2 of the OSI 7 layermodel and the lower part oflayer 3
S t A hit t O i
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 14/359
UMTS QoS Classes
▪ Conversational ▫ Speech over CS bearer
▫ Voice over IP, PS bearer
▫ Delay critical, imposed by human perception
▪ Streaming ▫ Multimedia streaming
▫ Using buffers, for non-real time delivery; real-video, real- audio
▪ Interactive ▫ Web browsing, database retrieval
▫ Round trip delay time is a key parameter
▪ Background ▫ E-mail
▫ Delay:- 10s of seconds or even minutes
System Architecture Overview
S t A hit t O i
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 15/359
Protocol Model for UTRAN Interfaces
▪ Protocol structures in UTRAN are designed in layers and planes.
▪ They are seen as logically independent of each other
▫ However they will physically interact.
▪ Being logically independent allows for changes to blocks in the
future
theoretically!
System Architecture Overview
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 16/359
General Protocol Model forUTRAN Terrestrial Interfaces
System Architecture Overview
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 17/359
Horizontal Layers in the General Protocol Model
▪ All UTRAN related issues are only visible in the Radio
Network Layer
▪ The Transport Layer simply represents standard transport
technology for use in UTRAN
▫ e.g. ATM and appropriate ATM Adaptation Layers
▪ AAL2 ( voice ) and AAL5 ( data/control)
▫ UDP/IP or RTP/UDP/IP ( release 6 ? )
System Architecture Overview
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 18/359
Vertical Planes in the General Protocol Model
▪ The Control Plane is provided for all UMTS
specific control signalling including:
▫ Application Protocols
▫ Signalling Bearers
▪ The User Plane is provided for all data sent and
received by the user including:
▫ Data Streams
▫ Data Bearers
System Architecture Overview
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 19/359
Vertical Planes in the General Protocol Model
▪ The Transport Network Control Plane also includes the Access
Link Control Application Part, ALCAP.
Transport Network User
Plane
Transport Network User
Plane
Transport Network Control
Plane
ALCAP
System Architecture Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 20/359
UMTS Technology Overview
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 21/359
CDMA - Direct Sequence Spread Spectrum
UMTS Technology Overview
f r e q u en c y
time
code
Frame Period (we may still needframes/timeslots for signaling)
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 22/359
CDMA Spreading
•Essentially Spreading involves changing the symbol rate on the air interface
Identical codes
Tx Bit Stream
P
f
Code Chip Stream
Spreading
P
f
Channel
Air InterfaceChip Stream
P
f
Code Chip Stream
Despreading
P
f
Rx Bit Stream
P
f
UMTS Technology Overview
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 23/359
Spreading and Despreading
Rx Bit Stream
Air InterfaceChip Stream
Tx Bit Stream 1
-1
Code Chip Stream
XSpreading
Code Chip Stream
X Despreading
UMTS Technology Overview
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 24/359
Spreading and Despreading with code Y
Air InterfaceChip Stream
Tx Bit Stream 1
-1
Code Chip Stream
XSpreading
X Despreading
UMTS Technology Overview
Code Chip Stream Y
Rx Bit Stream
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 25/359
Spreading
▪ If the Bit Rate is Rb, the Chip Rate is Rc, the energy per bitEb and the energy per chip Ec then
▪ We say the Processing Gain Gp is equal to:
▪ Commonly the processing gain is referred to as theSpreading Factor
b
ccb
R
R E E
b
c p
R
RG
UMTS Technology Overview
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 26/359
Spreading in noise
▪ The gain due to Despreading of the signal over widebandnoise is the Processing Gain
Signal
P
f
Spreading Code
Tx SignalP
f
Rx Signal (= Tx Signal + Noise)
f
P
Channel
Wideband Noise/Interference
P
f
Spreading Code Signal
P
f
gy
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 27/359
Visualising the Processing Gain
Signal
Intra-cell Noise
Inter-cell Noise
W/Hz
Before Spreading
f
W/Hz
After Spreading
f
W/Hz
Ec
Io With Noise
f
W/Hz
After Despreading /Correlation
f
W/HzEb
No
Post Filtering Orthog = 0
f
dBW/Hz
Eb
No
Eb /No
f
Eb
No
W/HzPost Filtering Orthog > 0
f
Eb
No
Eb /No dBW/Hz
f
gy
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 28/359
Separating BaseStations
▪ Summarising:
▫ Channelisation Codes
▪Are used to separate channelsfrom a single cell
▫ Scrambling Codes
▪Are used to separate cells fromeach other rather than purelychannels
▪ Different base stations will use thesame spreading codes withseparation being provided by theuse of different scrambling codes.
S1
S2
S3
C1 C2 C3
C1 C2 C3
C1 C2 C3
gy
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 29/359
Separating UE’s
▪ Summarising:
▫ Spreading/Channelisation Codes
▪Are time dependent and so are used in the UL to spread the signalbut not to separate the UE‟s
▫ Scrambling Codes
▪Are used to separate UEs from each other rather.
gy
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 30/359
Spreading Codes = Channelisation Codes
▪ Channelisation codes are orthogonal
▫ Which provides channel separation
▪ Number of codes available is dependant onlength of code
▪ Channelisation codes are used to spread thesignal
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 31/359
Channelisation Code Generation
▪ Channelisation codes can be generated from a Hadamard matrix
▪ A Hadamard matrix is:
▪ Where x is a Hadamard matrix of the previous level
▪ For example 4 chip codes are:
▫ 1,1,1,1
▫ 1,-1,1,-1
▫ 1,1,-1,-1
▫ 1,-1,-1,1
x x
x x
Note: These two codes correlateif they are time shifted
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 32/359
▪ Orthogonal Variable Spreading Factor Codes can
be defined by a code tree:
▫ SF = Spreading Factor of code (maximum 512 forUMTS in the DL, 256 in the UL)
SF = 1 SF = 2 SF = 4
Cch,1,0 = (1)
Cch,2,0 = (1,1)
Cch,2,1 = (1,-1)
Cch,4,0 =(1,1,1,1)
Cch,4,1 = (1,1,-1,-1)
Cch,4,2 = (1,-1,1,-1)
Cch,4,3 = (1,-1,-1,1)
OVSF codes
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 33/359
Vital Parameters
▪ Ec/Io of the Pilot Channel is used to
▫ estimate (“sound”) the channel (multipath characteristics)
▫ decide which server is “best server”
▫ make handover decisions
▫ Typical requirement > -15 dB
▪ Eb/No in both uplink and downlink affects error ratios.
▫ Typical requirement 1 to 10 dB
▫ Required value of Eb/No depends on propagationconditions and sophistication of receiver.
▫ This is your Quality Measure
▪ Noise rise limits path loss and coverage.
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 34/359
Rake Receiver
Correlator
Code Generators
(S & C)
ChannelEstimator
Phase Rotator Delay Equalizer
Matched Filter
I
Q
I
Q
A typical rake receiver withthree fingers
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 35/359
RAKE Receiver
▪ Auto-correlation function of PN sequence is used toproduce multipath estimate of propagation path.
▪ Each finger then acts as separate receiver to provide theoptimum signal
Autocorrelation
-0.5
0
0.5
1
0 1 2 3 4 5 6 7
Delay
A v e r a g e v a l u e
Direct Component Delayed Components
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 36/359
Wideband Implications
▪ Autocorrelation function of PN sequence is 1 for zero delay and zerofor all delays outside a chip period.
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 37/359
Multipath Situation
Autocorrelation can be processedto provide a channel estimation.
Tx
Rx
+
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 38/359
Resolution of Multipath
Chip Period = 0.26 microsecondsCorresponding path length difference = 78 metres
This indicates the sort of resolution possible with theUMTS Rake receiver
Tx
Rx
excel
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 39/359
Noise Rise
▪ The effective noise floor of the receiver increases as the
number of active mobile terminals increases.
▪ This rise in the noise level appears in the link budget andlimits maximum path loss and coverage range.
Background Noise
Three Users
One User
Two Users
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 40/359
Self Assessment Questions
▪ What is the Processing Gain (in dB) if a UMTS
system utilising a chip rate of 3840 kbps is used
with the following user rates?
i) 12.2 kbps
ii) 64 kbps
iii) 128 kbps
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 41/359
Self Assessment Questions
Solution:
dB98.2412200
3840000log10
3840000log10Gain
10
10
b R
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 42/359
Self Assessment Question
▪ If a 12200 bps voice channel has a
SNR of -16 dB. Determine Eb/No.
Processing Gain = 25 dB
Eb/No = 25 - 16 = 9 dB
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 43/359
Self Assessment Questions
▪ A mobile receives a 12200 bps voice channel from a base
station at a level of –106 dBm. The base station is transmitting
12 additional identical voice channels plus a pilot channel and
common channel that are both received with power levels of –
104 dBm. The background thermal noise level of the mobile is –99 dBm.
Determine the value of Eb/No at the output of the receiver for
the following orthogonality factors.
i) 0 ii) 0.5 iii) 1.0
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 44/359
Example Solution
W
W
W
13-
14-
13-
103.81Total
107.96dBm-1042
103.01dBm-10612
common)andpilot(incchannelsotherfromPowerTotal
Taking Orthogonality to be 0.5:
W 13-101.9PowerEffectiveModified
UMTS Technology Overview
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 45/359
Solution
W 13-101.9PowerEffectiveModified
W 13-101.26dBm99-PowerNoiseThermal
dBm95
10.163PwrEff.ModifiedNoiseThermal 13-
W
dB11(-95)--106SNR
dB14N
E
dB52GainProcessing
0
b
ebno
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 46/359
The Link Budget
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 47/359
The Link Budget
▪ GSM and UMTS compared
▫ More thermal noise in UMTS systems. KToB ~ -108.1 dBm
▫ Processing Gain (dBs) in UMTS
= 10 log (3840000/User Rate (bps))
▫ Power Control Margin for imperfect Fast Fading Power
Control must be considered in UMTS systems
▫ Interference Margin for Noise Rise must be considered in
UMTS systems
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 48/359
Recovering the Wanted Signal
jtotal
j
j
b
jtotal
j
P I
P
R
W
N
E
P I
PSNR
0
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 49/359
Recovering the Wanted Signal
total j
j jb
total
j
j
b
j
j
btotal
j
j
j
b
jtotal
j
I L
R
W
E
N
I
W
R
N
E
W
R
N
E I
P
W R
N E
P I P
0
0
0
0
1
1
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 50/359
Recovering the Wanted Signal
j jb
j
R
W
E
N L
01
1
95
1
1220038400003.01
1
12200 3840000 )dB5.2( 3.00
j
j
b
L
RW E
N
Inserting typical values for a voice service:
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 51/359
Recovering the Wanted Signal
Possibility 1:
•Single user.
•Thermal Noise responsible for
94/95ths of received power.
•SNR=1/94 = -19.8 dB
User j is responsible for 1/95th of received power
Possibility 2:
•Power level so high that thermalnoise is insignificant.
•95 identical users possible.
Practical situations will lie in between these two extremes.
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 52/359
Considering Thermal Noise
Total Power = Power from Users + Thermal Noise
ReceivedPowerTotaluserswantedfromreceivedPower
1
1
1
1
budget)link (affectsRiseNoise
1
1
UL
UL M j
j j
N
total
N
total
N total
M j
j jtotal
LP
I P
I
P I L I
•This is also the equivalent to loading factor – see later
C
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 53/359
Considering Thermal Noise
Total Power = Power from Own Cell + Power from Other Cells + Thermal
Noise
Thermal Noise
Own Cell
Other Cell 323
95
New Capacity
Th N i Ri E i
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 54/359
The Noise Rise Equation
jb
jUL
M j
j j
N
total
R
W
E
N L L
P
I
0
1
1
1
1
1
1
1
If we have M identical users:
jb
M j
j j
R
W
E
N M L
01 1
jb
N
total
R
W
E
N
M P I
01
1
1 RiseNoise
Th N i Ri E i
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 55/359
The Noise Rise Equation
jb
N
total
R
W
E
N
M P
I
01
1
1
RiseNoise
W E
N R
MR
b j
j
0
1
1RiseNoise
W E
N
b
0
ghputuser throusingle
tthroughputotal1
1
Eff t f N i hb i C ll
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 56/359
Effect of Neighbouring Cells
Users in other cells cause interference.
Typical ratio of power from other cells to power
from own cell, i, is 0.6
C id i th ll
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 57/359
Considering other cells
Total Power = Power from Users + Noise from other cells + Thermal Noise
ReceivedPowerTotal
userswantedfromreceivedPower
11
1
11
1
budget)link (affectsRiseNoise
1
11
UL
UL
M j
j
j N
total
N
total
N total
M j
j
jtotal
M j
j
jtotal
i Li
P
I
P
I
P I Li I L I
Th M difi d N i Ri E ti
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 58/359
The Modified Noise Rise Equation
jb
j
UL
M j
j
j N
total
R
W
E
N Li Li
P
I
0
1
1
1 11
1
11
1
If we have M identical users:
jb
M j
j j
R
W
E
N M L
01 1
jb
N
total
R
W
E
N
i M P
I
01
11
1
RiseNoise
Th M difi d N i Ri E ti
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 59/359
The Modified Noise Rise Equation
W E
N
i
b
0ghputuser throusingle
1tthroughputotal1
1RiseNoise
i
W E
N
b
1
ghputuser throusingle
tthroughpuTotal
as RiseNoise
0
C ll C it
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 60/359
Cell Capacity
i
W E
N
b
1
ghputuser throusingle 0
is known as “pole capacity”.
i N
b E
W
10
CapacityPole
usersof numberlargeFor
Cell Capacity
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 61/359
Cell Capacity
i N
E
W
b
1 CapacityPole
usersof numberlargeFor
0
kbps853
5.0133840000 CapacityPole
0.5 (4.77dB) 3Eb/No 3840000W
i
• 50% of this would give a Noise Rise of 3 dB• ( Eb/No goes from 3 to 6 )
•50% of 853 kbps = 426 kbps
Loading Factor
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 62/359
Loading Factor
R
W
i M N
E
i N
E
W
R M
R M
b
o
b
1
1
FactorLoading
:ratedatawithusersidenticalFor
CapacityPole
ThroughputActual
FactorLoading
0
Noise Rise and Loading Factor
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 63/359
Noise Rise and Loading Factor
▪ Loading (Capacity) is linked to Eb/No value
▪ Noise Rise is linked to maximum path loss
Noise Rise Loading Factor
1 dB 20%
3 dB 50%
6 dB 75%10 dB 90%
UL 1log10RiseNoise 10
Activity Factor
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 64/359
Activity Factor
factor.activityusertheis where
1
FactorLoading
R
W
i M N E
o
b
Users are not active 100% of the time.
It is necessary to adjust the loading factor to acknowledge this fact
Downlink Considerations
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 65/359
Downlink Considerations
The Downlink benefits from orthogonality between channelisation codes.
)1(
ghputuser throusingle
CapacityPole
0
i
W E
N
b
is orthogonality factor and has a value between zero and 1.
Downlink Considerations
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 66/359
Downlink Considerations
The Downlink loading factor.
W
i N
E
o
b
1throughput
DL
maxP
Ptotal DL
Varies between approximately 20% and 75%
Uplink Budget for 144 kbps service
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 67/359
Uplink Budget for 144 kbps service
Thermal Noise: -108 dBm, Noise Figure: 4 dB, Eb/No: 1.5 dB
Processing Gain: 14 dB (10 log[3840/144])
Sensitivity -116.5 dBm
Margins: Noise Rise: 3 dB, Fast Fading: 2 dB
Antenna Gains: 20 dBi
Tx Power: 21 dBm
•Allowable Path Loss: 152.5 dB
Downlink Budget for 144 kbps service
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 68/359
Downlink Budget for 144 kbps service
Allowable Path Loss: 152.5 dB
Sensitivity -113.5 dBm
Margins: Noise Rise: 3 dB, Fast Fading: 2 dB
Antenna Gains: 20 dBi
Required Tx Power: 24 dBm per channel
kbps19115.1
41.1
3840144 CapacityPole
For 3 dB Noise Rise, capacity is halved to 955 kbps or 6
channels.
Total transmitted power required = 6 x “24 dBm” = 31.8 dBm
Uplink Budget for 8 kbps service
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 69/359
Uplink Budget for 8 kbps service
Thermal Noise: -108 dBm, Noise Figure: 4 dB, Eb/No: 5 dB
Processing Gain: 27 dB (10 log[3840/8])
Sensitivity -126 dBm
Margins: Noise Rise: 3 dB, Fast Fading: 2 dB
Antenna Gains: 20 dBi
Tx Power: 21 dBm
Allowable Path Loss: 162 dB
Downlink Budget for 8 kbps service
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 70/359
Downlink Budget for 8 kbps service
Allowable Path Loss: 162 dB
Sensitivity -123 dBm
Margins: Noise Rise: 3 dB, Fast Fading: 2 dB
Antenna Gains: 20 dBi
Required Tx Power: 24 dBm per channel
For 3 dB Noise Rise, capacity is halved to 407 kbps or 51 channels.
Total required power is 51 x “24 dBm” = 41 dBm
kbps8155.1
3840316.08 CapacityPole
Coverage vs Capacity Comparisons
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 71/359
Coverage vs. Capacity Comparisons
Coverage vs. Capacity
145.00
150.00
155.00
160.00
165.00
170.00
175.00
180.00
100 200 300 400 500 600 700 800
Throughput (kbps)
M a x i m
u m P
a t h l o s s ( d B )
Uplink
Downlink
144 kbps service 8 kbps service
Coverage vs. Capacity
145.00
150.00
155.00
160.00
165.00
170.00
175.00
180.00
100 200 300 400 500 600 700 800
Throughput (kbps)
M a x i m
u m P
a t h l o s s ( d B )
Uplink
Downlink
Capacity Issues
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 72/359
Capacity Issues
▪ Cell Throughput affects Noise Rise
▪ Noise Rise affects Link budget.
▪ Adding more cells reduces pathloss andallows for more Noise Rise and hence highercapacity.
Capacity Issues: example
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 73/359
Capacity Issues: example
▪ Network originally provides coverage at NR = 2dB
Radius of each cell is RR
If number of cells isdoubled the radius
reduces to R/ 2
Path loss will reduce by 5.3 dB.
( Assuming simple model for Path loss 137 +
35logR)
NR can increase by 5.3 dB.
Capacity Issues: example
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 74/359
Capacity Issues: example
▪ NR of 2 dB corresponds to a loading factor of 37%.
▪ NR of 7.3 dB corresponds to loading factor of 81%
▪ Each cell can now handle more than double the traffic.
▪ Doubling the number of cells has increased the capacity
by a factor of 4.4 ( 0.81/0.37 x 2 )
Note that doubling the number again would give a NR of 12.6 dB
(loading factor 95%) which would not produce such a remarkable
improvement.
( 0.95/0.81 x 2 = 2.34 this is equivalent to 10.3 x capacity of original network )
Target Eb/No Values
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 75/359
Target Eb/No Values
▪ Capacity is linked to Eb/No value
▫ Typical values: Voice 4 dB; High Speed Data 1.5 dB
▪ Lower overhead in control data for higher speed data.
▪ Also, Eb/No value assumes processing gain of 3840/12.2 for 12200bps voice. Actual transmitted data is 30000 bps.
▪ Target Eb/No controlled by RNC and will depend upon prevailing
conditions.
Single Channel Cells
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 76/359
Single Channel Cells
▪ UMTS Flexibility allows for one user to take up entire cell‟scapacity.
▪ Noise Rise is now irrelevant as there can be no other users.
▪ Base station has higher capacity making coverage uplink
limited.
▪ Asymmetry in power can be reflected in asymmetry in
throughput. e.g. 2 Mbps possible in downlink with 144 kbpsin uplink.
Asymmetric Traffic
The Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 77/359
Asymmetric Traffic
▪ Combination of (symmetric) voice and (highly asymmetric) packet data traffic
will tend to make the service asymmetric.
▪ Predicted asymmetry is 3:1 in favour of downlink.
▪ Achievable through balancing the links so that for the same path loss, downlink
has greater throughput capability.
Coverage vs. Capacity
145.00
150.00
155.00
160.00
165.00
170.00
100 200 300 400 500 600 700 800
Throughput (kbps)
M a x i m u m P a t h l o s s ( d B )
Uplink
Dow nlink
ConclusionsThe Link Budget
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 78/359
Conclusions
▪ Eb/No and Capacity intimately linked.
▪ Link budgets are affected by fast fading and interference
margins.
▪ Uplink and downlink affected differently by increased loading.
▪ Flexibility allows high data rate services to be provided.
▪ Asymmetric traffic requirements can be designed in.
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 79/359
Analysis, Prediction and Optimisation ofDownlink Capacity
The Story so Far
Coverage versus Capacity
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 80/359
The Story so Far
▫ It is possible to make “ball park” estimates of the capacity on
the downlink.
▫ The first step is to estimate a nominal pole capacity
▫ Then estimate the noise rise that can be produced at the“magic spot”.
▫ Hence deduce loading factor.
▫ This is a useful “first pass” planning calculation to perform.
▫ However, it does not consider an unevenly loaded network, nor
does it help us optimise network performance.
i N
E b 1
3840
0
Further Analysis of the Downlink
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 81/359
Further Analysis of the Downlink
▫ The concept of the “identical user”.
Identical:
•Bit Rate
•Eb/No
•Path loss
•Orthogonality
•Interference
Further Analysis of the Downlink
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 82/359
Further Analysis of the Downlink
int
11)1( R
L
Tcom
L
user
L
user N P
L
P
L
P N
L
PP
▫ Power Received by each user:
PTcom
Puser
N-1“other users”
Bit Rate
Eb/No
Path loss
Orthogonality
i L
NPP
L
user Tcom
Further Analysis of the Downlink
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 83/359
u t e a ys s o t e o
RW
P L
P L
P N P
LP N E
R L
Tcom
L
user N
Luser b
int0
111 /
▫ Eb/No delivered to each user:
Total Transmitted Power
=
Tcomuser P NP
Capacity vs Link Loss
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 84/359
Capacity vs. Link Loss
Link Loss(dB) Tx Power for25 users(dBm)
Maximumusers for 43dBm TxPower
110 37.62 64
125 37.69 64
140 39.33 49
145 41.63 33
150 45.26 16
dBm102 ;bit/s12200 dBm;36
dB;6N ;6.0 ;6.0 0
N Tcom
b
P RP
E i
Rapid, Approximate Method
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 85/359
p , pp
▫ As Puser is allowed to approach infinity:
i N E
W RN
b
10
▫ The “Pole Capacity”.
Rapid, Approximate Method
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 86/359
p , pp
▫ Identical Users will experience identical noise rise.
▫ Noise rise can be converted to throughput.
▫ We can predict the noise rise for given circumstances.
0
5
10
15
20
25
0 0.2 0.4 0.6 0.8 1 1.2
Rapid, Approximate Method
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 87/359
p , pp
iP LP
LPiP
Tcom L N
L N T
1
1max
We can predict the noise rise for given circumstances.
The maximum noise rise that can be produced is
Then, capacity is given by
L N T
TcomT
b LPiPPP
N E W PC
1.
RiseNoise11
max
max
0
Effect of Link Loss on Capacity
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 88/359
p y
There is a maximum capacity at low levels of link loss. High
transmit power allows this capacity to be approached at significantlevels of link loss.
0
200
400
600
800
1000
1200
120 130 140 150 160
Link Loss (dB)
C a p a c i t y ( k b i t / s
)
+37 dBm +40 dBm +43 dBm +46 dBm
Maximum Capacity
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 89/359
At negligible levels of link loss, the expression
for noise rise becomes
Tcom
T
P
P max
kbit/s1.
1
3840
max0
T
Tcom
b P
P
i N
E
And capacity can be estimated from
Maximum Capacity
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 90/359
If is taken to be fixed at, for example, 5,
then capacity is given byTcom
T
P
P max
kbit/s
1
3072
0i
N E b
And maximum capacity can be estimated from
kbit/s2.01
1
3840
0
i N E b
Effect of Orthogonality
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 91/359
Graph shows the effect of orthogonality on the downlink
capacity for a link loss of 145 dB and i set at 0.6.
0
200
400
600800
1000
1200
0 0.2 0.4 0.6 0.8 1
Orthogonality
C a p a c i t y ( k
b i t / s )
BTS Power: 37 dBm 40 dBm 43 dBm 46 dBm
Effect of Out-of-Cell Interference
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 92/359
Graph shows the effect of variations in the value of i on the
downlink capacity for a link loss of 145 dB and orthogonality of 0.6.
0
200
400
600
800
1000
1200
1400
0 0.4 0.8 1.2 1.6 2
Out of Cell Interference
C a p a c i t y ( k b i t / s )
BTS Power: 37 dBm 40 dBm 43 dBm 46 dBm
Extending the Validity – The Evenly-loaded Network
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 93/359
So far, “identical users” have been considered.
Consideration is now given to an evenly loaded
network.
Crucially, is there a
representative value oflink loss and out-of-cell
interference that can
be used to estimate
downlink capacity?
Extending the Validity
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 94/359
Experimentation with Monte
Carlo simulation suggests
that:
The effective value of link loss
is 4 dB less than that to the
edge of the cell.
The effective out-of-cell
interference ratio is 0.85.
Max throughput = kbit/s2458
0 N E b
Extending the Validity- An Example
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 95/359
Network of cells with link loss
to edge of 133 dB.
Maximum throughput on
downlink at Eb/No of 7 dB
is 460 kbit/s for 43 dBm
transmit power.
Note:
Pole Capacity = 613 kbit/s
If 20% of power is for
common channels then
Max throughput = 490 kbit/s
for very low linkloss
Uplink-Downlink Balance
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 96/359
Approximate Downlink Max Capacity =
Approximate Uplink Max Capacity =
Initial expectation is that loading factors will be higher
on the downlink.
Uplink Diversity and MHA will favour the uplink.
kbit/s2458
0 N E b
kbit/s2400
0 N E b
The Unevenly-loaded Network
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 97/359
The situation is complicated by the fact that different
users experience different levels of noise rise.
For example, consider the case where there are 24
voice users, split into two, equal groups.
• Link loss = 120 dB
• DL i = 0.3
• NR = 2.1 dB
• Link loss = 140 dB
• DL i = 1.0
• NR = 1.4 dB
The Unevenly-loaded Network
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 98/359
For a more general
situation, the
maximum capacity is
often determined
using a Monte Carlo
simulator on a trial
and error basis.
The Unevenly-loaded Network
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 99/359
However, if the pole capacity is estimated from
Then a single simulation result can be used to estimatethe maximum downlink capacity.
Reports from the simulation include downlink traffic
channel power and throughput.
i N E b 1
3840
0
The Unevenly-loaded Network
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 100/359
As an example, it was found that a cell supported 300 kbit/s at an
Eb/No value of 6 dB using 33.9 dBm of traffic channel power.
Pole Capacity estimated at 772 kbit/s.
Hence representative noise rise estimated as 2.14 dB.
42 dBm of traffic channel power is available.
What noise rise (and hence throughput) would this cause?
The Unevenly-loaded Network
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 101/359
If 33.9 dBm causes 2.14 dB of noise rise then 42 dBm would
cause 7.1 dB of noise rise.
Loading factor of 80%.
Resulting throughput of 622 kbit/s at an Eb/No value of 6 dB.
Tested with Monte Carlo simulation and found to be valid for
general situations where the distribution of the new load was
similar to the existing load.
The Unevenly-loaded Network
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 102/359
For heavily concentrated “hot spot” situations.
A static analysis can result in an estimate for downlink values of i.
Optimising Throughput – using Pilot SIR
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 103/359
The value of i influences
throughput. Any hotspotsshould be located where i is
low. Examining the Pilot SIR
as part of a static analysis
when the network is heavily
loaded will indicate the
throughput possible.
Optimising Throughput – using Pilot SIR
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 104/359
If the pilot is at +33 dBm, the SIR reported will
be that for any traffic channel with the samepower.
This influences throughput.
E.g. SIR = -6 dB; target Eb/No = 4 dB
Maximum throughput for 33 dBm = 384 kbit/s.
33dBm = 2W so
192 kbit/s/watt (192 kbit/J).
Optimising Throughput – using Pilot SIR
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 105/359
Pilot SIR varies between -5 dB and -12dB.
kbit/J parameter varies by a factor of 5.
Re-directing antennas can cause
variation by a factor of 3 or more.
Conclusions
Downlink Analysis
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 106/359
▫ Downlink capacity can be estimated for dimensioning purposes.
▫ Estimates compared with Monte Carlo simulator predictions.
▫ Estimates less accurate where network is not evenly loaded.
Simulations can lead to a more accurate estimate.
▫ Site location and antenna azimuth have key role in optimising
downlink throughput.
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 107/359
Network Dimensioning
Session ObjectivesNetwork Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 108/359
▪ To answer the questions:
▫ What is dimensioning?
▫ How might we carry out dimensioning?
▫ What are the key issues with dimensioning?
▫ What are the key equations that we need?
▫ What is sensitivity analysis and how is it carried
out?
What is Dimensioning?Network Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 109/359
▪ Dimensioning is the task of estimating the site numbers in a network
▪ Why do we need to know the number of sites?
▫ Project Management
▫ Rollout Strategy
▫ Vendor Comparison
▫ Configuration Comparison
▫ Business Planning
▪ We are NOT talking about dimensioning individual sites/links forcapacity
Dimensioning OutputsNetwork Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 110/359
▪ Site Numbers
▫ By region
▫ By configuration
▫ By environment
▪ Project Milestones and Required Resource
▪ Turnkey Zones
▪ Core Network and Transmission Network DimensioningInputs
▪ Sensitivity Analysis
▫ How sensitive is the dimensioning to changes in inputs...
Types of Dimensioning
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 111/359
▪ There are many different ways to dimension anetwork
▫ There is no „right‟ way but there are many „wrong‟ways
▪ These can be generically grouped:
▫ Simple Coverage▫ Simple Capacity
▫ Simple Combined Coverage and Capacity
▫ Interactive Coverage and Capacity
▫ Benchmark Planning
„Spreadsheet‟ based
Planning Tool based
Dimensioning Inputs
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 112/359
Environment
SiteConfiguration
Geographic Demographic
Service
Simple Coverage
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 113/359
▪ Link Budget based
▫ i.e. simple numerical calculation
▪ Firstly a link budget is created
▪ The maximum path loss is used to calculate the cell
range using a propagation model
▪ The cell range is used to calculate the site area
▪ Site Numbers = (Total Area)/(Site Area)
Create Link Budget
Calculate Range
Calculate Site Area
Calculate Number
of Sites in a givenArea
Max PL
Max Range
Max Area
Key Issues with Simple CoverageDimensioning
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 114/359
Dimensioning
▪ Capacity!
▪ Building Penetration
▪ Shadow Fading
▪ Propagation Model
▪ Site Environment Parameters
Shadow Fading and Building Penetration
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 115/359
▪ Building Penetration
▫ Mean and standard deviation per environment ▪ Shadow Fading
▫ Typically calculated using „Jakes formula‟
▫ This assumes an isolated omni directional site…
b
aberf
b
abaerf F u
11
21exp1
2
12
2
0
xa
2
log10 10
enb Where: ;
x 0 - = Fade Margin
= location variability ( standard deviation )
n = Propagation Model Exponent
x0 -
x0 -
P(connect)
P(connect)
5.6
0
50%
76% 90%
75%
Point Location Probability
Area Location Probability
RH Clarke, "Statistical Theory of Mobile-Radio Reception," Bell SystemTechnical Journal 47, July 1968, pp. 957-1000
Shadow Fading
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 116/359
▪ Shadow Fading
▫ The distribution of power signals is known as log-normal distribution
▪That is the signal measured in decibels has a normal distribution
▫ The process by which this distribution comes about is known ashadowing or slow fading
▫ Any variation in received signal is of the order of 10s to 100s of metres
▫ The standard deviation in decibels is known as the locationvariability
▪On average 5 - 12 dB ( 8 dB at 2GHz in urban environment )
▪Has a tendency to increase with frequency
▪No relationship has been proven between range and
▪With A = 5.2 in urban and 6.6 in suburban and fc in MHz
A f f cc log3.1log65.02
Propagation Model
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 117/359
▪ COST 231 Hata typically used
▫ Rural very optimistic▫ No accounting for diffraction
▫ Typically considered inaccurate
▫ Accuracy limited to sites above30m, ranges above 1km,frequencies below 2GHz
▫ R in km
▫ fc in MHz
▫ hb base station height in metres
▫ hm mobile height usually 1.5m
▪ Very simple to implement...
▪ Assuming base station height tobe 30m and carrier of 1910MHzin a medium sized city
areasianmetropolit3dBG
citiessizedmedium0dBG
97.475.11log2.3
log55.69.44
log82.13log9.333.46
log
2
m
b
bc
dB
h E
h B
h f F
G E R BF L
R LdB log35137
Site Environment Parameters
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 118/359
▪ Different configurations in different environments
▫ MHA
▫ Xpolar/Space Diversity
▫ Antennas
▪ Requires different link budgets…
▪ Loading on sites may also differ withenvironment
▫ To take advantage of the capacity-coverage
tradeoff
▪ Also different number of sectors
Area Calculation
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 119/359
▪ Cells are complex shapes
▪ We assume in dimensioning that cellsconform to a regular shape
▫ Hexagons are commonly used because oftheir close packing properties
▫ K factors used to represent the differencebetween a circle of radius r and the site area
▫ The K factor will depend upon the number ofsectors
K = 0.827
K = 0.62
r
r
2
r K Area
Coverage-based Dimensioning: Example
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 120/359
▪ Area to be covered: 80 km2.
▪ Link Budget for NR of 3dB suggestsmaximum path loss of 151 dB can betolerated, assuming sectored antennas areused.
▪ In building margin and shadow fading marginreduce this to 131 dB
▪ Path loss model
K = 0.62
R
dBlog35137 R L
km674.01010 35635137 L R
Coverage-based Dimensioning: Example
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 121/359
▪ Area covered by 3-sectored site
▪ Number of sites required =
▪ 90 sites required (270 cells)
K = 0.62
R
km674.0101035635137
L
R
22 km88.062.0 R
9088.080
excel
Environment Distribution
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 122/359
▪ Spreadsheets don‟t dealwith topology ormorphology accurately
▫ Hills, parks anddistributed target areas
▫ Interference and trafficcaptured by sites will
vary
▪ Margins for siteacquisition and overlapare required
Urban Area Site Numbers
Suburban
Area
Site Numbers?
Simple Capacity Dimensioning
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 123/359
▪ Capacity calculation based
▪ Firstly calculate maximumcapacity per carrier
▪ Calculate maximum offeredtraffic per sector
▪ Calculate site area based ontraffic density
▪ Finally calculate themaximum number of sites inan area
Calculate CarrierCapacity
Calculate SectorOffered Traffic
Calculate MaximumSite Area
Calculate Numberof Sites in a Given
Area
Erlang-B
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 124/359
▪ Erlang-B formula provides an
estimate of the peak traffic (notexceeded more than x%(usually 2%) of the time giventhe average traffic (quoted inErlangs).
▪ Erlang-B should only be usedfor:
▫ circuit switched traffic
▫ single services
▪ UMTS is multi-service andpacket switched...
Variation of demand with time
time
demand
average
peak
Erlang B
Accommodating a multi-service system
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 125/359
▪ The Erlang B formula relies on the variance of the
demand equalling the mean (a Poisson distribution).
▪ If a particular service requires more than one “trunk”per connection, the demand is effectively linearlyscaled and the variance no longer equals the mean.
▪ Methods to investigate:
▫ Equivalent Erlangs
▫ Post Erlang-B
▫ Campbell‟s Theorem
Equivalent Erlangs Example
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 126/359
▪ Let us consider 2 services sharing the same resource:
▫ Service 1: uses 1 trunk per connection. 12 Erlangs of traffic.▫ Service 2, uses 3 trunks per connection. 6 Erlangs of traffic.
▪ We could regard the above as equivalent to 30 Erlangs of service 1:
▫ 30 Erlangs require 39 trunks for a 2% Blocking Probability
▪ Alternatively, we could regard the above as equivalent to 10 Erlangsof service 2.
▫ 10 Erlangs require 17 trunks, (equivalent to 51 “service 1 trunks”) for a 2%blocking probability
▪ Prediction varies depending on what approach you choose.
Post Erlang-B
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 127/359
▪ Consider 2 services sharing the same resource:
▫ Service 1: uses 1 trunk per connection. 12 Erlangs of traffic.
▫ Service 2: uses 3 trunks per connection. 6 Erlangs of traffic.
▪ We could calculate the requirement separately
▫ Service 1: 12 Erlangs require 19 trunks for a 2% Blocking Probability
▫ Service 2: 6 Erlangs require 12 trunks (equivalent to 36 “service 1trunks”).
▪ Adding these together gives 55 trunks.
▪ This method is known to over-estimate the number of trunks requiredas can be demonstrated by considering services requiring an equalnumber of trunks.
Post Erlang-B
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 128/359
▪ Consider 2 services requiring equal resource:
▫ Service 1: uses 1 trunk per connection. 12 Erlangs of traffic.
▫ Service 2: uses 1 trunk per connection. 6 Erlangs of traffic.
▪ We could calculate the requirement separately
▫ Service 1: 12 Erlangs require 19 trunks for a 2% Blocking Probability
▫ Service 2: 6 Erlangs require 12 trunks.
▪ Adding these together gives 31 trunks.
▪ The accepted method of treating the above would be to regard it as a
total of 18 Erlangs that would require 26 trunks.
▪ Post Erlang-B overestimates the requirement.
Campbell’s Theorem
▪ Campbell‟s theorem creates a composite distribution where:
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 129/359
▪ Campbell s theorem creates a composite distribution where:
▪ c is known as the capacity factor
▪ The amplitude used in the capacity is the amplitude of the target service
▪ Once the offered traffic and Capacity are derived, GoS can be derived withErlang-B -> similarly Required Capacity can be calculated if Offered Trafficand GoS target is known
caC Capacity ii
c
fficOfferedTra
i iii
i
iii
ba
ba
c
2
= mean = variance
i = arrival rateai = amplitude ofservicebi = mean holdingtime
iibγTrafficOfferedService
Campbell’s Theorem Example(1)
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 130/359
▪ Consider the same 2 services sharing the same
resource:▫ Service 1: uses 1 trunk per connection. 12 Erlangs of traffic.
▫ Service 2, uses 3 trunks per connection. 6 Erlangs of traffic.
▪ In this case the mean is:
▪ The variance is:
3063121Erlangs iiii aab
6636112Erlangs 2222iiii aab
Campbell’s Theorem Example(2)
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 131/359
▪ Capacity Factor c is:
▪ Offered Traffic for filtered distribution:
▪ Required Capacity for filtered distribution at2% GoS is 21
2.230
66
c
63.132.2
30 TrafficOffered c
α
Campbell’s Theorem Example(2)
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 132/359
▪ Required Capacity is different depending upon target
service for GoS (in service 1 Erlangs):
▫ Target is Service 1 C1=(2.2 x 21) + 1 = 47
▫ Target is Service 2, C2=(2.2 x 21) + 3 = 49
▪ Different services will require a different capacity for the
same GoS. In other words: for a given capacity, the
different services will experience a slightly different GoS.
campbell
Traffic Analysis Methods Compared
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 133/359
▪ Equivalent Erlangs
▫ Optimistic if you use the smallest amplitude of trunk (39)▫ Pessimistic if you use the largest amplitude of trunk (51)
▪ Post Erlang-B
▫ Pessimistic (55)
▫ Trunking efficiency improvement with magnitude ignored
▪ Campbell‟s theorem
▫ Middle band (47 - 49)
▫ Different capacities required for different services -realistic
▫ Preferred solution for dimensioning, but not ideal...
Capacity Dimensioning with Campbell’sTheorem
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 134/359
Theorem
▪ Consider the following service definition and traffic forecast.
Service Amplitude Forecast
Voice 1 250 E64 kbps data 2 63 E
144 kbps data 4 41 E
384 kbps data 8 12 E
Capacity Dimensioning with Campbell’sTheorem
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 135/359
Theorem
▪ Assuming we have n cells, we can determine the loadingper cell.
nnc
c
nnnnn
nnnnn
210028.3636meantrafficoffered
028.3636
1926
mean
variance
1926821414263250variance
636821414263250mean
222
Capacity Dimensioning with Campbell’sTheorem
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 136/359
Theorem
▪ Unfortunately, we cannot now look up “210/n” in the ErlangB tables.
▪ We need to introduce a notional capacity per cell in termsof “Service 1 trunks”.
▪ We will assume that each cell has a capacity of 32 suchtrunks.
nnc
210
028.3
636meantrafficoffered
Capacity Dimensioning with Campbell’sTheorem
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 137/359
Theorem
▪ Considering the equation
▪ C iis predefined as 32. a
idepends on the service we use as our
“benchmark”.
▪ Choosing service 3 as the “benchmark” service make aiequal to 4.
▪ Therefore 9.25 (or, rather, 9) trunks will service 4.34 Erlangs.
caC ii Capacity
25.9
028.3
4323
C
Capacity Dimensioning with Campbell’sTheorem
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 138/359
Theorem
▪ 9 trunks will service 4.34 Erlangs.
▪ Therefore,
▪ Cell requirement is established at 48cells.
▪ Each of the cells will service:
▫ 5.21 Erlangs of voice
▫ 1.32 Erlangs of 64 kbps data
▫ 0.85 Erlangs of 144 kbps data
▫ 0.25 Erlangs of 384 kbps data
48
34.4210
n
n
campbell
Key Issues with Simple capacitydimensioning
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 139/359
▪ What is the resource?
▫ Bitrate - no…
▫ Loading of individual user - yes…
▫ Calculate traffic analysis using the ratio of single channel loading fordifferent services
▪ Loading is affected by bitrate and E
b /N
0
▪ Note that uplink and downlink will yield different pole capacities
1amplitudefor1amplitudeforratebit
serviceforserviceforratebitamplitudeRelative
0
0
N
E
N E
b
b
Campbell amplitude
Simple Coverage and Capacity Dimensioning
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 140/359
▪ Simply carry out both coverage ANDcapacity dimensioning and combine them,taking the maximum value
▪ This should be carried out on a „per
environment basis‟, preferably per region.
Complex Coverage and CapacityDimensioning
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 141/359
▪ In this case a link is made between the coverage andcapacity
▪ The coverage is calculated from an initial link budget. Thislink budget will include an assumption of Noise Rise
▪ Then the number of subscribers captured per cell iscalculated
▪ The required loading to support the subscribers to thedesired GoS is calculated.
▪ This can be used to recalculate the Noise Rise…and fedback into the link budget
Complex Coverage and CapacityDimensioning (example)
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 142/359
▪ Link budget is created assuming 4 dB Noise Rise
▪ 120 Cells required.
▪ Analysis of traffic forecast suggest each cell will experience
2 dB Noise Rise
▪ Re-create link budget
▪ 90 Cells required
▪ Analyse loading: 3 dB Noise Rise
▪ Re-create link budget
▪ 100 Cells required
▪ Analyse loading: 2.8 dB Noise Rise……….
Benchmark Planning
etwor
Dimensioning
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 143/359
▪ A „first pass‟ nominal plan is created
▪ The problems of non-contiguous clutter and diffraction are
removed
▪ A capacity „check‟ is required to ensure that cells aren‟t
overloaded
▪ Very resource hungry…
▪ Sensitivity analysis is impossible
Other Dimensioning Key Failings...
GSM/UMTS Interaction
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 144/359
▪ GSM/UMTS Interaction
▫ Proportion a percentage of voice traffic to GSM
▫ Don‟t assume that UMTS carries all of the traffic
▪ Microcells
▫ Offer capacity relief to macrocells
▫ This allows macrocells to be larger, potentially with a lower loading
▪ Repeaters
▫ Extend the coverage of macrocells at a lower cost than a new Node-B
▪ Sharing the load▫ Analysis so far has assumed that each cell looks after its own traffic. If
capacity is fully allocated on best server, a connection may be establishedwith a neighbour.
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 145/359
Cell Breathing
Uplink Analysis: Cell Breathing
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 146/359
▪ The base station has to achieve the required Eb/No ratio from
a particular mobile.
▪ Noise and interference is present from:
▫ Thermal Noise
▫ Other mobiles in the same cell
▫ Mobiles in other cells
▪ Remember that all mobiles use the same frequency
Noise Rise vs. Throughput
e om na
Plan
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 147/359
▪ Each new userincreases the
throughput of
the cell but also
increases the
effective noiseexperienced by
all other users.
Noise Rise vs. Throughput
0.00
5.00
10.00
15.00
20.00
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Throughput (x100kbps)
N o i s e R i s e
Coverage vs. Capacity Example
e om na
Plan
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 148/359
Two polygons were createdto allow traffic served by aparticular cell to be spreadover two geographicregions.
Coverage vs. Capacity Example
e om na
Plan
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 149/359
Firstly 50 terminals were
spread over the large outerpolygon to demonstrate thatcoverage existed in that area.
Coverage vs. Capacity Example
e om na
Plan
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 150/359
Next 340 terminals were spreadover the same area. The quality of
service is significantly reducedsuggesting that there is a capacityproblem. The report suggests that306 users may be the maximum thesite can support.
Coverage vs. Capacity Example
However, redistributing these
e om na
Plan
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 151/359
, gterminals so as to reduce path lossresults in a very good quality ofservice being restored. An averageof 330 terminals were served.Capacity and coverage are linkedtogether.
Cell Breathing :- “good” or “bad” ?
e om na
Plan
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 152/359
.
▪ Cell Breathing is integral to WCDMA cellular radio systems.
▪ Its disadvantage is that it leads to the creation of gaps in the network
coverage.
▪Its advantage is that it maximises capacity when it is demanded.
▪ The amount of cell breathing can be controlled by limiting the
NoiseRise in the admission algorithm. It cannot, however be
eliminated.
Cell Breathing :- “Good” or “Bad” ?
e om na
Plan
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 153/359
.
▪ Limiting the Noise Rise to 3 dB will restrict throughput to 50% of
theoretical maximum and restrict coverage shrinkage to 33% of its
maximum area.
▪ Allowing Noise Rise to increase to 10 dB will allow throughput to rise
to approximately 90% of its theoretical maximum but coverage
shrinkage will rise to 73% of maximum.
▪ Planning to restrict Noise Rise to 3 dB will necessitate the provision of
extra sites.
Cell Breathing
e om na
Plan
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 154/359
.
▪ Very rough rule of thumb.
Area shrinkage (%) =
Coverage with 3 dBNoise Rise
Coverage with 10 dBNoise Rise
Unloaded Coverage
5.17101100
NR
breathing
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 155/359
Dimensioning Packet Scheduled Traffic
Packet Scheduled Traffic
mens on ng ac et c e u e
Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 156/359
▪ Issues
▫ Adding Packet Traffic into gaps in CS demand
▫ Trunking Efficiency when delays are tolerated.
Gaps in CS demand
mens on ng ac et c e u e
Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 157/359
▪ Each cell will have a notional capacity in kbps
▪ Simulation will provide details of mean demand during “busy
hour”.
▪ Difference between the two can be construed as “free
capacity”.
Efficiency using Erlang C
▪ Erlang C formula will predict probability of a given delay
mens on ng ac et c e u e
Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 158/359
Erlang C formula will predict probability of a given delayfor a given number of trunks and Erlangs of offered
traffic
▪ If unlimited delay can be tolerated, efficiency will be100%.
Example:
25 Servers, 20 Erlangs of Traffic offered.P(>0) = 0.21
P(T1/T2>0.4)=0.02
P(T1/T2>0.8)=0.0038
P(T1/T2>1.2)=0.00052
P(T1/T2>1.6)=0.00007
P(T1/T2>2.0)=0.00001
T1 is delay time
T2 is mean holding time
Erlang c
Delay Categories for UMTS
mens on ng ac et c e u e
Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 159/359
Category Example Delay
Conversational Interactive Games <250 ms
Interactive Web Browsing <4 s
Streaming ftp transmissions <10 s
Background e-mail >10 s
Example Result
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 160/359
▪ 25 “Trunks”, each with 12200 bps capacity.
▪ Mean delay of 10 ms acceptable. P(T1/T2>1)=0.02
▪ Result: 22 Erlangs of traffic carried.
▪ Note: Erlang B for 0.02 GoS would predict 17.5 Erlangs
carried.
▪ Accepting a delay increases trunking efficiency.
Erlang B
Erlang C for Shoppers
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 161/359
Factors affecting the delay
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 162/359
▪ Delay is proportional to the holding time.
▪ Holding time proportional to packet size.
▪ Need to consider “sensible” packet size.
▪ www model has standard packet size of 3840 bits.
▪ Short text message of 128 alphanumeric characters will
require approximately 1 kbit of capacity.
Probability of the Delay exceeding a givenvalue.
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 163/359
▪ The higher the delay, the lower the probability.▪ The curve is an exponential decay.
Probability vs. Delay
0
5
10
15
20
25
30
0 20 40 60 80 100 120 140
Delay (ms)
P r o b a b i l i t y ( % )
Line 1
Dimensioning “all packet” services
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 164/359
▪ Packet traffic may dominate demand in UMTS.
▪ Efficiency issues for packet data become very
significant.
Dimensioning “all packet” services
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 165/359
▪ Tolerance of delays implies tolerance of re-transmissions.
▪ Retransmission possibility allows for very low Eb/No
▪ Target Eb/No as low as 1 dB with 25% FER will provide maximumefficiency.
▪ 33% extra demand due to retransmission more than offset by
increased capacity.
Dimensioning “all packet” services
▪ If i = 0 5 Eb/No = 1 dB pole capacity = 2 Mbps
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 166/359
If i = 0.5, Eb/No = 1 dB, pole capacity = 2 Mbps.
▪ At 50% loading factor, a capacity of 1 Mbps could be
expected.
▪ 750 kbps of offered traffic will result in 1 Mbps carried if
retransmission ratio is 25%.
▪ High trunking efficiency will result in very high levels of traffic
serviced.
▪ Delay implications of retransmissions need to be addressed.
▪ Frame period of 10 ms significant.
Packet Scheduling for Web Browsing
▪ Web Browsing involves near-total asymmetry in downlink
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 167/359
Web Browsing involves near total asymmetry in downlinkdirection.
▪ It is possible to plan a cell so that, at the maximum pathloss, the downlink capacity is greater than the uplinkcapacity.
▪ Low Eb/No requirement of packet data can make great useof this extra capacity.
▪ E.g. 320 kbps at Eb/No of 1 dB imposes same loading as100 kbps at an Eb/No of 6 dB.
Packet Scheduling for Web Browsing
E l
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 168/359
▪ Example.
▫ 40 x 12.2 kbps speech channels on uplink and downlink.
▫ Analysis suggests that downlink transmitter is operating 4
dB below maximum.
▫ This extra 4 dB of power can be converted into
approximately 850 kbps of data throughput for packet
scheduled traffic.
Effect of Base Station Power on DownlinkNoise Rise
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 169/359
▪ Increasing Base Station Power will increase the Noise Rise
on the downlink but not in a particularly straightforward way.
▪ The Analysis involves imagining that the base station
transmits an effective thermal noise power along with the user
power.
Effect of Base Station Power on DownlinkNoise Rise
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 170/359
User Power B dBm
Noise Power A dBm
S
10
10
10
1010
10
10110log
10
101010logRise,Noise
A B
A
B A
X
Effect of Base Station Power on DownlinkNoise Rise
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 171/359
110
10
10
11010
10101
10
1010
1010
1010
X
B A
X A B
X A B
110
10log10
110
10log10
10
10poweruser
10
10
10
10
NR
X
B
A
The first step in calculating the new noise rise is to assign a value to thepower of the virtual noise source, A.
Effect of Base Station Power on DownlinkNoise Rise
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 172/359
1010 101log10RiseNoiseNew
AC
Then if the user power is increased from B dBm to C dBm:
Effect of Base Station Power on Downlink NoiseRise
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 173/359
Example: Original Noise Rise = 2 dB. Base Station Transmit Power
= 36 dBm.
Transmit Power is increased to 41 dBm.
Intermediate Parameter A = 38.33 dBm
New Noise Rise is 4.55 dB.
(Noise Rise has increased by 2.55 dB for a Base Station powerincrease of 5 dB. This increase in Noise Rise represents an increase
in loading factor and, hence, an increase in throughput.)
10
10 101log10RiseNoiseNew AC
110
10
log10
110
10log10
10
10poweruser
10
10
10
10
NR
X
B
A
New noise rise
Pause for Breath
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 174/359
▪ General Ideas:
▫ A small amount of extra capacity for CS traffic withhigh Eb/No can be converted to a larger amount oftraffic for a service at a low Eb/No.
▫ Extra Power available in the downlink can beconverted to extra capacity by increasing the Noise
Rise and hence the loading factor.
Calculating Extra Capacity
▪ Example
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 175/359
▪ Example.
▫ CS traffic Eb/No 6 dB, provides 225 kbps of traffic (uplink anddownlink) for a cell.
▫ i = 50%.
▫ Orthogonality on the downlink, = 0.6
▫ 32 dBm user power (max. available 42 dBm) on downlink.
▫ Our challenge is to determine the extra capacity available on the
uplink and downlink for a service requiring an Eb/No of 1.5 dB if
the base station user power can be increased to 42 dBm and the
Noise Rise limit on the uplink is set at 3 dB.
U li k l i
Calculating Extra Capacity
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 176/359
▪ Uplink analysis.
▫ At an Eb/No of 6 dB, the pole capacity isapproximately 640 kbps and throughput at 3 dBNoise Rise would be 320 kbps.
▫ 320 - 225 = 95 kbps available.
▫ 95 kbps at an Eb/No of 6 dB produces anequivalent loading as 268 kbps at an Eb/No of 1.5dB.
▫ 268 kbps of data can be added to the uplink.
Calculating Extra Capacity
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 177/359
▪ Downlink analysis.
▫ At an Eb/No of 6 dB, the pole capacity on the downlink is
approximately 1067 kbps and a throughput of 225 kbps
represents a loading factor of 21% and, hence, a Noise
Rise of 1 dB.
▫ Increasing the base station power from 32 dBm to 42 dBm
would increase the Noise Rise to 5.55 dB.
▫This represents 72% loading factor, an increase of 51%.
Downlink analysis continued
Calculating Extra Capacity
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 178/359
▪ Downlink analysis continued.
▫ At an Eb/No of 1.5 dB, the pole capacity on the downlink
is approximately 3000 kbps 51% loading factor increase
represents 1530 kbps.
▪ Conclusion is that the cell could accommodate additional 268kbps packet data on the uplink and 1530 kbps packet data onthe downlink.
▪ Provision of bearers to service the 1530 kbps additional
packet data would have to be addressed as a separateproblem.
Conclusions
mens on ng ac et c e u e Traffic
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 179/359
▪ Tolerating Delays increases trunking efficiency -
efficiencies higher than 80% can usually be assumed.
▪ Tolerating retransmissions reduces required Eb/No to
as low as 1 dB.
▪ Packet data that is heavily asymmetric in the downlink
direction can be planned for at the link
budget/dimensioning stage.
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 180/359
Optimisation Procedures
Optimisation Techniques
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 181/359
)1(
3840CapacityPole
0
i N
E b
Parameter i is crucial. Limiting mutual interference
between cells will increase capacity
Limiting mutual interference
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 182/359
• Downtilt antennas.• Consider mounting antennas onthe side of buildings.
Limiting mutual interference
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 183/359
Controlling the backlobe can produce asmall but significant improvement in
capacity.
0º
0ºElec 6ºMech
0º 0º
6º
6º 6º
6ºElec 0ºMech
0º
6º
6º 0º
6ºElec -6ºMech
0º
-6º
12º
0º
Limiting mutual interference
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 184/359
• Key parameter: Frequency Re-use Efficiency(FRE).
(W)ceinterferencell-intertheis
(W)ceinterferencell-intratheis
FRE
Inter
Intra
Inter Intra
Intra
N
N N N
N
Note FRE = 1/(1+i)
Limiting mutual interference
• Avoid high sites they become overloaded quickly
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 185/359
• Avoid high sites, they become overloaded quickly.
Balanced siteswill divide load
evenly.
• High sites will
“grab” too much
traffic.•They will gather a
lot of interferenceon the uplink.•They willcontribute a lot ofinterference on thedownlink.
Multi-User Detection
▪ Multi-User detection (MUD) is a method used toi th f f th i b d i th
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 186/359
improve the performance of the receiver by reducing the
noise contributions from other CDMA users.
▪ The concept is based on the fact that noise from CDMAusers, although usually approximated with AWGNcharacteristics, inherently consists of coherent signals.
▪ MUD reception decodes a number of userssimultaneously and subtracts their noise contributions tothe each other
▪ Essentially this results in a more sensitive receiver
Visualising the Processing Gain w/o MUD
W/Hz W/Hz W/Hz
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 187/359
W/Hz W/Hz W/Hz
Ec
No
SignalIntra-cell Noise
Inter-cell Noise
Before Spreading
After Spreading With Noise
f f f
W/Hz
After Despreading /Correlation
f
W/HzE
b
No
Post Filtering Orthog = 0
f
dBW/Hz
Eb
No
Eb /No
f
Visualising the Processing Gain with MUD
W/Hz
Post Filtering
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 188/359
W/Hz
Signal
Inter-cell Noise
After Despreading /Correlation
Filtering
f
Other Users
Eb
No
W/Hz
f
Eb
No
W/Hz
f
Eb
No
W/Hz
f
Eb
No
W/Hz
f
Because of MUD the contribution of the other users to the Noise is Reduced.
It is not completely eliminated because of the inaccuracies of the Multiple access interference estimation.
Multi-User Detection
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 189/359
▪ Modelling the MUD receiver can beachieved by adjusting the Eb/No of theservices to account for the improved noisecancellation.
▪ Compensates for cable loss between antenna and base station(t i ll 3dB)
Mast Head Amplifiers
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 190/359
(typically 3dB)
▪ MHA used to increase coverage range
▪ Gain 12dB (adjustable)
▪ MHA‟s are also called TMA (Tower Mounted Amplifiers)
- LNA‟s used on receive path
▪ Increase uplink capacity
▪ Only beneficial in uplink-limited situations
▪ 1.6 dB Noise Figure (NF), Gain 12 dB
▪ Drawback - Insertion loss on
Tx path (~ 1.3 dB)Ant
Bias-TDC
TMA
by pass
Uplink Receive Space Diversity
▪ Common to have two receive antennas per sector at the base station.
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 191/359
▪ Even if highly correlated, coherent combination should yield ~3 dB
improvement.
▪ In practice a gain of 4 dB or more is expected from antennas spaced 2-3 m
apart.
Receiveantenna 1
Receiveantenna 2
Uplink Receive Space Diversity
▪ This is not “conventional” space diversity
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 192/359
▪ This is not conventional space diversity.
▪ Each antenna is connected to a separate finger of the Rake receiver.
▪ This is possible due to the synchronisation and channel estimation
derived from the Pilot channel.
▪ Thus Eb/No is improved, rather than simply an effective power gain.
▪ Final limitation of this technique (why not 100 or 1000 antennas?) is
interesting to consider
▪ Very low individual Eb/No will probably mean a very low pilot level which
will lead to poor coherence and little gain - process becomes “self -
defeating”.
Downlink Transmit Diversity
▪ UMTS explicitly allows the use of transmit diversity from the base station
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 193/359
▪ However it is not possible to simply transmit simultaneously from two closeantennas as this would cause an interference pattern
▪ Mobile terminals must have the capability of implementing downlink transmitdiversity .
Transmitantenna 1
Transmitantenna 2
Downlink Transmit Diversity
▪ The following methods are suggested in the UMTS
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 194/359
Transmit DiversityMethod
Description
TSTD Time Switched Transmit antennaDiversity (open loop)
STTD Space Time block coding Transmitantenna Diversity (open loop)
Closed Loop Mode 1 Different Orthogonal Pilots
Closed Loop Mode 2 Same Pilot
▪ The following methods are suggested in the UMTS
standards to avoid the problem of the interferencepattern
Time Switched Transmit antenna Diversity(TSTD) Synch channel only
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 195/359
▪ Even numbered slots transmitted on Antenna 1, oddnumbered slots on Antenna 2
Antenna 1
Antenna 2
P-SCH
Slot #0 Slot #1 Slot #14Slot #2
P-SCH
P-SCH P-SCH
S-SCH
S-SCH
S-SCH S-SCH
Space Time block coding Transmit antennaDiversity (STTD)
STTD di i ti l i UTRAN STTD t i d t t th UE
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 196/359
b0 b1 b2 b3
b0 b1 b2 b3
-b2 b3 b0 -b1
Antenna 1
Antenna 2
Channel bits
STTD encoded channel bits
for antenna 1 and antenna 2.
▪ STTD encoding is optional in UTRAN. STTD support is mandatory at the UE
▪ Channel coding, rate matching and interleaving is done as in the non-diversity mod
▪ STTD encoding is applied on blocks of 4 consecutive channel bits
▪ The bit bi is real valued {0} for DTX bits and {1, -1} for all other channel bits.
Analysis of STTD
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 197/359
▪ STTD encoding effectively spreads a data bit across more than one bit period.
▪ This leads to a general improvement in noise performance.▪ Further, it allows a stronger signal from one antenna to dominate.
b0 b1 b2 b3-b2 b3 b0 -b1
b0-b2 b1+b3 b0+b2 b3-b1
Processing alternate bits will extract the data
Closed Loop Mode
▪ Channel coding, interleaving and spreading are done as in
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 198/359
g, g p gnon-diversity mode
▪ The spread complex valued signal is fed to both TXantenna branches, and weighted with antenna specificweight factors w1 and w2.
▪ The weight factors are determined by the UE, and signalled
using the FBI field of uplink DPCCH (Dedicated PhysicalControl Channel).
1 radio frame: T = 10 ms
Pilot
Npilot bits
TPC
NTPC bits
Slot #0 Slot #1 Slot #i Slot #14
Tslot = 2560 chips, 10 bits
f
DPCCHFBI
NFBI bitsTFCI
NTFCI bits
Closed Loop Mode
Spread/scramble
w1
CPICH1
Tx
Ant1
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 199/359
Spread/scramble
w2
DPCHDPCCH
DPDCH
Rx
Rx
CPICH2
Ant2
Tx
Weight Generation
w1 w2
Determine FBI message
from Uplink DPCCH
Soft Handover as an optimisation technique
▪ Soft handover is a form of space diversity.
pt m sat on Procedures
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 200/359
Soft handover is a form of space diversity.
▪ Issues.
▫ All except one signal may be poor
quality.
▫ Handover margin can influence the
gain achieved.
▫ 2 dB Eb/No gain possible.
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 201/359
UTRAN
Architecture and Protocols
Protocol Model for UTRAN Interfaces
▪ Protocol structures in UTRAN are designed in layers and planes
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 202/359
▪ Protocol structures in UTRAN are designed in layers and planes.
▪ They are seen as logically independent of each other
▫ However they will physically interact.
▪ Being logically independent allows for changes to blocks in the
future
theoretically!
General Protocol Model forUTRAN Terrestrial Interfaces
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 203/359
Horizontal Layers in the General Protocol Model
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 204/359
▪ All UTRAN related issues are only visible in the Radio
Network Layer
▪ The Transport Layer simply represents standard transport
technology for use in UTRAN
▫ e.g. ATM and appropriate ATM Adaptation Layers
▪ AAL2 ( voice ) and AAL5 ( data/control)
▫ UDP/IP or RTP/UDP/IP ( released 6 ? )
Vertical Planes in the General Protocol Model
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 205/359
▪ The Control Plane is provided for all UMTSspecific control signalling including:
▫ Application Protocols
▫ Signalling Bearers
▪ The User Plane is provided for all data sent andreceived by the user including:
▫ Data Streams
▫ Data Bearers
Vertical Planes in the General Protocol Model
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 206/359
▪ The Transport Network Control Plane also includes the AccessLink Control Application Part, ALCAP.
Transport Network User
Plane
Transport Network User
Plane
Transport Network Control
Plane
ALCAP
ALCAP - Access Link Control Application Part
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 207/359
▪ ALCAP sets up the transport bearers for the User
Plane
▪ The independence of the User and Control Planeassumes that
▫ ALCAP signalling transaction actually takes place.
Iu, the UTRAN-CN Interface
▪ The Iu has two different instances.
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 208/359
▫ Iu-CS for circuit switched connections
▫ Iu_PS for packet switched connections
▪ Originally proposed as one interface but the differences
between circuit and packet switched services has produced
two.
▪ The Control plane is the same for CS/PS as far as possible!
Protocol structure for the Iu-CS
UTRAN
▪ The Iu-CS control plane
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 209/359
▫ RANAP - Radio Access Network Application Part
▫ Broad-band SS7 ( Network-Network Interface NNI )
▫ ATM - AAL 5
▪ The Iu-CS transport network control plane
▫ Signaling protocol for setting up ATM AAL2
▫ Broad-band SS7
▫ ATM - AAL 5
▪ The Iu-CS user plane.
▫ ATM - AAL2 reserved for each CS connection
Protocol structure for the Iu-CS
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 210/359
▪ The Iu-PS control plane
Protocol structure for the Iu-PS
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 211/359
▫ RANAP - Radio Access Network Application Part
▫ Broad-band SS7 ( NNI ) or IP based signalling bearers
▫ ATM - AAL 5
▪ The Iu-PS transport network control plane
▫ NONE required
▪ the GPRS tunneling protocol only requires the IP addresses
which RANAP has supplied.
▪ The Iu-PS user plane.
Protocol structure for the Iu-PS
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 212/359
p
▫ ATM - AAL5
▫ multiple packet data flows are multiplexed onto several
permanent virtual connections PVCs
▫ Each flow uses
▪ GTP-U,
▪ UDP connectionless transport and IP
addressing.
Protocol structure for the Iu-PS
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 213/359
RANAP
▪ The functionality of the RANAP is performed through Elementary
Procedures EP‟s
RANAP has 12 defined functions
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 214/359
▪ RANAP has 12 defined functions
Relocation Iu Release Radio Access BearerManagement
Paging Location reporting Common IDManagement
Security Mode Control UE-CN signallingtransfer
OverloadManagement
Reset ReportingUnsuccessfultransmitted data
Management ofTracing
Iur, the RNC-RNC Interface
▪ The Iur provides support for four distinct functions:
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 215/359
▪ The Iur provides support for four distinct functions:
▫ Inter RNC Mobility
▫ Dedicated Channel Traffic
▫ Common Channel Traffic
▫ Global Resource Management
▪ The Radio Network System Application Part, RNSAP
is therefore divided into four modules.
RNSAP Iur1: Inter RNC Mobility
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 216/359
▪ Iur1 provides for the mobility of the user between two RNCs but
does not support the exchange of user data traffic.
▪ If the network fails to provide this then the only means of mobility
is to disconnect from RNS1 and reapply for connection in RNS2
Cell
Cell
RNSAP Iur2: Dedicated Channel Traffic
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 217/359
▪ Iur2 sets up and maintains a dedicated channel between two
RNCs
▪ Used in inter-RNC soft handover.
▪ Provides the serving SRNC with the capability to manage the
radio links in drift DRNCs
Cell
Cell SRNC
DRNC
RNSAP Iur3: Common Channel Traffic
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 218/359
▪ If Iur3 is not implemented then every time an inter-RNC cell
update takes place then that RNC becomes the serving RNC.
Cell
Cell
SRNC
RNC
RNC
SRNC
SRNC
RNC
No Iur3
RNSAP Iur4: Global Resource Management
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 219/359
▪ This function is considered optionally as it does not
transmit any user data across the Iur.
▪ However it does provide useful information
▫ transfer of cell measurement between RNCs
▫ transfer of Node B timing information between RNCs
Protocol Structure for Iur
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 220/359
Iub, RNC-Node B Interface
▪ Iub user plane, defines every type of transport channel, using
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 221/359
AAL2
▪ Iub signalling control plane, AAL5, is divided into 2 essential
components;
▫ NBAP-C Common Node B application part
▫ NBAP-D dedicated Node B application part
NBAP-C
Set-up of 1st radio link to the UE
Cell configuration
Handling of RACH/FACH andPCH
NBAP-D
Addition, release of radio links
Handling of softer handover
Handling of dedicated and sharedchannels on a Node B - UE.x basis
Protocol structure for the Iub
UTRAN
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 222/359
Note:
SSCF servicespecific co-ordination functionAAL set-up
UNI .. user -network interface
SSCOP servicespecific connectionoriented protocolAAL set-up
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 223/359
ATM Overview
ATM
▪ ATM Concepts
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 224/359
▫ ATM is based on a Virtual Circuit Technology
▫ Similar to Circuit Switching, ATM uses signalling protocols toestablish a Circuit before data communication commences.
▫ Unlike Circuit Switching, ATM is based on Statistical Multiplexing(similar to Packet Switching)
ATM
▪ Virtual Circuit Concept
▪ A connection is first established using signallingprotocols
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 225/359
▫ A route from the source to the destination is chosen
▫ The same route is used for all cells (fixed sizepackets) whilst the connection is maintained.
▪ Consequently no routing decision is needed forevery cell
▪ No dedicated path is required unlike CircuitSwitching
▪ Each Link of the network is shared by a set of
virtual channels▫ Each cell uses only one virtual channel number
ATM
▪ Each packet contains enough information forthe node (switch) to forward it towards the
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 226/359
destination.
▪ Tables at nodes are filled with connection data
▪ Parameters used for establishing VirtualCircuits
▫ Calling and Called Party Addresses
▫ Traffic Characteristics
▫ QoS Parameters
ATM VCI & VPI Assignment
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 227/359
IN LINK IN VCI VPI OUT LINK OUT VCI VPIA 20 10 AB 4 3
AB 4 3 B 19 7
ATM
▪ Advantages of Virtual Circuits
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 228/359
g
▪ In-order delivery of packets or cells
▪ Fast Delivery (no routing decision foreach packet)
▪ Less Header Overhead▪ High efficiency when two stations
exchange data for long time
ATM
▪ Handling Congestion with Virtual Circuits
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 229/359
▪ Establishing Virtual Circuits alone is not sufficientto avoid congestion
▪ We must declare Traffic Characteristics and QoSrequirements.
▪ Resources must be reserved while establishingVirtual Circuits
▪ ATM network is made up of Switches &
ATM
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 230/359
Endpoints ▪ Switch
▫ Accepts incoming cell
▫ Reading and updating cell header
▫ Switching cell towards output interface
▪ Endpoints
▫ Contains ATM network interface adapter
▫ Ex. Workstations, routers, DSUs, LAN Switches,Video CODECs
ATM
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 231/359
ATM SwitchRouter
LAN Switch
Work Station
ATM Switch Router
ATM End points
NNI
UNI UNI
▪ Interfaces
▫ ATM supports two primary interfaces
ATM
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 232/359
▪ User Network Interface (UNI)
▫ Connects end system to an ATM switch
▫ RNC - Node B
▪ Network Network Interface (NNI)
▫ Connects two ATM switches
▫ RNC - MSC, RNC - RNC
▫ UNI and NNI can further be subdivided to Public and PrivateUNIs and NNIs
ATM - Interfaces
Private ATM Public ATM
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 233/359
PrivateNNI
PrivateUNI
PublicNNI
PublicUNI
Cell Format
▪ ATM Cell Format
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 234/359
▪ Each cell consists of 53 bytes
▪ First 5 bytes for cell header
▪ Remaining 48 bytes for payload
8 bits
53 bytes
Payload (48 Bytes)
Header (5 Bytes)
Cell-Header
GFC VPI VCI PT HECC
L
P
at
UNI
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 235/359
VPI VCI PT HEC
12 16 3 1 8 bits
C
L
P
4 8 16 3 1 8 bits
at
NNI
GFC : Generic Flow Control VPI : Virtual Path IdentifierVCI : Virtual Circuit Identifier PT : Payload TypeCLP : Cell Loss Priority HEC : Header error CheckUNI : User Network Interface NNI : Network-Network Interface
ATM
▪ ATM Cell Header Bits
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 236/359
▫ Generic flow control GFC▫ Only in the cells transported over UNI
▫ Enables a local switch to regulate flowcontrol
▫ In NNI these 4bits are part of VPI
▫ Virtual Path Identifier VPI
▫ Used for identification/routing purposeswithin network
▫ 8 bits at UNI
▫ 12bits at NNI
ATM
▫ Virtual Circuit Identifier VCI
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 237/359
▫ Identification/routing purposes within network▫ 16bit field
▫ Payload Type PT
Type of payload carried within a cell
▫ user data
▫ operation and maintenance data (OAM)
▫ congestion indication (CI) bit
▫ CI bit may be modified by any switch to indicate
congestion to end users
ATM
▫ PT Interpretation
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 238/359
000 User Data; type 0; no congestion
001 User Data, type 1; no congestion
010 User Data; type 0; congestion
011 User Data; type 1; congestion
100 OAM Cell
101 OAM Cell
110 Resource Management Cell (to be defined)
111 Reserved for future use
ATM
▫ Cell loss Probability CLP : (1 bit)
▪ Indicates relative priority of a cell
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 239/359
▪ Indicates if a cell can be discarded in case ofcongestion
▪ CLP = 0; High priority; cell not to be discarded
▪ CLP = 1; Low priority; cell may be discarded
▪ CLP bit is set by the user or by the service
provider
▫ Header error check HEC
▪ 8bit field
▪ cyclic redundancy check CRC on first four bytes(32 bits) of header
▪ Permanent Virtual Circuits (PVC)
▫ Direct Connectivity
Circuit Provision
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 240/359
▫ No call setup procedure
▪ Switched Virtual Circuits (SVC)
▫ Created & released dynamically
▫ Flexible Connection
▪ Connectionless services
▫ Multipoint SVC’s used for broadcast IPpackets
•IP over ATM implementation
•RFC-2684 ( PVC‟s, SVC‟s and multipoint SVC‟s )
•RFC-2225 (SVC‟s)
VCs are grouped together to create VPs and VPs are
Recap - Virtual Connections
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 241/359
bundled together to create a transmission path
Virtual Channel
Virtual Path
Transmission Path
ATM Reference Model
OSI Reference Model ATM Reference Model
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 242/359
Higher Layers
ATM Adaptation Layer
(AAL)
ATM Layer
Physical Layer
MANAGEMENT
LAYER
P
LA
N
E
CONTROL USER
Physical Layer
Data Link
Network
Transport
Session
Presentation
Application
Planes (Application Functions)
ATM Reference Model
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 243/359
▪ Control Plane
▪ Generating & managing signallingrequests
▪User Plane
▪ Managing the transfer of user data
▪ Management Plane
▪ Layer management
▪ Plane management
ATM Reference Model
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 244/359
Layers▫ ATM Layer
▫ ATM Adaptation Layer (AAL)
▫ Higher layers above AAL accept user data,
arrange it into packets and hand it to AAL.
▪ The ATM Layer is responsible for:
ATM Layers
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 245/359
▫ Addition/removal of correct cell header
▫ Multiplexing cells into a single stream received fromAAL
▫ Demultiplexing of cells and relaying to AAL
▫ Handling of Connection Identifiers
▫ Generic flow control
ATM Ad i L AAL
ATM Layers
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 246/359
▪ ATM Adaptation Layer, AAL
▫ Convergence function between user layer and ATM layer
▫ Conversion of source information to 48 byte segments
▫ Allocates service classes to support various information
sources
ATM Adaptation Layer
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 247/359
AAL1 AAL2 AAL3/4 AAL5
Timing
Bit rate
Mode
Yes No
Constant Non-real time
Connection oriented Connectionless
Yes No
Real-time, variable NRT Variable
ATM
▪ AAL Types▫ AAL1
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 248/359
▪ Constant bit rate CBR Example: Circuit Emulation
▪ Connection oriented
▪ Timing information exists
▫ AAL2
▪ Real Time variable bit rate VBR Eg: Traffic UMTS Node B -RNC
▪ Connection oriented
▪ Requires timing information
▪ Ex: Compressed video
ATM
▪ AAL 3/4
ATM
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 249/359
▪ Non-real time VBR Example: FrameRelay
▪ Connection oriented or connectionless
▪ No timing information
▪ AAL5▪ Variable bit rate Eg: Packet data and
signalling UMTS
▪ Connection oriented
▪ No timing information
▪ Simpler than AAL 3/4▪ Started in ITU; Completed in ATM Forum
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 250/359
ATM Segmentation & Assembly
ATM AAL2
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 251/359
▪ The principle ATM adaptation layer specified forUTRAN connection is AAL2
▪ AAL2 provides for real-time fixed delay traffic
ATM AAL2
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 252/359
▪ The AAL2 receives from the ATM layerinformation in the form of a 48 byte ATM ServiceData Unit (ATM-SDU).
▪ The AAL2 passes to the ATM layer information inthe form of a 48 byte ATM-SDU.
AAL2 Common Part Sublayer (CPS)
▪ The AAL2 CPS provides the capabilities to transfer CPS-SDUsfrom one CPS user to another CPS user through an ATM network.
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 253/359
▪ Two types of CPS users are supported:
▫ Service Specific Convergence Sublayer, SSCS entities
▫ Layer Management.
Iub interface
▪ The service offers a peer-to-peer operation:
▫ data transfer of CPS SDUs of up to 45 (default) or 64 bytes
AAL2 Common Part Sublayer (CPS)
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 254/359
▫ data transfer of CPS-SDUs of up to 45 (default) or 64 bytes▫ multiplexing and demultiplexing of multiple AAL type 2 channels
▪ CPS-SDU sequence integrity is maintained on each AAL type2 channel. ( useful for voice ! )
▪ However some CPS-SDUs may be lost; and lost CPS-SDUswill not be retransmitted. ( pointless for voice )
▪ The AAL2 CPS possesses the following characteristics:
AAL2 Common Part Sublayer (CPS)
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 255/359
▫ The AAL2 channel is a bi-directional virtual channel.
▫ The same value of channel identifier value is used for bothdirections.
▫ AAL2 channels can be established over
▪ ATM layer Permanent Virtual Circuit (PVC)
▪ Switched Virtual Circuit (SVC).
▪ The Common Part Sublayer merges several streams of CPS-
Packets onto a single ATM connection
Procedure of AAL2 Common Part Sublayer (CPS)
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 256/359
Packets onto a single ATM connection.
▪ The Common Part Sublayer receives CPS-SDUs from one ormore SSCS transmitter processes.
▪ CPS then multiplexes and packs the CPS-Packets into CPSProtocol Data Units, CPS-PDUs.
▪ At the CPS receiver, the CPS-Packets are unpacked anddemultiplexed and passed to one of the SSCS receivers.
▪ A CPS-Packet consists of a 3 byte CPS-Packet Header (CPS-
PH) followed by a CPS Packet Payload (CPS PP)
Format and coding of the CPS-Packet
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 257/359
PH) followed by a CPS-Packet Payload (CPS-PP).▫ The size and positions of the fields of the CPS-Packet are shown
below
▪ Channel Identifier (CID) „maximum 255 different sources‟
CPS Packet
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 258/359
▪ Channel Identifier (CID) maximum 255 different sources
▫ The CID value identifies the AAL2 CPS user of the channel.
▫ The AAL2 channel is a bi-directional channel.
▫ The same value of channel identification is used for bothdirections.
▫ Only 248 used for user data
▪ Length Indicator (LI)
▫ The LI field is binary encoded with a value that is one less than the
number of bytes in the CPS-Packet Payload
CPS Packet
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 259/359
number of bytes in the CPS-Packet Payload.
▫ The default maximum length of the CPS-Packet Payload is 45bytes; otherwise, the maximum length can be set to 64 bytes.
▫ The maximum length is channel specific, i.e. its value need not becommon to all AAL2 channels.
▫ However, for a given CID value, all CPS-Packet payloads mustconform to a common maximum value.
▪ User-to-User Indication (UUI)
▫ The UUI field serves two purposes:
CPS Packet
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 260/359
▫ The UUI field serves two purposes:
▪ to convey specific information transparently betweenthe CPS users, i.e. between SSCS entities orbetween Layer Management
▪ to distinguish between the SSCS entities and LayerManagement users of the CPS
▪ User-to-User Indication (UUI)
▫ The 5-bit UUI field uses
CPS Packet
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 261/359
The 5-bit UUI field uses
▪ 0 to 27 for SSCS entities,
▪ 30 to 31 for Layer Management.
▪ The values 28, 29 are reserved for futurestandardization.
▫ The contents of the UUI field are used to transport the UUIparameters
▪ CPS-UNITDATA primitive
▪ MAAL-UNITDATA primitive
▪ Header Error Control (HEC)
▫ The transmitter shall calculate the remainder of the division(modulo 2) by the generator polynomial x5 x2 1 of the
CPS Packet
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 262/359
The transmitter shall calculate the remainder of the division(modulo 2), by the generator polynomial x5 x2 1, of theproduct of x5 and the contents of CID,LI and UUI (19 bits) ofthe CPS-PH.
▫ The coefficients of the remainder polynomial shall beinserted in the HEC field with the coefficient of the x4 term
in the most significant bit of the HEC field.▫ The receiver uses the contents of the HEC field to detect
errors in the CPS-PH.
▫ Now that the CPS-Packet is formed we can put it into aCPS Protocol Data Unit. This has a single byte headerconstructed as follows.
CPS-PDU ▪ Format and coding of the Common Packet Sublayer-Protocol Data Unit
CPS-PDU
▫ The CPS-PDU consists of a single byte start field and a 47-byte payload.
▫ The 48-byte CPS-PDU is the ATM-SDU.
Th i d iti f th fi ld f th CPS PDU h b l
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 263/359
▫ The size and positions of the fields of the CPS-PDU are shown below.
▪ CPS-PDU start field (STF) (8 bits)
▪ a) Offset Field (OSF) (6 bits)
▫ This field carries the binary value of the offset, measured in number ofbytes between the end of the STF and the first start of a CPS Packet or in
CPS-PDU
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 264/359
s e d ca es t e b a y a ue o t e o set, easu ed u be obytes, between the end of the STF and the first start of a CPS-Packet or, inthe absence of a first start, to the start of the PAD field.
▫ The value 47 indicates that there is no start boundary in the CPS-PDUpayload.
▫ Values greater than 47 are not allowed.
▪ b) Sequence Number (SN) (1 bit)
▫ This bit is used to number (modulo 2) the stream of CPS-PDUs.
▫ Alternating 0,1.
▪ c) Parity (P) (1 bit)
▫ This bit is used by the receiver to detect errors in the STF. The transmittersets this bit value such that the parity over the 8-bit STF is odd.
▪ The CPS-PDU payload may carry zero, one or more (completeor partial) CPS-Packets
CPS-PDU Payload
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 265/359
▪ CPS-Packets have a maximum size of 64 bytes.
▪ Unused payload is filled with padding bytes coded with thevalue zero.
▪ A CPS-Packet may overlap one or two ATM cell boundaries.
▪ The overlap point where the CPS-Packet becomes partitionedmay occur anywhere in the CPS-Packet, including the CPS-Packet header.
▪ AUU = ATM-User-to-ATM-User Indication = 0
▪ SLP = Submitted (Cell) Loss Priority = 0
▪ CI = Congestion Indication = 0
Additional ATM-DATA
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 266/359
CI Congestion Indication 0▫ CI "1" indicates that congestion has been encountered
either before transmission or during transfer
▪ CPS Interface Data (CPS-INFO)
▫ This parameter specifies the interface data unitexchanged between the CPS and the SSCS entity.
▫ The interface data is an integral multiple of one byte.
▫ The CPS Interface Data represents a complete CPS-SDU.
▪ CPS User-to-User Indication (CPS-UUI) ▫ This parameter is transparently transported by the CPS
between peer CPS users.
ITU SDL diagrams
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 267/359
▪ The process of forming ATM CPS-Packets will beattempted through the use of SDL diagrams providedby the ITU.
▪ Copyright has been approved for the use of thesediagrams in this course.
SDL
▪ Due to the international nature of telecommunications and theenormous investment needed to develop large telecommunicationsystems, a structured method for recording designs has been
proposed.
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 268/359
p p
▪ The Specification and Description Language enables teams of peopleto work on specific aspects of a system and to communicate theirdesigns in a coordinated way.
▪ The objective of SDL is to break down a system into manageableblocks and processes. These processes can be defined using states,signals and tasks.
SDL Concepts
▪ In complex systems like PABX‟s a large number of independent events occur simultaneously, and any
specification method must be able to cater for theseconcurrent events in a simple manner
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 269/359
pconcurrent events in a simple manner.
▪ The concept of a process for describing concurrent events isimportant.
▪ A process may be a physical item such as a line interface, oran abstract item such as an administrative procedure.
▪ Events are the occurrences that are relevant to the behaviourin question.
▪ The following pictures illustrate how a user's requirements can havevarious interpretations (Meek and Heath, 1980).
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 270/359
Graphical SDL
▪ Start
▪ State
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 271/359
▪ Input
▪ Output
▪ Task
▪ Decision
▪ Stop
▪ Connection
ATM Packet Assembler
ATM Segmentation Process
I.363.2 SDL Diagrams for the CPS transmitter
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 272/359
ATM Packet Assembler
ATM Segmentation Process
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 273/359
▪ This ATM-DATA is an ATM-SDU
▪ With the addition of a cell header, 5 bytes this becomes anATM cell
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 274/359
UTRAN Signals & Signaling
Radio Frame Structure
▪ Radio Frame Period Tf = 10ms
▪ Frames are used for channel format control
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 275/359
▪ 15 slots, #0…#14
▪ Slots are used for power control, and synchronisation.
Tslot = 666.7s = 2560 chips
#0 #1 #2 #i #14
Tf = 10ms = 38400 chips
The Common Pilot Channel CPICH
▪ UTRA has two types of pilot channel
▫ Primary
▪ always uses primary scrambling code – choice of 512fixed channelisation code all 1‟s
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 276/359
▪ fixed channelisation code, all 1 s
▪ only one per cell/sector
▫ Secondary
▪ can have any channelisation code of length 256
▪ may use a secondary scrambling code – choice of 1 -15for each P-CPICH
▪ The P-CPICH is defined by the power in its signal and takesthe form
ceinterferen overalltheof thattobitspilotin theenergyof ratio o
c
I
E
The Primary Synchronisation Channel
▪ The Downlink Synchronisation Channel SCH, transmits the PrimarySynchronisation Code, P-SCH
▪ This is a 256 chip sequence and is the same in all cells in the network
▪ The channel is transmitted at the start of a timeslot, for the first 66.67s
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 277/359
The channel is transmitted at the start of a timeslot, for the first 66.67s
P-SCH P-SCH P-SCH
256 chips66.67s
2560 chips666.7s
Timeslot # 0 Timeslot # 1 Timeslot # 2
The Primary Synchronisation Channel
▪ The Primary Synchronisation Code, P-SCH is formed by:
▫ a = <1,1,1,1,1,1,-1,-1,1,-1,1,-1,1,-1,-1,1> 16 bits
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 278/359
▫ Cp-sch= (1+ j)×< a,a,a,-a,-a,a,-a,-a,a,a,a,-a,a,-a,a,a>
a 16×16 complex array
▪ The P-SCH is chosen for good aperiodic auto correlation
▪ No scrambling is used, & the P-SCH is already spread
▪ This is defined for all networks
The Secondary Synchronisation Channel
▪ The downlink SCH also transmits the Secondary Synchronisation Code.
▪ This is a 256 chip sequence and uses a defined bit pattern
▪ The channel is transmitted at the start of a timeslot, for the first 66.67s, atthe same time as the P-SCH
▪ The S-SCH indicates which one of 64 groups of downlink scrambling
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 279/359
g p gcodes is in use in the cell
P-SCH P-SCH P-SCH
256 chips66.67s
S-SCH S-SCH S-SCH
2560 chips666.7s
Timeslot # 0 Timeslot # 1 Timeslot # 2
The Secondary Synchronisation Channel
▪The Secondary Synchronisation Code can use time-switched transmit antenna diversity (TSTD) and is the only
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 280/359
y ( ) ychannel in UTRA FDD that uses TSTD.
▪ The S-SCH have identical real and imaginary components,
does not use scrambling & is already spread.
▪ This is defined for all networks
The Secondary Synchronisation Channel
▪ The S-SCH is used to find the scrambling code group, 1 of 64and the timeslot number.
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 281/359
▪ By following the sequence of S-SCH codes the UE candetermine where it is in the matrix of group codes and itscolumn position.
Exercise
Locate which group code and time slot position the UE is at, from the following data
15,16,10,7,8,1,10,8,2,16,9,15,1,9,2,15
The Secondary Synchronisation Channel
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 282/359
▪ The PCCPCH is transmitted continuously at constant power from eachcell and carries the Logical Broadcast Channel BCH
▪ Uses one of the 512 Primary Scrambling Codes
▪ There is only one PCCPCH per cell
The Primary Common Control Physical Channel
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 283/359
There is only one PCCPCH per cell
27 kbps, SF=256
P-SCH P-SCH P-SCH
256 chips
66.67s
S-SCH S-SCH S-SCH2560 chips
666.7s
Timeslot # 0 Timeslot # 1 Timeslot # 2
2304 chips600s
Data (18 bits)Data (18 bits) Data (18 bits)PCCPCH
30 kbps, SF=256
The Primary Common Control Physical Channel
▪ The PCCPCH does not transmit in the first 256 chips ofeach timeslot.
▫ This provides space for the Primary and SecondarySynchronisation codes to be broadcast.
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 284/359
y
▪ Used with the Common Pilot Channel to provide forchannel estimation.
▪ To improve performance
▫ The PCCPCH can use open-loop transmission diversity.
▪ There is only one PCCPCH per cell
▪ The SCCPCH carries two different commontransport channels
▫ Forward Access Channel FACH
The Secondary Common Control Physical Channel
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 285/359
▫ Paging Channel PCH
▪ These two channels can share the same physicalchannel or use separate ones.
▪ There can be multiple SCCPCH channelsdepending on the load
▪ SCCPCH can carry Packet Switched traffic if theFACH load is low
▪ The spreading factor for SCCPCH is fixed at themaximum data rate for FACH and PCH.
The Secondary Common Control Physical Channel
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 286/359
▪ Sharing the same channel saves power as the signalmust be broadcast at a power level to cover entire cell
▪ A separate paging indicator channel PICH is used to getthe attention of the UE.
▪ The Paging channel and Paging Indicator channel areoperated together.
Paging Indicator Channel PICH
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 287/359
▪ The paging indicators use a channelisation code of 256.
▪ A PICH radio frame is of length 10ms.
▪ PICH carries 288 bits and 12 bits are left idle
Paging a mobile: PCH
▪ Paging information is carried on the paging channel,PCH, a downlink common channel.
▪ Each terminal is allocated a paging group.
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 288/359
▪ Each group listens periodically to a Paging IndicationChannel, PICH.
▪ How often a mobile must listen is governed by itspaging group. If the mobile was always on, battery lifewould be very short.
Paging Indication Channel PICH▪ Fixed rate (SF=256, 30 kbps) so that 300 bits occupy a full frame.
▪ N Paging Indicators {PI0, …, PINp-1} are transmitted in each PICH
frame, where Np=18, 36, 72, or 144.
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 289/359
b1b0
288 bits for paging indication 12 bits (undefined)
One radio frame (10 ms)
b287 b288 b299
▪ These are mapped into 300 bits of which 288 bits are defined.
PICH
▪ A „1‟ in the appropriate bit position indicates to the UE that it
should decode the next PCH
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 290/359
PICH
S-CCPCH
Paging Indicators
Paging Message
▪ Physical Random Access Channel PRACH
▫ Used to carry the Random Access Channel RACH
Common Uplink Physical Channel
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 291/359
▫ Based on Slotted ALOHA
▫ UE can start the random-access transmission atdefined time intervals
▫ There are 15 access slots per frame, spaced 5120chips apart.
▪ The RACH procedure has the following steps;
▫ UE decodes the BCH to find
▪ RACH sub-channels
Random Access Channel RACH
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 292/359
▪ Scrambling codes & signatures
▫ UE randomly selects one of the RACH signatures
▫ UE randomly selects one of the RACH sub-channels
▫ The RACH access slots are taken over 2 × 10ms frames.
▪ The RACH procedure continues with the following steps;
▫ Downlink power level is measured
▫ A preamble of 4096 chips is sent with selected signature
Random Access Channel RACH
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 293/359
p p g
▫ This is a repeating sequence of a 16 bit preamble
▫ Waits for a Acquisition Indicator channel AICH preamble
▫ When AICH is detected a 10ms or 20ms message istransmitted.
RACH Signatures ( 16 bits )
P0(n) 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
P1(n) 1 -1 1 -1 1 -1 1 -1 1 -1 1 -1 1 -1 1 -1
P2(n) 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1
P3(n) 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1
P4(n) 1 1 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1 -1
P5(n) 1 -1 1 -1 -1 1 -1 1 1 -1 1 -1 -1 1 -1 1
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 294/359
P5(n) 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
P6(n) 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1 -1 1 1
P7(n) 1 -1 -1 1 -1 1 1 -1 1 -1 -1 1 -1 1 1 -1
P8(n) 1 1 1 1 1 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1
P9(n) 1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1
P10(n) 1 1 -1 -1 1 1 -1 -1 -1 -1 1 1 -1 -1 1 1
P11(n) 1 -1 -1 1 1 -1 -1 1 -1 1 1 -1 -1 1 1 -1
P12(n) 1 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 1 1
P13(n) 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 1 -1 1 -1
P14(n) 1 1 -1 -1 -1 -1 1 1 -1 -1 1 1 1 1 -1 -1
Orange(dark) represents a 1 and Blue(light) represents a -1
▪ The AICH is a fixed rate, SF=256, physical channel whichcarries the Acquisition Indicators, AI.
▪ The AI corresponds to the PRACH signature chosen by the
Acquisition Indicator Channel AICH
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 295/359
▪ The AIs corresponds to the PRACH signature chosen by theUE.
▪ The phase reference for the AICH is the primary CPICH
▪ Frame duration is 20msec, slot length 5120 chips ( 2 normalTS‟s)
▪ 4096 chips are used with 1024 chips with no transmission
▪ AICH consists of 32 symbols created as follows:
▪ AICH consists of 32 symbols created as follows:
Acquisition Indicator Channel AICH
jjbAIa 15
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 296/359
jss
s jb AI a
,0
▪ a j = <a0,a1,a2…….a31>
▪ bs,j is the sequence bs,0, …. bs,31
▪ AIs can take values 1, or 0
AICH signature patterns
0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1
2 1 1 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1 -1
3 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1 -1 1 1
4 1 1 1 1 1 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 1 1 1 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1
5 1 1 -1 -1 1 1 -1 -1 -1 -1 1 1 -1 -1 1 1 1 1 -1 -1 1 1 -1 -1 -1 -1 1 1 -1 -1 1 1
P0(n) 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
P1(n) 1 -1 1 -1 1 -1 1 -1 1 -1 1 -1 1 -1 1
P2(n) 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1 -1
P3(n) 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1
P4(n) 1 1 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1
P5(n) 1 -1 1 -1 -1 1 -1 1 1 -1 1 -1 -1 1 -1
AIs bs,j
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 297/359
6 1 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 1 1 1 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 1 1
7 1 1 -1 -1 -1 -1 1 1 -1 -1 1 1 1 1 -1 -1 1 1 -1 -1 -1 -1 1 1 -1 -1 1 1 1 1 -1 -1
8 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
9 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1
10 1 1 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1 -1 1 1 1 1
11 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1 -1 1 1 -1 -1 1 1 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1
12 1 1 1 1 1 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 1 1 1 1 1 1
13 1 1 -1 -1 1 1 -1 -1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1 -1 -1 1 1 1 1 -1 -1 1 1 -1 -1
14 1 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1 -1 1 1 1 1 1 1 1 1 -1 -1 -1 -1
15 1 1 -1 -1 -1 -1 1 1 -1 -1 1 1 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1 1 1 -1 -1 -1 -1 1 1
( )
P6(n) 1 1 -1 -1 -1 -1 1 1 1 1 -1 -1 -1 -1 1
P7(n) 1 -1 -1 1 -1 1 1 -1 1 -1 -1 1 -1 1 1
P8(n) 1 1 1 1 1 1 1 1 -1 -1 -1 -1 -1 -1 -1
P9(n) 1 -1 1 -1 1 -1 1 -1 -1 1 -1 1 -1 1 -1
P10(n) 1 1 -1 -1 1 1 -1 -1 -1 -1 1 1 -1 -1 1
P11(n) 1 -1 -1 1 1 -1 -1 1 -1 1 1 -1 -1 1 1
P12(n) 1 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 1 1 1
P13(n) 1 -1 1 -1 -1 1 -1 1 -1 1 -1 1 1 -1 1
P14(n) 1 1 -1 -1 -1 -1 1 1 -1 -1 1 1 1 1 -1
P15(n) 1 -1 -1 1 -1 1 1 -1 -1 1 1 -1 1 -1 -1
▪ a j is formed by incrementing s from 0 to 15, whichever „s‟ is used bythe UE will have a value 1, the rest will be 0
AICH signature patterns
▪ Example lets choose P8(n) = AI8
115151221111001
0,15150,220,110,000
.....
.....
b AI b AI b AI b AI a
b AI b AI b AI b AI a
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 298/359
31,151531,2231,1131,0031
2,15152,222,112,002
1,15151,221,111,001
.....
.
.....
.....
b AI b AI b AI b AI a
b AI b AI b AI b AI a
bbbba
0,80,880 1 bb AI a As all other elements have AI = 0
▪ Therefore „a‟ will select the AICH „b‟ signature pattern corresponding to thetransmitted RACH preamble
AICH signature patterns
▪ Continuing
▫ a0 to a31 has been found.
▫ „a‟ is then spread and modulated as bits with a SF of 256
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 299/359
▫ a is then spread and modulated as bits, with a SF of 256
▫ The modulation scheme halves the bit rate ie 1 symbol for 2
bits
▫ 32 × 256 ÷ 2 = 4096 chips + 1024 chips per slot = 5120
chips
▫ 5120 × 15 = 76800 chips in 20msec = 3,840,000 chips
▫ Once the UE has detected the AICH, the UE becomes part
of the active set for that cell.
PRACH
▪ The PRACH consists of two parts
▫ A preamble
▪ To initiate access▫ A message
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 300/359
A message
▪ Which can contain a request for a dedicated channel or asmall packet of user data
2 frames = 20 ms
1 PRACH slot = 2 normal timeslots 1 PRACH preamble = 4096 chips
PRACH Message
Acquisition Indicator Channel AICH
▪ The AICH indicates whether the PRACH preamble has beenreceived.
▪ If the Node-B receives the preamble it mirrors the preamblesignature back on the AICH
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 301/359
g
2 frames = 20 ms
1 PRACH slot = 1.25ms
1 PRACH preamble = 4096 chips
PRACH
AICH
Message
1 PRACH preamble = 4096 chips
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 302/359
UTRAN Synchronisation
Node Synchronisation
▪ Node Synchronisation is required to provide a common timingreference among different Node Bs.
▪ In the UTRAN a common timing reference among all the nodes
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 303/359
g gis not provided.
▪ To minimise delay and buffering on the air interface, Uu,
estimates of the timing differences between RNC and Node Bs,are made without the need to compensate for the phasedifferences between RNC's and Node B's clocks.
Synchronisation
▪ Time Alignment
RNC
CN
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 304/359
▪ Transport ChannelSynch
▪ Radio InterfaceSynch
Node B
Node B
UE
TDD only
Time Alignment Synchronisation
▪ The time alignment handling procedure over Iu relates to the
control of DL transmission timing in the CN nodes in order tominimise the buffer delay in the SRNC.
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 305/359
▪ The SRNC invokes this procedure whenever a Iu User PlaneProtocol Data Unit, PDU, is inappropriately timed.
▪ The SRNC indicates to the CN by means of a Time Alignmentcontrol frame.
▪ The delay is by a number of +/- 500µsec steps.
Time Alignment Synchronisation
CN
ACK
Timing Alignment n × 500µs
▪ A supervision timer TTA is started after the SRNC sends a timingalignment control frame
1ms
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 306/359
SRNC
TTA
or
NACK
If TTA expires then the SRNC will sendanother timing alignment control FrameTTA starts
Transport Channel: RNC - Node B Synchronisation
▪ RNC - Node B synchronisation is used for determining gooddownlink DL and uplink UL offset values.
▪ Measurements of node offsets can be made at the start or
restart, as well as during normal operation, to supervise thestability of the nodes.
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 307/359
y
▪ Procedure:-
▫ the RNC sends a DL Node Synchronisation control frame to NodeB containing the parameter t1.
▫ Upon reception of a DL Synchronisation control frame, the NodeB shall respond with UL Synchronisation Control Frame to theRNC, indicating t2 and t3, as well as t1.
Conference Call Example
▪ Assume Conference call in Dallas
▪ You are in England, the time difference is 6 hours
▪ You must control the meeting ( ie CRNC )
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 308/359
▪ You send out email requests for staff to meet at 6pm
▪ The meeting room can be thought of as the Node B
▪ The staff meeting in the room are the UE‟s, they aresynchronised with England time as is the Node B
▪ You ( CRNC ) need to be in your office delivering your speech at12 noon
RNC - Node B Synchronisation
SRNC 4094 4095 0 1 2
3
RFN
t1 t4
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 309/359
Node B
UL Node Synch
[t1=4094.3, t2=201.2, t3=202.25 ]
200 201 202 203 204 205
BFN
DL Node Synch [t1=4094.3]t2 t3
RNC - Node B Synchronisation
SRNC
▪ These two paths ( t2 - t1 ) + ( t4 - t3) give the Round Trip Delay RTD
4094 4095 0 1 23
RFN
t1 t4
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 310/359
Node B
200 201 202 203 204 205
BFN
t2 t3
[t1=4094.3, t2=201.2, t3=202.25, t4=2.3 ]
(t2-t1) + (t4-t3) = (t4-t1) – (t3-t2)
So 2.3 – 4094.3 = -4092 + 4096 = 4, 202.25 – 201.2 = 1.05
Therefore RTD = 4 – 1.05 = 2.95 x 10 msec = 29.5 msec
Transport Channel Synchronisation
▪ The Transport Channel (or L2) synchronisation, provides anL2 common frame numbering between UTRAN and UE.
▪ This frame number is the Connection Frame Number(CFN) and it is associated at L2 to every transport block
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 311/359
(CFN), and it is associated at L2 to every transport blockset, TBS and passed to L1.
▪ The CFN is not transmitted in the air interface for eachTBS, but is mapped by L1 to the cell System FrameNumber, SFN of the first radio frame used for thetransmission of the TBS (the SFN is broadcast at L1 in thebroadcast channel, BCH).
▪ The mapping is performed via the Frame_offset parameter.
▪ In case of soft handover, the Frame_offsets of the differentradio links are selected in order to have a timedtransmission of the diversity branches in the air interface.
Counters and parameters
▪ Counters and parameters as used in the different UTRANsynchronisation procedures.
▪ BFN & RFN, Node B and RNC Frame Number counter. These are theNode B/RNC common frame number counters BFN/RFN can be
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 312/359
Node B/RNC common frame number counters. BFN/RFN can beoptionally frequency-locked to a Network sync reference.
▫ Range: 0 to 4095 frames, 12 bits.
▪ SFN Cell System Frame Number counter. SFN is sent on BCCH onLayer 1. SFN is used for paging groups and system informationscheduling etc.
▫ In FDD SFN = BFN adjusted with T_cell.
Counters and parameters
▪ CFN, Connection Frame Number, is the frame counter used forthe L2/transport channel synchronisation between UE andUTRAN.
▫ A CFN value is associated to each TBS and it is passed together withit through the MAC-L1 SAP.
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 313/359
▪ CFN provides a common frame reference (at L2) to be used e.g.for synchronised transport channel reconfiguration.
▫ Range: 0 to 255 frames, 8 bits.
▫ When used for PCH the range is 0 to 4095 frames, 12 bits.
Counters and parameters
▪ Frame_offset is a radio link specific L1 parameter used to map the CFN intothe SFN that defines the specific radio frame for the transmission on the airinterface.
▪ At the L1/L2 interaction, the mapping is performed as:▫ LSB8(SFN) = CFN + Frame_offset (from L2 to L1)
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 314/359
▫ CFN = LSB8(SFN) - Frame_offset (from L1 to L2)
▫ The resolution of all three parameters is 1 frame.
▫ Frame_offset and CFN have the same range (8 bits, 0…255) and only the 8 leastsignificant bits of the SFN are used.
▫ The operations are modulo 256. ie 0 .. 255
▪ In the UTRAN, the Frame_offset parameter is calculated by the SRNC andprovided to the Node B.
Counters and parameters
▪ OFF The parameter OFF is calculated by the UE and reportedto the UTRAN only when the UTRAN has requested the UE tosend this parameter.
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 315/359
▪ In the neighbouring cell list, the UTRAN indicates for each cellif the Frame_offset is already known by the UTRAN or shall bemeasured and reported by the UE.
▫ OFF has a resolution of 1 frame and a range of 0 to 255.
Counters and parameters
▪ The DOFF (Downlink-Offset) is used to define Frame_offset and Chip_offsetduring the first radio link RL setup.
▫ The resolution should be good enough to spread load over the Iub and Node B (basedon certain load distributing algorithms).
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 316/359
▫ In addition it is used to spread out the location of the Pilot Symbol in order to reducethe peak downlink power since the Pilot symbol is transmitting at the slot boundary(the largest chips for one symbol is 512 chips).
▪ The SRNC sends a DOFF parameter to the UE when the new RL makes the UEchange its state to the dedicated channel state.
▫ Resolution: 512 chips; Range :0 to 599 (<80ms).
Counters and parameters
▪ The Chip_offset is used as the offset for the DL DPCH relative to thePCCPCH timing.
▫ The Chip_offset parameter has a resolution of 1 chip and a range of 0 to 38399 (<
10ms).▫ The Chip_offset parameter is calculated by the SRNC and provided to the Node B.
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 317/359
▪ Frame_offset + Chip_offset (sent via NBAP) are rounded together to theclosest 256 chip boundary in Node B.
▪ The 256 chip boundary is used regardless of the used spreading factor, alsowhen the spreading factor is 512. The rounded value (which is calculated inNode B) controls the DL DPCH air-interface timing.
Counters and parameters
▪ The reported Tm parameter has a resolution of 1 chip and a range of 0to 38399. The Tm shall always be sent by the UE.
▪ T_cell represents the Timing delay used for defining the start of SCH,CPICH and the downlink Scrambling Code(s) in a cell relative to BFN.
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 318/359
▪ The main purpose is to avoid having overlapping SCHs in different cellsbelonging to the same Node B.
▫ A SCH burst is 256 chips long.
▫ SFN in a cell is delayed T_cell relative to BFN.
▫ Resolution: 256 chips. Range: 0 .. 9 x 256 chips.
Counters and parameters
▪ t1 RNC specific frame number (RFN) that indicates the time whenRNC sends the frame through the SAP to the transport layer.
▪ t2 Node B specific frame number (BFN) that indicates the time
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 319/359
t2 Node B specific frame number (BFN) that indicates the timewhen Node B receives the corresponding DL synchronisationframe through the SAP from the transport layer.
▪ t3 Node B specific frame number (BFN) that indicates the timewhen Node B sends the frame through the SAP to the transportlayer.
Counters and parameters
▪ TOAWS (Time of Arrival Window Startpoint) is the window startpoint.
▫ DL data frames are expected to be received after this window startpoint.
▫ TOAWS is defined with a positive value relative Time of Arrival WindowEndpoint (TOAWE).
▫ A data frame arriving before TOAWS gives a Timing Adjustment Control frameresponse The resol tion is 1 ms the range is {0 CFN length/2 1 ms}
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 320/359
response. The resolution is 1 ms, the range is: {0 .. CFN length/2 –1 ms}.
▪ TOAWE (Time of Arrival Window Endpoint) is the window endpoint.
▫ DL data frames are expected to be received before this window endpoint.
▫ TOAWE is defined with a positive value relative Latest Time of Arrival (LTOA).
Counters and parameters
▪ LTOA (Latest Time of Arrival) is the latest time instant a Node Bcan receive a data frame and still be able to process it.
▫ Data frames received after LTOA can not be processed (discarded).
▫ LTOA is defined internally in Node B to be a processing time beforethe data frame is sent in air-interface.
Th i ti (T ) ld b d d i d d t
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 321/359
▫ The processing time (Tproc) could be vendor and service dependent.
▪ TOA (Time of Arrival) is the time difference between theTOAWE and when a data frame is received.
▫ A positive TOA means that data frames are received beforeTOAWE.
▫ A negative TOA means that data frames are received after TOAWE.
Counters and parameters
▪ Data frames that are received after TOAWE but before LTOA(TOA+TOAWE>=0) are processed by Node B.
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 322/359
▫ TOA has a resolution of 125 µs.
▫ TOA is positive when data frames are received before TOAWE.
▫ The range is: {0 .. +CFN length/2 –125 µs}.
▫ TOA is negative when data frames are received after TOAWE.
▫ The range is: { –125 µs .. –CFN length/2}.
Transport Channel Synchronisation ▪ The DL Data frame number is calculated with the delay in mind. So the
RNC assumes the Node B will process the data at CFN 202.
RNC
194 195 196 197 198 199
CFN
DL Data Frame 202 UL Data Frame 202
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 323/359
Node B
200 202
1235
204
1237 SFN
UE DL
CFN
CFN
200
1233
202 204
CFN to SFN is the frame offset
8Lsb(SFN) - CFN = 8
Receiving Window
TOA
Synchronisation Example
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 324/359
Synchronisation Example
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 325/359
Frame arrives too early
Synchronisation Example
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 326/359
Frame arrives too late
FDD Radio Interface Synchronisation
▪ FDD Radio Interface Synchronisation makes sure that the
UE gets the correct frames when receiving from severalcells.
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 327/359
▪ The UE measures the Timing difference between its DPCHand SFN in the target cell when doing handover and reportsit to SRNC.
▪ SRNC sends this Time difference value in two parametersFrame_offset and Chip_offset over Iub to Node B.
FDD Radio Interface Synchronisation
▪ Example: The following set of timing diagrams show an examplewith two cells connected to one UE where handover is done fromsource cell (Cell 1) to target cell (Cell 2).
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 328/359
BFN - Node B Frame Number
SFN - System Frame Number SFN is delayed by Tcell relative to BFN. Tcell is used to
skew cells in the same Node B in order to avoid
colliding with synchronisation channel SCH bursts.
FDD Radio Interface Synchronisation
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 329/359
DL - DownLink
DPCH - Dedicated Physical Channel
CFN - Connection Frame Number Chip_offset 2.6ms about a quarter of a frame
FDD Radio Interface Synchronisation
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 330/359
Tp1 is the propagation delay between Node B and the UE
TUE Tx The time when UE transmits on the UL DPCH
T0 is a constant of 1024 chips ( 266µs )
FDD Radio Interface Synchronisation
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 331/359
OFF is 2 frames
Tm is 3840chips
Tm - measured at UE at handover,
Tm has a range of 0 - 38399 chips
FDD Radio Interface Synchronisation
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 332/359
Frame_offset = 2
Chip_offset = 10240Tp - Propagation delay in uplink
FDD Radio Interface Synchronisation ▪ How to determine Tm at UE
▫ Select a time instant where frame N starts at DL SFN2 e.g. frame number
2058, the time from that time instant to the next frame border of DL
DPCHnom equals Tm (if these are in phase with each other, Tm is zero).▫ In the example Tm is 3840 which is 1ms
UTRAN Sychronisation
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 333/359
▪ How to determine OFF at UE
▫ The difference between the frame number selected for time instant (2058)and the frame number starting at instant (8) mod 256 frames equals OFF.
▪ Example:▫ (2058 – 8) mod 256 = 2, another example is (2056 – 6) mod 256 = 2.
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 334/359
UTRAN Signalling
UTRAN Signalling Procedures
▪ The signalling procedures shown in the following sections do not represent thecomplete set of possibilities.
▪ The standard specifies a set of EPs for each interface, which may be combined indifferent ways in an implementation.
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 335/359
▪ Therefore these sequences are at present merely examples of a typicalimplementation.
System Information Broadcasting
3. BCCH:System Information
1. System Information Update Request
UE Node B RNC CN
NBAPNBAP
RRCRRC
2. System Information Update Response NBAPNBAP
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 336/359
1. The RNC forwards a request to node B via a Node B application part NBAP message„System Information Update Request’.Parameters: Master/Segment Information Block(s) (System information to be broadcasted), BCCHmodification time.
4. BCCH: System Information RRCRRC
5. BCCH:System Information RRCRRC
2. The Node B confirms the ability to broadcast the information sendingSystem Information Update Response message to the RNC via NBAP.(If the Node B cannot Broadcast the information as requested, SystemInformation Update Failure is returned to the RNC).
3./4./5. The information is broadcast via BCCH, on the air interface byRRC message System Information.Parameters: Master/Segment Information Block(s) (System information).
Paging for a UE in RRC Idle Mode
▪ This example shows how paging is performed for an UE in radio resource controlRRC Idle Mode.
▪ The UE may be paged for both CS and PS services.
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 337/359
▪ Since the UE is in RRC Idle Mode, the location is only known at CN level andtherefore paging is distributed over a defined geographical area (Location Area).
Paging for a UE in RRC Idle Mode
UE Node B
1.1
Node B
2.1
RNC
1
RNC
2
CN
RANAPRANAP 1. Paging
RANAP RANAP1. Paging
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 338/359
2. PCCH : Paging Type 1
3. PCCH : Paging Type 1
1. CN initiates the paging of a UE over a LA spanning two RNCs (i.e.RNC1 and RNC2) via RANAP message Paging.Parameters: CN Domain Indicator, Permanent NAS UE Identity,Temporary UE Identity, Paging Cause.
2. Paging of UE performed by cell1 using Paging Channel PCCH PagingType 1 message.3. Paging of UE performed by cell2 using Paging Type 1 message. The UE detects page message from RNC1 (as example) and the procedure for
non-access stratum NAS signalling connection establishment follows. NAS
message transfer can now be performed.
NAS Signalling Connection Establishment
▪ This example shows establishment of a non-access stratum NASSignalling Connection.
▪ This establishment could be requested by the terminal by itself (forexample to initiate a service) or could be caused by a paging from
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 339/359
the CN.
NAS Signalling Connection Establishment
UE Serving
RNC
CN
1. RRC Connection Establishment
RRCRRC2. DCCH : Initial Direct Transfer
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 340/359
RANAP RANAP
3. Initial UE Message
1. Radio resource control RRC Connection isestablished.2. UE sends RRC Initial Direct Transfer to SRNC.Parameters: Initial NAS Message CN node indicator (it indicates the
correct CN node into which the NAS message shall be forwarded).
3. SRNC initiates signalling connection to CN, and sends theRANAP message Initial UE Message.Parameters: NAS PDU
The NAS signalling connection between UE and CN can now be usedfor NAS message transfer.
RRC Connection Establishment
▪ The following example shows establishment of a RRCconnection in a dedicated transport channel (DCH)state
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 341/359
Dedicated transport Channel DCH Establishment
Parameters: Signalling link termination,T l dd i i f i
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 342/359
1. The UE initiates set-up of an RRC connection by sending RRC messageConnection Request on common control channel CCCH.
Parameters: Initial UE Identity, Establishment cause, Initial UE Capability.
2. When a DCH is set-up, a NBAP message Radio Link Setup Request is sent to Node B.
3. Node B responses with NBAP message Radio Link Setup Response. 4.SRNC initiates set-up of Iub Data Transport bearer using ALCAP protocol.This request contains the AAL2 Binding Identity to bind the Iub DataTransport Bearer to the DCH. The request for set-up of Iub Data Transportbearer is acknowledged by Node B.
Transport layer addressing information(AAL2 address, AAL2 Binding Identity)for the Iub Data Transport Bearer.
Node B allocates resources, starts physical layer PHY receptionThe SRNC decides to use a DCH for this RRC connection, allocates aradio network temporary identity RNTI and radio resources for the RRCconnection.
DCH Establishment
Parameters: Initial UE Identity,RNTI, Capability update
Requirement, Transport FormatSet, Transport FormatCombination Set, frequency, DL
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 343/359
5./6.The Node B and SRNC establish synchronism for the Iub and IurData Transport Bearer by means of exchange of the appropriate DCHFrame Protocol frames Downlink Synchronisation and Uplink
Synchronisation.
7. Message RRC Connection Setup is sent on CCCH from SRNC toUE.
8. Message RRC Connection Setup Complete is sent on DCCH fromUE to SRNC.
Parameters: Integrity information, ciphering information.
Then Node B starts downlink DL transmission.
scrambling code (FDD only),Time Slots (TDD only), UserCodes (TDD only), Power
control information.
Soft Handover (FDD)
▪ Radio Link Addition (Branch Addition)
▫ This example shows establishment of a radio link via a Node Bcontrolled by a RNC other than the serving RNC.
▫ This is the first radio link to be established via this RNS, thus macro-
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 344/359
diversity combining/splitting with already existing radio links withinDRNS is not possible.
Radio Link Addition (Branch Addition)
If this is the first radio link
via the DRNC for this UE,a new Iur signallingconnection is established.
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 345/359
1. SRNC requests DRNC for radio resources by sending RNSAP messageRadio Link Setup Request.
Parameters: Cell id, Transport Format Set per DCH, Transport Format Combination Set,
frequency, UL scrambling code
SRNC decides to setup a radio link RL via a new cell controlled by another RNC.
This Iur signallingconnection will be usedfor all RNSAP signalling
related to this UE.
2. If requested resources are available, DRNC sends NBAP message Radio Link
Setup Request to Node B.
Parameters: Cell id, Transport Format Set per DCH, Transport Format Combination
Set, frequency, UL scrambling code.
Then Node B start the uplink UL reception. 3. Node B allocates requested resources. Successful outcome is reportedin NBAP message Radio Link Setup Response.Parameters: Signalling link termination, Transport layer addressing information (AAL2
address, AAL2 Binding Identitie(s)) for Data Transport Bearer(s).
4. DRNC sends RNSAP message Radio Link Setup Response to SRNC.Parameters: Transport layer addressing information (AAL2 address, AAL2 BindingIdentity) for Data Transport Bearer(s), Neighbouring cell information.
Radio Link Addition (Branch Addition)
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 346/359
5. SRNC initiates setup of Iur/Iub Data Transport Bearer using access link controlapplication part ALCAP protocol.
This request contains the AAL2 Binding Identity to bind the Iub DataTransport Bearer to DCH. This may be repeated for each Iur/Iub Data Transport Bearer to be setup.
6./7. Node B and SRNC establish synchronism for the Data TransportBearer(s) by means of exchange of the appropriate DCH Frame Protocol
frames Downlink Synchronisation and Uplink Synchronisation, relativeto already existing radio link(s). Then Node B starts DL transmission.
8. SRNC sends RRC message Active Set Update (Radio Link Addition)to UE on dedicated control channel DCCH.
Parameters: Update type, Cell id, DL scrambling code, Power control information,Ncell information.9. UE acknowledges with RRC message Active Set Update Complete.
Soft Handover „continued‟
▪ Radio link Deletion (Branch Deletion)
▫ This example shows deletion of a radio link belonging to a
Node B controlled by a RNC other than the serving RNC.
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 347/359
Radio link Deletion (Branch Deletion)
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 348/359
1. SRNC sends RRC message Active Set Update (Radio Link Deletion) toUE on DCCH.Parameters: Update type, Cell id.
2. UE deactivates DL reception via old branch, and acknowledges with
RRC message Active Set Update Complete. 3. SRNC requests DRNC to deallocate radio resources by sendingRNSAP message Radio Link Deletion Request.Parameters: Cell id, Transport layer addressing information.
4. DRNC sends NBAP message Radio Link Deletion Request toNode B.Parameters: Cell id, Transport layer addressing information.
SRNC decides to remove a radio link via an old cell controlled by another RNC.
Radio link Deletion (Branch Deletion)
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 349/359
5. Successful outcome is reported in NBAP message Radio Link Deletion
Response. 6. DRNC sends RNSAP message Radio Link Deletion Response to SRNC. 7. SRNC initiates release of Iur/Iub Data Transport Bearer using ALCAP protocol.Node B deallocates radio resources.
Hard Handover
▪ This example shows Inter-RNS Hard Handover
▪ A switch in the core network CN, is in a situation in which the UE is connected to two CNnodes simultaneously.
▪ The CN will end up using one Node B directly under the target RNC after the hardhandover.
▪ The Serving RNC makes the decision to perform the Hard Handover via CN.
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 350/359
▪ The SRNC also decides into which RNC (Target RNC) the Serving RNC functionality isto be relocated.
Hard Handover with switching in the CN
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 351/359
1./2. SRNC sends Relocation Required messages to both CN nodes.
Parameters: target RNC identifier, Information field transparent to the CN
node and to be transmitted to the target RNC.
Upon reception of Relocation Required message, the CN elementprepares itself for the switch and may also suspend data traffic between
UE and itself for some bearers
..
3./4. When CN is aware of preparation , CN node conveys aRelocation Request message to the target RNC to allocate resources.
Parameters: bearer ID's requested to be re-routed towards the CN node, fromwhich the Relocation Request originated.
CN indicates in the message whether it prefers point to multipoint typeof connections within CN or hard switch in CN. In this example the
latter is assumed.
Target RNC allocates necessary resources within the UTRAN to supportthe radio links to be used after completion of the Hard Handoverprocedure.
5. Target RNC and CN node establish the new Iu transport bearers for
each Radio Access Bearer related to the CN node.
Hard Handover with switching in the CN
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 352/359
6./7./8. The target RNC allocates RNTI and radio resources for the RRCconnection and the Radio Link, then sends the NBAP message Radio LinkSetup Request to the target Node-B.
Parameters: Cell id, Transport Format Set, Transport Format Combination Set,frequency, UL scrambling code (FDD only), Time Slots (TDD only), User Codes (TDDonly), Power control information etc.
Node B allocates resources, starts PHY reception, and responds withNBAP message Radio Link Setup Response. Target RNC initiates set-upof Iub Data Transport bearer using ALCAP protocol. This request containsthe AAL2 Binding Identity to bind the Iub Data Transport Bearer to the DCH.
Hard Handover with switching in the CN
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 353/359
9./10. When RNC has completed preparation phase, RelocationRequest Acknowledge is sent to the CN elements.
Parameters: transparent field to the CN that is to be transmitted to the Source RNS.
11./12. When CN is ready for the change of SRNC, CN node sends aRelocation Command to the RNC. Message contains the transparentfield provided by Target RNC.
Parameters: information provided in the Information field from the target RNC.
Hard Handover with switching in the CN
13 S RNC d RRC Ph sical Channel
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 354/359
13. Source RNC sends a RRC message Physical Channel
Reconfiguration to the UE.
NOTE 1: The messages used here are only one example of the various
messages which can be used to trigger a handover, to confirm it or to
indicate the handover failure.
The different possibilities are specified in the RRC specification (25.331)
Hard Handover with switching in the CN
14 /15 When target RNC has detected the UE Relocation Detect
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 355/359
•14./15. When target RNC has detected the UE, Relocation Detectmessage is sent to the CN nodes. The Target RNC switches theconnection towards the new Iu, when UE is detected.
•After the switch, UL traffic from Node-B's is routed via the newlyestablished Mobile data channel to the new MAC/RLC entities andfinally to the correct Iu transport bearer.
•DL data arriving from the new Iu link is routed to newly establishedradio link control RLC entities, to the medium access control MAC
and to the Nodes B.
Hard Handover with switching in the CN
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 356/359
16. When the UE switch from the old RL to the new RL, the source NodeB detect a failure on its RL and send a NBAP message Radio LinkFailure Indication to the source RNC.
17. When the RRC connection is established with the target RNC and
necessary radio resources have been allocated the UE sends RRCmessage Physical Channel Reconfiguration Complete to the targetRNC.
NOTE 1: The messages used here are only one example of the various
messages which can be used to trigger a handover, to confirm it or to
indicate the handover failure. The different possibilities are specified in the RRC specification (25.331)
Hard Handover with switching in the CN
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 357/359
18./19 After a successful switch and resource allocation at targetRNC, RNC sends Relocation Complete messages to the involved CNnodes.
Note: At any phase, before the Relocation Complete message issent, the old communication link between the CN and UE is all thetime existing and working and the procedure execution can bestopped and original configuration easily restored.
20./21. The CN node initiates the release of the Iu connections to thesource RNC by sending RANAP message Iu Release Command.
Hard Handover with switching in the CN
UTRAN Signalling
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 358/359
22. Upon reception of the release requests from the CN nodes the
old SRNC executes all necessary procedures to release all visible
UTRAN resources that were related to the RRC connection inquestion.
23./24. SRNC confirm the IU release to the CN nodes sending
the message Iu Release Complete.
7/29/2019 UMTS System
http://slidepdf.com/reader/full/umts-system 359/359