0% found this document useful (0 votes)
17 views107 pages

2 - OWP112020 WCDMA Radio Network Capacity Dimensioning ISSUE1.24 WOM Colombia

The document discusses WCDMA radio network capacity dimensioning, highlighting the relationship between capacity, coverage, and interference factors. It outlines objectives for understanding 3G traffic models, estimating multi-service capacity, and analyzing uplink and downlink interference. Key components include traffic models for circuit-switched and packet-switched services, as well as methods for capacity and bandwidth dimensioning.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
17 views107 pages

2 - OWP112020 WCDMA Radio Network Capacity Dimensioning ISSUE1.24 WOM Colombia

The document discusses WCDMA radio network capacity dimensioning, highlighting the relationship between capacity, coverage, and interference factors. It outlines objectives for understanding 3G traffic models, estimating multi-service capacity, and analyzing uplink and downlink interference. Key components include traffic models for circuit-switched and packet-switched services, as well as methods for capacity and bandwidth dimensioning.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 107

WCDMA Radio

Network Capacity
Dimensioning
www.huawei.com

Huawei Technologies Co., Ltd. All rights reserved.


Foreword
 WCDMA is a self-interference system
 WCDMA system capacity is closely related to coverage
 WCDMA network capacity has the “soft capacity” feature
 The WCDMA network capacity restriction factors in the radio
network part include the following:
 Uplink interference
 Downlink power
 Downlink channel code resources (OVSF)
 Channel element (CE)
 IUB Bandwidth

Huawei Technologies Co., Ltd. All rights reserved. Page1


Objectives
 Upon completion of this course, you will be able to:

 Grasp the parameters of 3G traffic model

 Understand factors that restrict WCDMA network capacity

 Understand methods and procedures of estimating multi-


service capacity

Huawei Technologies Co., Ltd. All rights reserved. Page2


Contents

1. Traffic Model

2. Interference Analysis

3. Capacity Dimensioning

4. CE Dimensioning

5. IUB Bandwidth Dimensioning

6. Network Dimensioning Flow

Huawei Technologies Co., Ltd. All rights reserved. Page3


Contents

1. Traffic Model

1.1 Overview of traffic model

1.2 CS traffic model

1.3 PS traffic model

Huawei Technologies Co., Ltd. All rights reserved. Page4


QoS Type
It is necessary to maintain the time relationship Voice service,
Conversation between the information entities in the stream. videophone
Real-time category

al Small time delay tolerance, requiring data rate


symmetry

Typically unidirectional services, high Streaming


Streaming requirements on error tolerance, high multimedia
requirements on data rate
Non real-time category

Request-response mode, data integrity must be Web page


Interactive maintained. High requirements on error tolerance, browse,
low requirements on time delay tolerance network game

Data integrity should be maintained. Small delay Background


Background restriction, requiring correct transmission download of
Email

Huawei Technologies Co., Ltd. All rights reserved. Page5


Traffic Model

Service Pattern

Traffic Model
Results

User Behaviour

System Configuration

Huawei Technologies Co., Ltd. All rights reserved. Page6


The Contents of Traffic Model
 Service pattern refers to the service features
 User type (indoor ,outdoor, vehicle)

 User’s average moving speed

 Service Type

 Uplink and downlink service rates

 Spreading factor

 Time delay requirements of the service

 User behaviour refers to the conduct of people in using the


service

Huawei Technologies Co., Ltd. All rights reserved. Page7


Contents

1. Traffic Model

1.1 Overview of traffic model

1.2 CS traffic model

1.3 PS traffic model

Huawei Technologies Co., Ltd. All rights reserved. Page8


CS Traffic Model

 Voice service is a typical CS services. Voice data arrival conforms


to the Poisson distribution. Its time interval conforms to the
exponent distribution

 Key parameters of the model


 Penetration rate
 BHCA: busy-hour call attempts
 Mean call duration (s)
 Activity factor
 Mean rate of service (kbps)

Huawei Technologies Co., Ltd. All rights reserved. Page9


CS Traffic Model Parameters

 Mean busy-hour traffic (Erlang) per user = BHCA × mean call


duration /3600

 Mean busy hour traffic volume per user (kbit) = BHCA × mean call
duration × activity factor × mean rate

 Mean busy hour throughput per user (bps) = mean busy hour
traffic volume per user × 1000/3600

Huawei Technologies Co., Ltd. All rights reserved. Page10


Contents

1. Traffic Model

1.1 Overview of traffic model

1.2 CS traffic model

1.3 PS traffic model

Huawei Technologies Co., Ltd. All rights reserved. Page11


PS Traffic Model
Session

Packet Call Packet Call


Downloading Downloading

Active Dormant Dormant Active

Packet Call

Data Burst Data Burst Data Burst

Huawei Technologies Co., Ltd. All rights reserved. Page12


PS Traffic Model Parameters

Packet Call Num/Session

Packet Num/Packet Call


Traffic Model Packet Size (bytes)

Reading Time (sec)

Typical Bear Rate (kbps)

BLER

Huawei Technologies Co., Ltd. All rights reserved. Page13


Parameter Determining
 The basic parameters in the traffic model are determined in
the following ways:

 Obtain numerous basic parameter sample data from the


existing network
 Obtain the probability distribution of the parameters through
processing of the sample data
 Take the distribution most proximate to the standard probability
as the corresponding parameter distribution through
comparison with the standard distribution function

Huawei Technologies Co., Ltd. All rights reserved. Page14


PS User Behaviour Parameters

Penetration Rate

User Behaviour BHSA

User Distribution
(High, Medium, Low end)

Huawei Technologies Co., Ltd. All rights reserved. Page15


PS User Behaviour Parameters

 Penetration Rate

 BHSA
 The times of single-user busy hour sessions of this service

 User Distribution (High, Medium, Low end)


 The users are divided into high-end, mid-end and low-end
users.

Huawei Technologies Co., Ltd. All rights reserved. Page16


PS Traffic Model Parameters

 Data Transmission time (s): The time in a single session of


service for purpose of transmitting data.

SessionTrafficVolume × 8 / 1000 1
DataTransmissionTime = ×
1 − BLER TypicalRate
 Holding Time (s): Average duration of a single session of service
PacketCallNum
HoldingTime = ( − 1) × Re adingTime + DataTransmissionTime
Session
 Activity Factor:
DataTransmissionTime
ActivityFactor =
HoldingTime

Huawei Technologies Co., Ltd. All rights reserved. Page17


PS Traffic Model Parameters

 Mean busy-hour traffic (Erlang) per user


MeanBusyHourTrafficPerUser(Erlang) = UserNum× HoldingTime
TransmissionTime
= 1×
ActivityFactor

TrafficVolumePerUser × (1 + RetransmissionRate)
=1 ×
TypicalRate × ActivityFactor × 3600

 Question:What other factors do we need to consider?

Huawei Technologies Co., Ltd. All rights reserved. Page18


Contents

1. Traffic Model

2. Interference Analysis

3. Capacity Dimensioning

4. CE Dimensioning

5. IUB Bandwidth Dimensioning

6. Network Dimensioning Flow

Huawei Technologies Co., Ltd. All rights reserved. Page19


Basic Principles

 In the WCDMA system, all the cells use the same frequency,
which is conducive to improve the WCDMA system capacity.
However, for reason of co-frequency multiplexing, the
system incurs interference between users. This multi-
access interference restricts the capacity in turn.

 The radio system capacity is decided by uplink and


downlink. When planning the capacity, we must analyze
from both uplink and downlink perspectives.

Huawei Technologies Co., Ltd. All rights reserved. Page20


Contents

2. Interference Analysis

2.1 Interference Analysis Overview

2.2 Uplink Interference Analysis

2.3 Downlink Interference Analysis

Huawei Technologies Co., Ltd. All rights reserved. Page21


Interference Analysis Overview

 Why do we analyze interference in network dimensioning?


 No matter uplink or downlink dimensioning, the Eb/No
requirement should be met:
 Eb/No = Ec/No × PG

 Eb/No and PG is pre-defined, so we should calculate


expected Ec and No through interference analysis

 The interference increase which is load factor could be


predicted

 The load factor of each service rate is different

Huawei Technologies Co., Ltd. All rights reserved. Page22


Contents

2. Interference Analysis

2.1 Interference Analysis Overview

2.2 Uplink Interference Analysis

2.3 Downlink Interference Analysis

Huawei Technologies Co., Ltd. All rights reserved. Page23


Uplink Interference Analysis

 Uplink interference analysis is based on the following


formula:

I TOT = I own + I other + PN

Huawei Technologies Co., Ltd. All rights reserved. Page24


Uplink Interference Analysis

 Receiver noise floor: PN

PN = 10 log( K * T *W ) + NF

 For Huawei NodeB, the typical value is -106.4dBm/3.84MHZ

Huawei Technologies Co., Ltd. All rights reserved. Page25


Uplink Interference Analysis

 I own: Interference from users of this cell


 Interference that every user must overcome is : I total − Pj

 Pj is the receiving power of the user j , ρ j is UL activity factor


( Eb / No ) Avg _ j
Under the ideal power control : 10 Pj W 1
 10
= ⋅ ⋅
I TOT − Pj R j ρ j
 Hence: I TOT
Pj =
1 W 1
1+ ( Eb / No ) Avg _ j ⋅ ⋅
Rj ρ j
10 10

 The interference from users of this cell is the sum of power of


all the users arriving at the receiver:
N
I own = ∑ Pj
1

Huawei Technologies Co., Ltd. All rights reserved. Page26


Uplink Interference Analysis

 I other :Interference from users of adjacent cell

 The interference from users of adjacent cell is difficult to


analyze theoretically, because it is related to user distribution,
cell layout, and antenna direction diagram.

 Adjacent cell interference factor :

I other
f =
I own
Huawei Technologies Co., Ltd. All rights reserved. Page27
Uplink Interference Analysis
N
I TOT
I TOT = I own + I other + PN = (1 + f )∑ + PN
1 W 1
1
1+ ( Eb / No ) Avg _ j ⋅ ⋅
Rj ρ j
10 10

1
Define: Lj =
1 W 1
1+ ( Eb / No ) Avg _ j ⋅ ⋅
Rj ρ j
10 10

N
Then: I TOT = I TOT ⋅ (1 + f ) ⋅ ∑ L j + PN
1

1
Obtain: I TOT = PN ⋅ N
1 − (1 + f ) ⋅ ∑ L j
1
Huawei Technologies Co., Ltd. All rights reserved. Page28
Uplink Interference Analysis
 Suppose that:
 All the users are 12.2 kbps voice users, Eb/NoAvg = 5dB

 Voice activity factor ρ j = 0.67

 Adjacent cell interference factor f=0.55

Huawei Technologies Co., Ltd. All rights reserved. Page29


Uplink Interference Analysis
 According to the above mentioned relationship, the noise will rise:

I TOT 1 1
NoiseRise = = N
=
PN 1 − ηUL
1 − (1 + f )∑ L j
1

Huawei Technologies Co., Ltd. All rights reserved. Page30


Uplink Interference Analysis
 Define the uplink load factor for one user:

1
η j = (1 + f ) × L j = (1 + f ) ×
1 W 1
1+ ( Eb / No )Avg _ j ⋅ ⋅
Rj ρ j
10 10

 Define the uplink load factor for the cell:

N N
1
ηUL = (1 + f ) × ∑ L j = (1 + f ) × ∑
1 W 1
1 1
1+ ( Eb / No )Avg _ j ⋅ ⋅
Rj ρ j
10 10

Huawei Technologies Co., Ltd. All rights reserved. Page31


Contents

2. Interference Analysis

2.1 Interference Analysis Overview

2.2 Uplink Interference Analysis

2.3 Downlink Interference Analysis

Huawei Technologies Co., Ltd. All rights reserved. Page32


Downlink Interference Analysis

 Downlink interference analysis is based on the following


formula:

I TOT = I own + I other + PN

Huawei Technologies Co., Ltd. All rights reserved. Page33


Downlink Interference Analysis

 Receiver noise floor: PN

PN = 10 log( K * T *W ) + NF

 For commercial UE, the typical value is -101dBm/3.84MHZ

Huawei Technologies Co., Ltd. All rights reserved. Page34


Downlink Interference Analysis

 I own :Interference from downlink signal of this cell


 The downlink users are identified with the mutually orthogonal
OVSF codes. In the static propagation conditions without multi-
path, no mutual interference exists.

 In case of multi-path propagation, certain energy will be


detected by the RAKE receiver, and become interference
signals. We define the non-orthogonal factor α to describe this
phenomenon:

( Iown ) j = α × PTX

Huawei Technologies Co., Ltd. All rights reserved. Page35


Downlink Interference Analysis
 I other : Interference from the downlink signal of adjacent cell

 The transmitting signal of the adjacent cell NodeB will cause


interference to the users in the current cell. Since the
scrambling codes of users are different, such interference is
non-orthogonal

 Hence we obtain:

( Iother ) j = f × PTX

Huawei Technologies Co., Ltd. All rights reserved. Page36


Downlink Interference Analysis

 Ec/Io for User j is:

Pj
Ec 10 CL / 10 Pj
( )j = =
Io (α + f ) × PTX PN / 10 (α + f ) × P + 10 ( CL + PN ) / 10
+ 10 TX
10CL /10

Huawei Technologies Co., Ltd. All rights reserved. Page37


Downlink Interference Analysis

 Under the ideal power control:

( Eb / No ) j
Ec W 1
10 10
= ( )j × ×
Io Rj ρ j

 Then we can get:

( Eb / No ) j
10(CL + PN ) /10
10 10
× ρ j × PTX × (α + f + )
PTX
Pj =
W / Rj

Huawei Technologies Co., Ltd. All rights reserved. Page38


Downlink Interference Analysis

 Define the downlink load factor for user j:

( Eb / No ) j
PTX 10(CL + PN ) /10
10 10
×ρj × × (α + f + )
Pj Pmax PTX
ηj = =
Pmax W / Rj

 Define the downlink load factor for the cell:

PTX
η DL =
Pmax

Huawei Technologies Co., Ltd. All rights reserved. Page39


Downlink Interference Analysis

 According to the above mentioned relationship, the noise will rise:

I total PN + I own + I other No + (α + f )× PMax ×η DL / CL


NoiseRise = = =
PN PN PN

Huawei Technologies Co., Ltd. All rights reserved. Page40


Contents

1. Traffic Model

2. Interference Analysis

3. Capacity Dimensioning

4. CE Dimensioning

5. IUB Bandwidth Dimensioning

6. Network Dimensioning Flow

Huawei Technologies Co., Ltd. All rights reserved. Page41


Capacity Dimensioning Flow
Dimensioning Start

Assumed
Subscribers
Load per Connection of
R99

CS Average Cell CS Peak Cell Load PS Average Cell


HSPA Cell Load
Load (ErlangB) Load

Total Cell Load

No
=Target Cell Load?

Yes
Dimensioning End

Load cell −total _ UL = max{ Load CS − peak , Load CS − avg + Load PS − avg + Load HSUPA }
Load cell −total _ DL = max{ Load CS − peak , Load CS − avg + Load PS − avg + Load HSDPA } + Load CCH

Huawei Technologies Co., Ltd. All rights reserved. Page42


Contents

3. Capacity Dimensioning

3.1 R99 Capacity Dimensioning

3.2 HSDPA Dimensioning

Huawei Technologies Co., Ltd. All rights reserved. Page43


Capacity Dimensioning Differences
GSM WCDMA

 Hard blocking  Soft blocking


 Capacity --- hardware dependent  Capacity --- interference dependent
 Single service  Multi services (CS&PS)
 Single GoS requirement  Respective quality requirements of
each service

Huawei Technologies Co., Ltd. All rights reserved. Page44


CS Capacity Dimensioning (1)
 CS services
ErlangB Model
 Real time
GoS requirements

......

Channels

Capacity

?
Blocking probability ErlangB Cell Loading

Huawei Technologies Co., Ltd. All rights reserved. Page45


Best Effort for Packet Services
Total Load
 PS Services:

Load
CS Peak Load
 Best Effort
Load occupied by PS
 Retransmission

 Burst Traffic CS Average Load

Load occupied by CS

Time

 PS will use the spare load apart from that used by CS

Huawei Technologies Co., Ltd. All rights reserved. Page46


CS Capacity Dimensioning

 Average load:

AverageLoad j = Traffic j × LoadFactor j


N
AverageLoadTotal = ∑ AverageLoad j
1

 Peak load:
 Query the peak connection through ErlangB table

PeakLoad j = PeakConn j × LoadFactorj

PeakLoadTotal = ErlangB ( PeakLoad j )

Huawei Technologies Co., Ltd. All rights reserved. Page47


PS Capacity Dimensioning
 Average load:

AverageLoad j = Traffic j × (1 + Re txRate) × (1 + BurstRate) × LoadFactor j

N
AverageLoadTotal = ∑ AverageLoad j
1

 Peak load:
 None

 Why don’t we calculate PS peak load?

Huawei Technologies Co., Ltd. All rights reserved. Page48


Case Study (Import)

 Common parameters:
 Maximum cell transmission power: 20W
 Subscriber number per Cell: 800
 Overhead of SHO (including softer handover): 40%
 Retransmission of PS is 5%
 R99 PS traffic burst: 20%
 Activity factor of PS is 0.9
 Power allocation for CCH is 20% in downlink

Huawei Technologies Co., Ltd. All rights reserved. Page49


Case Study (Import)

 Traffic Model, GoS and load factors:

UL DL GoS Load Factors (UL) Load Factors (DL)

AMR12.2k (Erl) 0.02 0.02 2% 1.18% 0.83%

CS64k (Erl) 0.001 0.001 2% 4.99% 4.65%

PS64k (Kbit) 50 100 N/A 4.21% 2.96%

PS128k (Kbit) 0 100 N/A 5.94%

PS384 (Kbit) 0 0 N/A

Huawei Technologies Co., Ltd. All rights reserved. Page50


Case Study (Export)
 Uplink Average Traffic  Downlink Average Traffic

CSTrafficPerCell(Erlang) = NumOfUser× TrafficPerUser

AMR12.2k: 0.02×800=16 AMR12.2k: 0.02×800=16

CS64k: 0.001×800=0.8 CS64k: 0.001×800=0.8

TrafficVolumePerUser× (1 + ReTransmissionRate)
PSTrafficPerCell(Erlang) = NumOfUser×
TypicalRate × ActivityFactor × 3600

50 × (1 + 5%) 100 × (1 + 5%)


PS64k: 800 × = 0.20 PS64k: 800 × = 0.41
64 × 0.9 × 3600 64 × 0.9 × 3600

100 × (1 + 5%)
PS128: 800 × = 0.20
128 × 0.9 × 3600

Huawei Technologies Co., Ltd. All rights reserved. Page51


Case Study (Export)
 Uplink Average Load  Downlink Average Load

CSUplinkAverageLoad CSDownlinkAverageLoad
= TrafficPerCell × LoadFactor = TrafficPerCell × LoadFactor × (1 + SoftHoFactor)
AMR12.2k: AMR12.2k:
16×1.18%=18.88% 16×0.83%×(1+40%)=18.59%
CS64k: CS64k:
0.8×4.99%=3.99% 8×4.65% ×(1+40%)=5.21%
PSUplinkAverageLoad = TrafficPerCell PSDownlinkAverageLoad = TrafficPerCell
× LoadFactor × (1 + BurstRate) × LoadFactor × (1 + BurstRate)× (1 + SoftHoFactor)
PS64k: 0.2×4.21%×(1+20%) PS64k: 0.41×2.96%×(1+20%)×(1+40%)
=1.01% =2.04%
PS128: 0.20×5.94%×(1+20%)×(1+40%)
=2.00%

Huawei Technologies Co., Ltd. All rights reserved. Page52


Case Study (Export)
 Uplink Peak Load  Downlink Peak Load

CSUplinkPe akConn CSDownlinkPeakConn


= ErlangB (ULAverageTraffic, Gos ) = ErlangB ( DLAverageTraffic × (1 + SoftHoFactor ), Gos )
AMR12.2k: AMR12.2k:
ErlangB(16,2%)=24 ErlangB(16×(1+40%),2%)=31
CS64k: CS64k:
ErlangB(0.8,2%)=4 ErlangB(0.8×(1+40%),2%)=5
CSPeakLoad= PeakConn× LoadFactor
AMR12.2k: AMR12.2k:
=24×1.18%=28.32% =31×0.83%=25.73%
CS64k: CS64k:
=4 ×4.99%=19.94% =5×4.65%=23.25%

Huawei Technologies Co., Ltd. All rights reserved. Page53


Case Study (Export)
 Uplink Cell Load

CS Peak Load= 28.32%+19.96%=48.28%

CS Average Load=18.88%+3.99%=22.87%

PS Average Load=1.01%

 Downlink Cell Load

CS Peak Load= 25.73%+23.25%=48.98%

CS Average Load= 18.59%+5.21%=23.8%

PS Average Load= 2.04%+2.00%=4.04%

Huawei Technologies Co., Ltd. All rights reserved. Page54


Contents

3. Capacity Dimensioning

3.1 R99 Capacity Dimensioning

3.2 HSDPA Dimensioning

Huawei Technologies Co., Ltd. All rights reserved. Page55


HSDPA Capacity Dimensioning (1)
 HSDPA Capacity Dimensioning
 The purpose is to obtain the required HSDPA power to satisfy
the cell average throughput.

 HS-DSCH will use the spare power apart from that of R99

Power 3GPP Release 99 Power 3GPP Release 5

Unused power
Pmax-R99 HS-DSCH

Dedicated channels (power controlled) Dedicated channels (power controlled)

Common channels Common channels

Power usage with dedicated t t


HS-DSCH with dynamic power allocation
channels channels

Huawei Technologies Co., Ltd. All rights reserved. Page56


HSDPA Capacity Dimensioning (2)
4.00% DU Cell coverage Radius=300m

Capacity Based on Simulation

Distribution probability
3.50%

3.00%
2.50%

 to simulate Ior/Ioc distribution in the 2.00%


1.50%

network with certain cell range 1.00%


0.50%
0.00%

to simulate cell throughput distribution

4.22
2.98
2.04
1.39
0.96
0.66
0.45
0.31
0.21
0.14
0.1
0.07
0.05
0.03
0.02
0.01
0.01
0.01
0
0
0
0

Ioc/Ior

based on Ec/Io distribution in the cell


Conditions of Simulation
 Dimensioning Procedure Channel model-TU3
5 codes

Cell coverage
radius
Simulation

HSDPA Power
Ior/Ioc distribution
Allocation

Cell average
Ec/Io distribution Ec/Io =>throughput throughput

Huawei Technologies Co., Ltd. All rights reserved. Page57


Case Study
 Input parameters
 Subscriber number per cell: 800
 HSDPA Traffic model: 1200kbit per subs
 HSDPA Retransmission rate: 10%
 HSDPA Data burst rate:20%
 The power for HS-SCCH: 5%
 Cell radius: 1km

 HSDPA cell average throughput:

800 *1200
× (1 + 10%) × (1 + 20%) = 352kbps
3600
 The needed power for HS-DSCH including that for HS-SCCH is 18.38%

Huawei Technologies Co., Ltd. All rights reserved. Page58


Case Study (Export)

 Uplink Cell Load

Cell Load =Max(CS Peak Load , CS Average Load + PS Average Load)

=Max((28.32%+19.96%) ,(18.88%+3.99%+1.01%))

=48.28%
 Downlink Cell Load

Cell Load =Max(CS Peak Load , CS Average Load + PS Average Load+HSPA Load)

+CCH Load

=Max((25.73%+23.25%) ,(18.59%+5.21%+2.04%+3.99%+18.38%))+20%

=68.98%

UL Cell Load <UL Target Load & DL Cell Load <DL Target Load

One Carrier is enough!

Huawei Technologies Co., Ltd. All rights reserved. Page59


Contents

1. Traffic Model

2. Interference Analysis

3. Capacity Dimensioning

4. CE Dimensioning

5. IUB Bandwidth Dimensioning

6. Network Dimensioning Flow

Huawei Technologies Co., Ltd. All rights reserved. Page60


Overview

 Definition of a CE:
 A Channel Element is the base band resource required in the Node-B
to provide capacity for one voice channel, including control plane
signaling, compressed mode, transmit diversity and softer handover.

 NodeB Channel Element Capacity


 One BBU3900
 UL 1,536 CEs with full configuration

 DL 1,536 CEs with full configuration

Huawei Technologies Co., Ltd. All rights reserved. Page61


Huawei Channel Elements Features
 Channel Elements pooled in one NodeB
 No need extra R99 CE resource for CCH
 reserved CE resource for CCH

 No need extra CE resource for TX diversity


 No need extra CE resource for Compressed Mode
 reserved resources for Compressed Mode

 No need extra CE resource for Softer HO


 HSDPA does not occupy R99 CE resource
 separate module for HSDPA

 HSUPA shares CE resource with R99 services


 No additional CE resource for AGCH RGCH and HICH

Huawei Technologies Co., Ltd. All rights reserved. Page62


CE Dimensioning Flow
Dimensioning Start

--Subscribers per NodeB


--Traffic model

CS Average CS Peak CE PS Average


HSPA CE
CE (ErlangB) CE

Channel Elements per


NodeB

Dimensioning End

CEUL _ Total = Max(CE CS _ Peak _ UL , CE CS _ Average _ UL + CE PS _ UL + CE A _ UL + CE HSUPA )


CE DL _ Total = Max(CE CS _ Peak _ DL , CE CS _ Average _ DL + CE PS _ DL + CE A _ DL )

Huawei Technologies Co., Ltd. All rights reserved. Page63


CE Mappings for R99 Bearers

Channel Elements Mapping for R99 Bearers

Bearer Uplink Downlink

AMR12.2k 1 1

CS64k 3 2

PS64k 3 2

PS128k 5 4

PS144k 5 4

PS384k 10 8

Huawei Technologies Co., Ltd. All rights reserved. Page64


R99 CE Dimensioning Principle
 Peak CE occupied by CS can be obtained through ErlangB algorithm

 Average CE needed by CS and PS depend on the traffic of each service,


i.e.

 Average CE = Traffic * CE Factor CE resource shared


among each service
Total CE
CE

CS Peak CE ErlangB Model


CE occupied by PS AM
R1
and HSPA 2.2
k

......
CS Average CE CE
Resources
CE occupied by CS

6 4k
CS
Time

Huawei Technologies Co., Ltd. All rights reserved. Page65


HSDPA CE Dimensioning
 In uplink, no CE consumption for HS-DPCCH if corresponding UL
DCH channel exists

 In uplink, CE consumed by one A-DCH depends on its bearing


rate

 In downlink, A-DCH is treated as R99 DCH.

 No additional CE needed for HS-DSCH and HS-SCCH

Associated Dedicated Channels

One HSDPA link need


one A-DCH in uplink and
downlink respectively
Site 1 Site 2

Huawei Technologies Co., Ltd. All rights reserved. Page66


CE Mappings for HSDPA Bearers

HSDPA Channel Elements Consumption


Traffic Uplink Downlink
HSDPA Traffic --- 0 CE
HS-DPCCH 0 CE ---
UL A-DCH 3 CE ---
DL A-DCH --- 1 CE

Huawei Technologies Co., Ltd. All rights reserved. Page67


Case Study (Import)
 Common Parameters Traffic Model UL DL GoS
AMR12.2k (Erl) 0.02 0.02 2%
 Subscribers number per NodeB: 2000 CS64k (Erl) 0.001 0.001 2%
PS64k (kbit) 50 100 N/A
 Overhead of SHO: 30%
PS128k (kbit) 0 80 N/A
 R99 PS traffic burst: 20% HSPA (kbit) 0 1200 N/A

 Retransmission rate of R99 PS: 5%


 PS Channel element utilization rate: 0.7
 Average throughput requirement per user of HSDPA: 400kbps
 HSDPA traffic burst is 25%
 Retransmission rate of HSDPA is 10%

Huawei Technologies Co., Ltd. All rights reserved. Page68


Case Study (Export)
 Uplink Average Traffic  Downlink Average Traffic

CSTrafficPerNodeB(Erlang) = NumOfUser× TrafficPerUser

AMR12.2k: 0.02×2000=40 AMR12.2k: 0.02×2000=40

CS64k: 0.001×2000=2 CS64k: 0.001×2000=2

TrafficVolumePerUser× (1 + ReTransmissionRate)
PSTrafficPerNodeB(Erlang) = NumOfUser×
TypicalRate × CEUtilizationRate × 3600

50 × (1 + 5%)
PS64k:2000 × = 0.65 PS64k:2000 × 100 × (1 + 5%) = 1.30
64 × 0.7 × 3600 64 × 0.7 × 3600

80 × (1 + 5%)
PS128:2000 × = 0.52
128 × 0.7 × 3600

Huawei Technologies Co., Ltd. All rights reserved. Page69


Case Study (Export)
 Uplink Average CE  Downlink Average CE

E = Roundup (TrafficPerNodeB× (1 + SoftHoFactorwithoutSof ter ))× CEConsumption


CSAverageC

AMR12.2k: AMR12.2k:
Roundup(40×(1+30%) )×1=52 Roundup(40×(1+30%) )×1=52
CS64k: CS64k:
Roundup(2×(1+30%) )×3=9 Roundup(2×(1+30%) )×2=6

E = Roundup (TrafficPerNodeB× (1 + BurstRate)× (1 + SoftHoFactorwithoutSof ter ))


PSAverageC
× CEConsumpt
ion

PS64k: Roundup(0.65×(1+20%) PS64k: 6


×(1+30%) )×3 =6 PS128k: 4

Huawei Technologies Co., Ltd. All rights reserved. Page70


Case Study (Export)
 DL HSDPA CE

NumOfHSPAU ser = RoundDown( NumOfUserPerNodeB


TrafficVolumePerUser× (1 + HSPAReTran smissionRate) × (1 + HSPABurstR
ate)
× )
TargetRate× 3600

1200 × ( 1 + 10%) × ( 1 + 25%)


= RoundDown( 2000 × )
400 × 3600
=2

= NumOfHSPAUser × CEConsumption A- DCH


DLHSDPACE
= 2 ×1 = 2

Huawei Technologies Co., Ltd. All rights reserved. Page71


Case Study (Export)
 Uplink Peak CE  Downlink Peak CE

CSPeakConn= ErlangB ( AverageTraffic × (1 + SoftHoFactorwithoutSofter ), Gos )

AMR12.2k: AMR12.2k:
ErlangB(40×(1+30%),2%) =63 ErlangB(40×(1+30%),2%) =63

CS64k: CS64k:
ErlangB(2 ×(1+30%),2%)= 7 ErlangB(2 ×(1+30%),2%)= 7
CSPeakCE= PeakConn× CEConsumption
AMR12.2k: AMR12.2k:
=63×1=63 =63×1=63
CS64k: CS64k:
=7×3=21 =7×2=14

Huawei Technologies Co., Ltd. All rights reserved. Page72


Case Study (Export)
 Uplink NodeB CE  Downlink NodeB CE

CS Peak CE= 63+21 CS Peak CE= 63+14

CS Average CE=52+9 CS Average CE= 52+6

PS Average CE=6 PS Average CE= 6+4

HSDPA CE=2

TotalCE

=Max(CS Peak CE , CS Average CE+PS Average CE+HSDPA CE)

UL CE=84 DL CE=77

Question : which WBBP board should be selected if NodeB


configuration is S2/2/2?

Huawei Technologies Co., Ltd. All rights reserved. Page73


Contents

1. Traffic Model

2. Interference Analysis

3. Capacity Dimensioning

4. CE Dimensioning

5. IUB Bandwidth Dimensioning

6. Network Dimensioning Flow

Huawei Technologies Co., Ltd. All rights reserved. Page74


IUB Transport Overview

 ATM over E1/T1 E1/T1


TDM network
E1/T1

Node B RNC

 IP over E1/T1 E1/T1


TDM network
E1/T1

Node B RNC

 IP over Ethernet FE
IP network
FE

Node B RNC

Huawei Technologies Co., Ltd. All rights reserved. Page75


IUB Protocol Stack

ATM IP over E1/T1 IP over Ethernet


Control Plane User Plane Control Plane User Plane Control Plane User Plane
Radio Network Transport Network Radio Network Radio Network

DCH-FP
RACH-FP
FACH-FP
PCH-FP
HSDSCH-FP
EDCH-FP
Radio Network Layer
DCH-FP
RACH-FP
FACH-FP
PCH-FP
HSDSCH-FP
EDCH-FP

DCH-FP
RACH-FP
FACH-FP
PCH-FP
HSDSCH-FP
EDCH-FP
NBAP NBAP NBAP

FP-MUX

Transport Layer
ALCAP
Q.2630.2

Q.2150.2
SSCF-UNI SSCF-UNI

SSCOP SSCOP SCTP UDP SCTP UDP


AAL5 AAL5 AAL2 IP (IPHC) IP

ATM PPP (MUX+Compression) MAC

L1(PHY) L1(PHY) L1(PHY)

Huawei Technologies Co., Ltd. All rights reserved. Page76


IUB Bandwidth Composing
 IUB Bandwidth Composing
 Radio Network layer User Plane Data
 DCH User Data Bandwidth

– CS Voice Traffic Bandwidth

– CS VP Traffic Bandwidth

– R99 PS Traffic Bandwidth

– SRB Signaling Bandwidth

 HSPA Service Traffic Bandwidth

 Common Transport Channel Data Bandwidth

– RACH / FACH /PCH

 FP Control Frame

Huawei Technologies Co., Ltd. All rights reserved. Page77


IUB Bandwidth Composing(Cont.)

 IUB Bandwidth Composing


 Radio Network Layer Control Plane Data
 NBAP Common Procedures

 NBAP Dedicated Procedures

 Transport Network Layer Control Plane Data

 O&M Channel Bandwidth


 Either of UL and DL physical layer average bandwidth is 64Kbits/s

 Protocol Processing Overhead

Huawei Technologies Co., Ltd. All rights reserved. Page78


Bandwidth Dimensioning Flow
Dimensioning Procedure
input output

User Num / NodeB CS IUB Bandwidth

IUB Bandwidth
CS Traffic PS IUB Bandwidth
Voice Traffic
VP Traffic
Traffic
The Qos of CS Service HSPA IUB Bandwidth

PS Traffic
PS64 Throughput
Service Bandwidth
PS128 Throughput
PS384 Throughput
PS Retransmission
CCH Bandwidth

HSPA Traffic

Signaling Bandwidth

O&M Channel Bandwidth

Page79
Huawei Technologies Co., Ltd. All rights reserved.
Bandwidth Dimensioning Formula
Total Bandwidth

 CS and PS share IUB PS/HSPA Occupied Bandwidth


CS Peak
bandwidth Bandwidth

CS Peak Bandwidth-

Iub Bandwidth

ErlangB Algorithm CS Average


Bandwidth
 The Bandwidth for PS CS Occupied Bandwidth

and HSPA-BE Service


Signaling Bandwidth

CCH Bandwidth

O&M Bandwidth

Time
IubTotal = Max [ IubCS _ Peak , ( IubPS _ Avg + IubCS _ Avg + IubHSPA )]
+ IubSignalling + IubCCH + IubO & M

Huawei Technologies Co., Ltd. All rights reserved. Page80


Dimensioning Principle Case
 CS AMR Bandwidth Dimensioning

Huawei Technologies Co., Ltd. All rights reserved. Page81


Dimensioning Principle Summary
IUB Bandwidth IUB Overhead
R99
Service Type ATM IP over E1/T1 IP over Ethernet ATM IP over E1/T1 IP over Ethernet

AMR12.2k 22 kbps 20 kbps 20 kbps 80% 64% 64%


CS64k 88 kbps 70 kbps 71 kbps 38% 9% 11%
PS64k 92 kbps 74 kbps 75 kbps 44% 16% 17%
PS128k 180 kbps 140 kbps 144 kbps 41% 9% 13%
PS384k 540 kbps 415 kbps 418 kbps 41% 8% 9%

HSPA IUB Overhead ATM IP over E1/T1 IP over Ethernet


Uplink 27% 7% 7%
Downlink 35% 10% 10%

IP over
CCH IUB Overhead ATM IP over E1/T1
Ethernet
UL Bandwidth for 1 RACH / Cell 60 kbps 50 kbps 50 kbps
DL Bandwidth for 1 SCCPCH(FACH/PCH)
/ Cell
73 kbps 70 kbps 70 kbps

Huawei Technologies Co., Ltd. All rights reserved. Page82


CS Bandwidth Dimensioning

 CS IUB Peak Bandwidth Dimensioning

 Use ErlangB Algorithm to Estimate CS IUB Peak Bandwidth

Required Iub Bandwidth


Service Traffic GoS ErlangB Algorithm
Individual Total
AMR 12.2kbps 50 Erl 2% 1.342Mbps
2.838Mbps
CS 64kbps 10 Erl 2% 1.496Mbps

Huawei Technologies Co., Ltd. All rights reserved. Page83


CS Bandwidth Dimensioning(Cont.)
 CS IUB Average Bandwidth Dimensioning
 The below formula is used to estimate CS IUB Average
Bandwidth:

IubCS _ Average = ∑[Traffic _ Servicei * Iub _ BW _ Servicei * Activity _ Factori * (1 + SHO _ Factor )]
i

Voice Traffic Iub Bandwidth of + Soft HO factor


Voice Service

CS Average
Iub
VP Traffic Iub Bandwidth of + Soft HO factor Bandwidth
VP Service

Huawei Technologies Co., Ltd. All rights reserved. Page84


PS Bandwidth Dimensioning
 PS IUB Bandwidth Dimensioning

 PS IUB Bandwidth Dimensioning must consider the below


factors:

 When PS is BE Service, PS can share IUB Bandwidth with CS

 Retransmission for PS

 PS actual data rate is bursting, sometimes the service data rate is


high, sometimes the data rate is low

Huawei Technologies Co., Ltd. All rights reserved. Page85


PS Bandwidth Dimensioning(Cont.)
 PS IUB Bandwidth Dimensioning
 PS IUB Bandwidth Dimensioning formula:

[ Traffic _ Service _ i * Iub _ BW _ service _ i * ( 1 + Re transmission _ Ratio _ i )


IubPS _ Average = ∑* ( 1 + Burst _ Ratio _ i ) * ( 1 + SHO _ Factor ) ]
i

Traffic of PS IUB Bandwidth of + Retransmission + Burst + SHO


Service 1 PS Service 1 Ratio Ratio Factor

PS Average
IUB
. Bandwidth
.
.

PS Service i

Huawei Technologies Co., Ltd. All rights reserved. Page86


HSUPA Bandwidth Dimensioning
 HSUPA IUB Bandwidth Dimensioning
 Usually, HSPA is used to bear BE Service, so the
Dimensioning Algorithm is similar to PS:

IubHSUPA = (TrafficHSUPA / Sub) * ( Num _ Subs / NodeB) * (1 + HSUPA _ Overhead )


* (1 + Re transmissionHSUPA) * (1 + Burst _ RatioHSUPA) * (1 + SHO _ Ratio)

Traffic of + IUB
+ Retransmission + Burst + SHO HSUPA IUB
HSUPA Overhead
Ratio Ratio Factor Bandwidth

Huawei Technologies Co., Ltd. All rights reserved. Page87


HSDPA Bandwidth Dimensioning
 HSDPA IUB Bandwidth Dimensioning
 HSDPA can not support Soft Handover, so the HSDPA IUB
Bandwidth Dimensioning will not consider the SHO Factor:

IubHSDPA = (TrafficHSDPA / Sub) * ( Num _ Subs / NodeB) * (1 + HSDPA _ Overhead )


* (1 + Re transmissionHSDPA) * (1 + Burst _ RatioHSDPA)

Traffic of + Iub Overhead + Retransmission + Burst HSDPA Iub


HSDPA Ratio Ratio Bandwidth

Huawei Technologies Co., Ltd. All rights reserved. Page88


Relation Between PS/HSPA and CS
Bandwidth Dimensioning
 IUB User Plane Bandwidth Dimensioning

Huawei Technologies Co., Ltd. All rights reserved. Page89


Relation between PS/HSPA and CS
Bandwidth Dimensioning(Cont.)

 IUB User Plane Bandwidth Dimensioning

 Usually, PS/HSPA is BE Service, so these service can use the


rest IUB Bandwidth of CS

Iubtraffic = Max[ IubCS _ Peak , ( IubCS _ Avg + IubPS _ Avg + IubHSPA )]

Huawei Technologies Co., Ltd. All rights reserved. Page90


Signaling Bandwidth Dimensioning
 Signaling IUB Bandwidth Dimensioning

Bandwidth Dimensioning need to consider follow signalings:

 NBAP Signaling

 ALCAP Signaling(ATM Transport)

 FP Control Frame

 SRB(RRC Signaling)

Usually, we think Signaling Bandwidth is 10% of Traffic


Bandwidth

Huawei Technologies Co., Ltd. All rights reserved. Page91


CCH Bandwidth Dimensioning
 CCH IUB Bandwidth Dimensioning
 DL : FACH and PCH map to one SCCPCH, typical IUB
Bandwidth is 70 kbps (IP) / 74 kbps (ATM) per one SCCPCH

 UL : RACH, typical IUB Bandwidth is 50 kbps (IP) / 60 kbps


(ATM)

 Case : 1 NodeB (Configuration : S1/1/1), DL IUB Bandwidth


Dimensioning
 70 kbps * 3 Cells = 210 kbps (IP)

 74 kbps * 3 Cells = 222 kbps (ATM)

Huawei Technologies Co., Ltd. All rights reserved. Page92


O&M Bandwidth Dimensioning

 O&M IUB Bandwidth Dimensioning

 NodeB

 UL IUB Bandwidth: 64kbps

 DL IUB Bandwidth: 64kbps

Huawei Technologies Co., Ltd. All rights reserved. Page93


NodeB Bandwidth Dimensioning

 NodeB IUB Bandwidth Dimensioning

IubTotal = Max ( Iubuplink , Iubdownlink ) + IubO & M

IubUL = IubTraffic _ UL + IubCCH _ UL + IubSignalling _ UL IubDL = IubTraffic _ DL + IubCCH _ DL + IubSignalling _ DL

Huawei Technologies Co., Ltd. All rights reserved. Page94


IUB Bandwidth Dimensioning Case(Import)
 Input
 Iub interface: IP over E1T1

 NodeB Configuration : S1/1/1

 User Num of the NodeB : 2000

 SHO Factor : 30% (except Softer Handover)

 R99 PS Burst Ratio : 20%

 HSPA Burst Ratio : 25%

 R99 PS Retransmission Ratio : 5%

 HSPA Retransmission Ratio : 1%

 Voice Activity Factor : 0.5

Huawei Technologies Co., Ltd. All rights reserved. Page95


IUB Bandwidth Dimensioning Case (Import)

 Input

Traffic Model of Single


UL DL GoS Requirement
User for Busy Time

AMR12.2k 20 mErl 20 mErl 2%

CS64k 1 mErl 1 mErl 2%

PS64k 50 kbits 100 kbits N/A

PS128k 0 200 kbits N/A

HSPA 1000 kbits 5000 kbits N/A

Huawei Technologies Co., Ltd. All rights reserved. Page96


Case Study (Export)
 Uplink Average Traffic  Downlink Average Traffic

CSTrafficPerNodeB(Erlang) = NumOfUser× TrafficPerUser

AMR12.2k: 0.02×2000=40 AMR12.2k: 0.02×2000=40

CS64k: 0.001×2000=2 CS64k: 0.001×2000=2

TrafficVolumePerUser× (1 + ReTransmissionRate)
PSTrafficPerNodeB(Erlang) = NumOfUser×
TypicalRate × 3600

50 × (1 + 5%) 100 × (1 + 5%)


PS64k:2000 × 64 × 3600
= 0.46 PS64k: 2000 × = 0.91
64 × 3600

200 × (1 + 5%)
PS128:2000 × = 0.91
128 × 3600

Huawei Technologies Co., Ltd. All rights reserved. Page97


Case Study (Export)
 Uplink Average BW  Downlink Average BW

W(Kbit/s) = TrafficPerNodeB× IubBW × ActivityFactorVoicecall × (1 + SoftHoFactorwithoutSof ter )


CSAverageB
AMR12.2k: AMR12.2k:
40×20×0.5 × (1+30%)=520 40×20×0.5 × (1+30%)=520
CS64k: CS64k:
2×70 × (1+30%)= 182 2×70 × (1+30%)= 182

W(Kbit/s) = TrafficPerNodeB× IubBW × (1 + BurstRate)× (1 + SoftHoFactorwithoutSof ter )


PSAverageB

PS64k: 0.46×74×(1+20%) ×(1+30%) PS64k: 0.91×74×(1+20%) ×(1+30%)


=53.1 =105.05

PS64k: 0.91×140×(1+20%) ×(1+30%)

=198.74

Huawei Technologies Co., Ltd. All rights reserved. Page98


Case Study (Export)
 HSUPA Average BW

HSPAThroughputPerNodeB(Kbit/s)
TrafficVolumePerUser× (1 + ReTransmissionRate)
= NumOfUser×
3600
1000 × (1 + 1%)
= 2000 ×
3600
= 561.11

HSUPAAverageBW(Kbit/s)
= HSUPAThrou
ghputPerNodeB× (1 + Overhead)× (1 + BurstRate)× (1 + SoftHoFactorwithoutSof ter )
= 561.11 × (1 + 7%) × (1 + 25% ) × (1 + 30% )
= 975.63

Huawei Technologies Co., Ltd. All rights reserved. Page99


Case Study (Export)
 HSDPA Average BW

HSPAThroughputPerNodeB(Kbit/s)
TrafficVolumePerUser× (1 + ReTransmissionRate)
= NumOfUser×
3600
5000 × (1 + 1%)
= 2000 ×
3600
= 2805.56

HSDPAAvera geBW(Kbit/s)
= HSDPAThrou ghputPerNodeB× (1 + Overhead)× (1 + BurstRate)
= 2805.56 × (1 + 10%) × (1 + 25% )
= 3857.6

Huawei Technologies Co., Ltd. All rights reserved. Page100


Case Study (Export)
 Uplink Peak Load  Downlink Peak Load

CSPeakConn= ErlangB(AverageTraffic × (1 + SoftHoFactorwithoutSof er ),Gos)

AMR12.2k: AMR12.2k:
ErlangB(40×(1+30%),2%)=63 ErlangB(40×(1+30%),2%)=63
CS64k: CS64k:
ErlangB(2×(1+30%),2%)=7 ErlangB(2×(1+30%),2%)=7
bit/s) = PeakConn× IubBW × ActivityFactorVoicecall
CSPeakBW(K
AMR12.2k: AMR12.2k:
=63×20×0.5=630 =63×20×0.5=630
CS64k: CS64k:
=7×74=518 =7×74=518

Huawei Technologies Co., Ltd. All rights reserved. Page101


Case Study (Export)
 Uplink Traffic BW

ULTrafficBW = Max(CSPeak
BW,
CS Peak BW= 630+518=1148
CSAverageB W + PSAverageBW
CS Average BW=520+182=702 + HSUPAAvera geBW)
PS Average BW=53.1

HSUPA Average BW=975.63 DLTrafficBW = Max(CSPeakBW,


 Downlink Traffic BW CSAverageB W + PSAverageBW
+ HSDPAAvera
geBW)
CS Peak BW= 630+518=1148

CS Average BW=520+182=702

PS Average BW= 105.05 +198.74=303.79

HSDPA Average BW= 3857.6

Huawei Technologies Co., Ltd. All rights reserved. Page102


Case Study (Export)
IubBW = Max(ULBW,DLBW)+ IubO& M

ULBW = ULTrafficBW + ULSignalling + ULCCHBW

DLBW = DLTrafficBW + DLSignalling + DLCCHBW

ULBW = Max(1148,702 + 53.1 + 975.63) × (1 + 10%) + 50 × 3

DLBW = Max(1148,702 + 303.79 + 3857.6) × (1 + 10%) + 70 × 3

IubBW = Max(1730.73,4863.44) + 64

Question : How many E1T1 should be configure?

Huawei Technologies Co., Ltd. All rights reserved. Page103


Contents

1. Traffic Model

2. Interference Analysis

3. Capacity Dimensioning

4. CE Dimensioning

5. IUB Bandwidth Dimensioning

6. Network Dimensioning Flow

Huawei Technologies Co., Ltd. All rights reserved. Page104


Network Dimensioning Flow
Start

Coverage Requirement UL/DL Link Budget

Cell Radius=Min (RUL, RDL)

UL/DL Capacity
Capacity Requirement
Dimensioning

No
Satisfy Capacity Requirement? Adjust Carrier/NodeB

Yes
CE Dimensioning

IUB Bandwidth Dimensioning

Output NodeB Amount/


NodeB Configuration

End

Huawei Technologies Co., Ltd. All rights reserved. Page105


Thank you
www.huawei.com

You might also like