TJCWR1207067E0 WCDMA RNS Product Training Manual I

WCDMA RNS Product Training Manual
Volume I

ZTE CORPORATION

COPYRIGHT

Copyright ZTE Corporation
All rights reserved.
All information contained herein are confidential information of ZTE and must be
handled with highest care. Nobody can, for any purpose, copy, save, link to searching
tools, or distribute by any means (including but not limited to electronic, mechanical,
photocopying, recording means) of the above mentioned information without prior
written consent of ZTE.

* * * *
ZTE UNIVERSITY
ZTE University, Dameisha, Yantian District, Shenzhen, P.R.China
Postcode: 518083
Tel: (+86755) 26778800
Fax: (+86755) 26778999
ZTE CORPORATION
ZTE Plaza, Keji Road South, Hi-Tech Industrial Park, Nanshan District, Shenzhen, P.R.China
Postcode: 518057
ZTE University Websithttp://univ.zte.com.cn
Client Support Hot line+8675526770800 800-830-1118
Fax: (+86755) 26770801
* * * *
Version: 2012, first edition
S.N.: TJCWR1207067E0
General

Preface
Thanks for using Manual for WCDMA RNS Product Training Manual Volume I In order to use the
Manual properly, please read the Preface first.
1. Application
It should not be used for the purpose of on-site installation or trouble shooting.
2. About This Manual
This manual is composed of Five volumesand the table of contents of each volume is shown below:
Volume Course Name
I
WR_BT1003_E01_0 Channel Structure and function
WR_BT1011_E01_0 HSDPA Technology
WR_BT1012_E01_0 HSUPA Technology
WR_SS1011_E02_0 ZXWR RNC Introduction
II
GU_SS1021_E02_0 ZXSDR BTS Introduction
WR_OC2011_E02_0 ZXWR RNC Quick Data Configuration
III
WR_OC2031_E03_0 ZXSDR Node B Data Configuration
WR_TS3012_E01_0 ZXWR RNC Troubleshooting (Equipment Class)
WR_TS3013_E01_0 ZXWR RNC Troubleshooting (Interconnection Class)
WR_TS3014_E01_0 ZXWR RNC Troubleshooting (NM Class)
WR_TS3031_E01_0 ZXSDR Node B Troubleshooting

Volume Course Name
OM_SS201_E01_0 RAN EMS Software Platform Introduction
IV
OM_OC202_E01_0 RAN EMS Fault Management
OM_OC203_E01_0 RAN EMS Performance Management
V
OM_OC204_E01_0 RAN EMS Security Management
OM_OC205_E01_0 RAN EMS Maintenance Management
OM_SM201_E01_0 RAN EMS Maintenance and Troubleshooting

3. Manual Update history
Version Date Comments
1.0 2012 New

4. From the Author
Thank you for using this manual and your continuous support. We would appreciate your comments
and suggestions on this Manual.
We can be reached at
Telephone+8675526778072
Fax+8675526778999

ZTE UNIVERSITY

Channel Structure and
function
ZTE CORPORATION
ZTE Plaza, Keji Road South,
Hi-Tech Industrial Park,
Nanshan District, Shenzhen,
P. R. China
518057
Tel: (86) 755 26771900 800-9830-9830
Fax: (86) 755 26772236
URL: http://support.zte.com.cn
E-mail: doc@zte.com.cn

Contents

Chapter 1 .......................................................................... 1
Chanel structure .............................................................. 1
Channels of UTRAN ......................................................... 1
Logical channels ................................................................... 2
Transport channels ................................................................ 3
Physical channels .................................................................. 5
Channel Mapping ................................................................ 19
Physical layer porcedure ................................................ 20
Cell search procedure .......................................................... 20
Random Access Procedure .................................................... 21

Confidential and Proprietary Information of ZTE CORPORATION 1
Ch a p t e r 1
Chanel structure

knowledgement
mapping of logical channels onto transport channels
mapping of transport channels and physical channels
physical layer procedure
Channels of UTRAN
Channels of UTRAN are divided into
logical channels
transport channels
physical channels
In air interface protocol modele of UTRAN, MAC layer accomplish
the mapping of logical channels onto transport channels, PHY
layer accomplish the mapping of transport channels onto
physical channels. Figure 1 shows the position of logical
channels and transport channels

Channel Structure and function

2 Confidential and Proprietary Information of ZTE CORPORATION

FI GURE 1 L OGI CAL CHANNEL S AND TRANSPORT CHANNEL S
DuplicationAvoidance
GC Nt DC
GC Nt DC
RLC
RLC
RLC
RLC
RLC
RLC
RLC
RLC
BMC
PDCP
PDCP
C-planesignalling U-planeinformation
UuS boundary
c
o
n
t
r
o
l
c
o
n
t
r
o
l
c
o
n
t
r
o
l
c
o
n
t
r
o
l
L3
Radio
Bearers
L2/PDCP
L2/BMC
L2/RLC
Logical
Channels
L2/MAC
Transport
Channels
L1
control
RRC
MAC
PHY

Logical channels
MAC layer implement the mapping between logical channel and
transport channel and provide data transport service for logical.
Logical channels are divided into control channel and traffic
channel. Fig illustrates the structure of logical channels:

Chapter 1 Chanel structure

FI GURE 2 L OGI CAL CHANNEL STRUCTURE

Control channels only used to transport control plane
information. Including BCCH,PCCH,CCCH,DCCH,SHCCH.
Traffic channels only used to transport user plane information.
Inculuding DTCH,CTCH.
Transport channels
Transport channels are services offered by Layer 1 to the higher
layersA transport channel is defined by how and with what
characteristics data is transferred over the air interface. A
general classification of transport channels is into two groups:
Dedicated channel: used by dedicated user.
Common channel: used by all users within one cell.
Show as Figure 3



FI GURE 3 TRANSPORT CHA NNEL S
BCH
PCH
FACH
DSCH
RACH
CPCH
DCH
DCH

There exists only one type of dedicated transport channel, the
Dedicated Channel (DCH). The Dedicated Channel (DCH) is a
downlink or uplink transport channel. The DCH is transmitted
over the entire cell or over only a part of the cell using e.g.
beam-forming antennas.
There are six types of common transport channels: BCH, FACH,
PCH, RACH, CPCH and DSCH.
Broadcast Channel -BCH
The Broadcast Channel (BCH) is a downlink transport
channel that is used to broadcast system- and
cell-specific information. The BCH is always transmitted
over the entire cell and has a single transport format.
Forward Access Channel -FACH
The Forward Access Channel (FACH) is a downlink
transport channel. The FACH is transmitted over the
entire cell. It is used to transmit control imformation after
Basestation have received the random access
requirement sent by UE. FACH is also can be used to
send paket data.
There is one or several FACH within one cell. One of them
must has low datarate so all terminals in this cell can
receive it. The others can have high datarate.
Paging Channel - PCH
The Paging Channel (PCH) is a downlink transport
channel. The PCH is always transmitted over the entire
cell. The transmission of the PCH is associated with the
transmission of physical-layer generated Paging
Indicators, to support efficient sleep-mode procedures.
Random Access Channel - RACH
The Random Access Channel (RACH) is an uplink
transport channel. The RACH is always received from the
entire cell. It is used to carry control information (such as
call setup request) sent by UE.The RACH is characterized


by a collision risk and by being transmitted using open
loop power control.
Commnon Packet Channel - CPCH
The Common Packet Channel (CPCH) is an uplink
transport channel. CPCH is associated with a dedicated
channel on the downlink which provides power control
and CPCH Control Commands (e.g. Emergency Stop) for
the uplink CPCH. The CPCH is characterised by initial
collision risk and by being transmitted using inner loop
power control.
Downlink Shared Channel - DSCH
The Downlink Shared Channel (DSCH) is a downlink
transport channel shared by several UEs The DSCH is
associated with one or several downlink DCH. The DSCH
is transmitted over the entire cell or over only a part of
the cell using e.g. beam-forming antennas.
Physical channels
Physical channels are defined by a specific carrier frequency,
scrambling code, channelization code (optional), time start &
stop (giving a duration) and, on the uplink, relative phase (0 or
t/2). There is 2 types of physical channel. Uplink physical and
downlink physical.
Uplink physical channels
There are 2 types of uplink dedicated physical channels(Uplink
Dedicated Physical Data Channel and Uplink Dedicated Physical
Control Channel) and 2 types of uplink common physical
channels( Physical Random Access Channel and Physical
Common Packet Channel) illustrate as Figure 4
FI GURE 4 UPL I NK PHYSI CAL CHANNEL S

Uplink dedicated physical channels



There are two types of uplink dedicated physical channels, the
uplink Dedicated Physical Data Channel (uplink DPDCH) and the
uplink Dedicated Physical Control Channel (uplink DPCCH).The
DPDCH and the DPCCH are I/Q code multiplexed within each
radio frame
The uplink DPDCH is used to carry the DCH transport channel.
There may be zero, one, or several uplink DPDCHs on each radio
link.
The uplink DPCCH is used to carry control information generated
at Layer 1. The Layer 1 control information consists of known
pilot bits to support channel estimation for coherent detection,
transmit power-control (TPC) commands, feedback information
(FBI), and an optional transport-format combination indicator
(TFCI). The transport-format combination indicator informs the
receiver about the instantaneous transport format combination
of the transport channels mapped to the simultaneously
transmitted uplink DPDCH radio frame. There is one and only
one uplink DPCCH on each radio link.

Figure 5 shows the frame structure of the uplink dedicated
physical channels. Each radio frame of length 10 ms is split into
15 slots, each of length T
slot
= 2560 chips, corresponding to one
power-control period.
FI GURE 5 FRAME STRUCTURE OF UPL I NK DEDI CATED PHYSI CAL CHA NNEL

The parameter k in figure determines the number of bits per
uplink DPDCH slot. It is related to the spreading factor SF of the
DPDCH as SF = 256/2
k
. The DPDCH spreading factor may range
from 256 down to 4. The spreading factor of the uplink DPCCH is
always equal to 256, i.e. there are 10 bits per uplink DPCCH
slot.
The exact number of bits of the uplink DPDCH and the different
uplink DPCCH fields (N
pilot
, N
TFCI
, N
FBI
, and N
TPC
) is configured by
higher layers and can also be reconfigured by higher layers.


The FBI bits are used to support techniques requiring feedback
from the UE to the UTRAN Access Point, including closed loop
mode transmit diversity and site selection diversity transmission
(SSDT).
There are two types of uplink dedicated physical channels; those
that include TFCI (e.g. for several simultaneous services) and
those that do not include TFCI (e.g. for fixed-rate services). It is
the UTRAN that determines if a TFCI should be transmitted and
it is mandatory for all UEs to support the use of TFCI in the
uplink.
Npilot34567and 8The shadowed column part of pilot
bit pattern is defined as FSW and FSWs can be used to confirm
frame synchronization. (The value of the pilot bit pattern other
than FSWs shall be "1".)
TPC is corresponding to power contrl command.
Multi-code operation is possible for the uplink dedicated physical
channels. When multi-code transmission is used, several parallel
DPDCH are transmitted using different channelization codes.
However, there is only one DPCCH per radio link.
Uplink common physical channels
Physical Random Access Channel (PRACH)
The random-access transmission is based on a Slotted
ALOHA approach with fast acquisition indication. The UE can
start the random-access transmission at the beginning of a
number of well-defined time intervals, denoted access slots.
There are 15 access slots per two frames and they are
spaced 5120 chips apart Information on what access slots
are available for random-access transmission is given by
higher layers.The random-access transmission consists of
one or several preambles of length 4096 chips and a
message of length 10 ms or 20 ms.
FI GURE 6 FRAME STRUCTURE OF RANDOM ACCESS CHANNEL



Pilot
N
pilot
bits
Data
N
data
bits
Slot #0 Slot #1 Slot #i Slot #14
T
slot
=2560 chips, 10*2
k
bits (k=0..3)
Message part radio frame T
RACH
=10 ms
Data
Control
TFCI
N
TFCI
bits

Each preamble is of length 4096 chips and consists of 256
repetitions of a signature of length 16 chips. There are a
maximum of 16 available signatures .
The 10 ms message part radio frame is split into 15 slots,
each of length T
slot
= 2560 chips. Each slot consists of two
parts, a data part to which the RACH transport channel is
mapped and a control part that carries Layer 1 control
information. The data and control parts are transmitted in
parallel.
A 10 ms message part consists of one message part radio
frame, while a 20 ms message part consists of two
consecutive 10 ms message part radio frames. The message
part length is equal to the Transmission Time Interval of the
RACH Transport channel in use. This TTI length is configured
by higher layers.
The data part consists of 10*2
k
bits, where k=0,1,2,3. This
corresponds to a spreading factor of 256, 128, 64, and 32
respectively for the message data part.
The control part consists of 8 known pilot bits to support
channel estimation for coherent detection and 2 TFCI bits.
This corresponds to a spreading factor of 256 for the
message control part. The pilot bit pattern is described in
table 8. The total number of TFCI bits in the random-access
message is 15*2 = 30. The TFCI of a radio frame indicates
the transport format of the RACH transport channel mapped
to the simultaneously transmitted message part radio frame.
In case of a 20 ms PRACH message part, the TFCI is
repeated in the second radio frame.
Physical Common Packet ChannelPCPCH
The Physical Common Packet Channel (PCPCH) is used to
carry the CPCH CPCH. The CPCH transmission is based on
DSMA-CD approach with fast acquisition indication. The UE
can start transmission at the beginning of a number of
well-defined time-intervals, relative to the frame boundary of
the received BCH of the current cell.
The PCPCH access transmission consists of one or several
Access Preambles [A-P] of length 4096 chips, one Collision
Detection Preamble (CD-P) of length 4096 chips, a DPCCH


Power Control Preamble (PC-P) which is either 0 slots or 8
slots in length, and a message of variable length Nx10 ms.
FI GURE 7 FRAME STRUCTURE OF CPCH

Pilot
N
pilot
bits
TPC
N
TPC
bits
Data
N
data
bits
T
slot
=2560 chips, 10*2
k
bits (k=0..6)
1 radio frame: T
f
=10 ms
Data
Control
FBI
N
FBI
bits
TFCI
N
TFCI
bits

CPCH access preamble part
Similar to RACH preamble part. The RACH preamble signature
sequences are used. The number of sequences used could be
less than the ones used in the RACH preamble. The scrambling
code could either be chosen to be a different code segment of
the Gold code used to form the scrambling code of the RACH
preambles or could be the same scrambling code in case the
signature set is shared.

CPCH collision detection preamble part
Similar to RACH preamble part. The RACH preamble signature
sequences are used. The scrambling code is chosen to be a
different code segment of the Gold code used to form the
scrambling code for the RACH and CPCH preambles

CPCH power control preamble part



The power control preamble segment is called the CPCH Power
Control Preamble (PC-P) part. The Power Control Preamble
length is a higher layer parameter, L
pc-preamble
, which shall take
the value 0 or 8 slots. The TFCI field is filled with "1" bits.

CPCH message part
Similar to uplink dedicated channel, Each 10 ms frame is split
into 15 slots, each of length T
slot
= 2560 chips. Each slot consists
of two parts, a data part that carries higher layer information
and a control part that carries Layer 1 control information. The
data and control parts are transmitted in parallel. The sf of
CPCH message part is 256
Downlink physical channel
Downlink physical channels include Detedicated physical channel
\ one Shared Physical Channel\ five Commnon Control Channel
downlink Detedicated physical channel -DPCH
primary and secondary Commnon Pilot Channel - CPICH
primary and secondary Common Control Physicl Channel
- CCPCH
Synchronous Channel - SCH
Physical Downlink Shared Channel - DSCH
Acquisition Indication Channel - AICH
Paging Indication Channel - PICH
downlink physical are illustrated as Figure 8
FI GURE 8

There is only one type of downlink dedicated physical channel,
the Downlink Dedicated Physical Channel (downlink DPCH).
Downlink Dedicated
Physical Channel


Within one downlink DPCH, dedicated data generated at Layer 2
and above, i.e. the dedicated transport channel (DCH), is
transmitted in time-multiplex with control information generated
at Layer 1 (known pilot bits, TPC commands, and an optional
TFCI). The downlink DPCH can thus be seen as a time multiplex
of a downlink DPDCH and a downlink DPCCH.
Each frame of length 10 ms is split into 15 slots, each of length
T
slot
= 2560 chips, corresponding to one power-control period.
FI GURE 9 FRAME STRUCTURE OF DL DPCH

The parameter k in figure 9 determines the total number of bits
per downlink DPCH slot. It is related to the spreading factor SF
of the physical channel as SF = 512/2
k
. The spreading factor
may thus range from 512 down to 4.
The exact number of bits of the different downlink DPCH fields
(N
pilot
, N
TPC
, N
TFCI
, N
data1
and N
data2
) is given in table 11. What slot
format to use is configured by higher layers and can also be
reconfigured by higher layers.
There are basically two types of downlink Dedicated Physical
Channels; those that include TFCI (e.g. for several simultaneous
services) and those that do not include TFCI (e.g. for fixed-rate
services). It is the UTRAN that determines if a TFCI should be
transmitted and it is mandatory for all UEs to support the use of
TFCI in the downlink.
Npilot248 & 16
TPC symbol is corresponding to transimission power control
command T 0or1
Downlink Common Physical Channel
Common Pilot Channel (CPICH)
The CPICH is a fixed rate (30 kbps, SF=256) downlink
physical channel that carries a pre-defined bit sequence.
There are two types of Common pilot channels, the Primary
and Secondary CPICH. They differ in their use and the



limitations placed on their physical features.Fig show the
frame structure of CPICH.
Pre-defined bit sequence
Tslot =2560 chips , 20 bits
1 radio frame: Tf =10 ms

slot #1
Frame#i+1 Frame#i
slot #14
Antenna2
Antenna1
slot #0
FrameBoundary
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
11 11 00 00 11 11 00 00 11 00 11 11 00 00 11 11 00 00 11 11 00 00 11 11

The Primary Common Pilot Channel (P-CPICH) has the following
characteristics:
The same channelization code is always used for the P-CPICH,
see [4];
The P-CPICH is scrambled by the primary scrambling code,
see [4];
- There is one and only one P-CPICH per cell;
- The P-CPICH is broadcast over the entire cell.
The Primary CPICH is a phase reference for the following
downlink channels: SCH, Primary CCPCH, AICH, PICH AP-AICH,
CD/CA-ICH, CSICH, DL-DPCCH for CPCH and the S-CCPCH. By
default, the Primary CPICH is also a phase reference for
downlink DPCH and any associated PDSCH. The UE is informed
by higher layer signalling if the P-CPICH is not a phase reference
for a downlink DPCH and any associated PDSCH.The Primary
CPICH is always a phase reference for a downlink physical
channel using closed loop TX diversity.
A Secondary Common Pilot Channel (S-CPICH) has the following
characteristics:
An arbitrary channelization code of SF=256 is used for the
S-CPICH, see [4];
A S-CPICH is scrambled by either the primary or a secondary
scrambling code,
There may be zero, one, or several S-CPICH per cell;


A S-CPICH may be transmitted over the entire cell or only
over a part of the cell;
A Secondary CPICH may be a phase reference for a
downlink DPCH. If this is the case, the UE is informed about
this by higher-layer signalling.The Secondary CPICH can be
a phase reference for a downlink physical channel using
open loop TX diversity, instead of the Primary CPICH being a
phase reference.
Note that it is possible that neither the P-CPICH nor any
S-CPICH is a phase reference for a downlink DPCH.
Primary Commnon Control Physical Channel (P-CCPCH)
Common control physical channel consists of Primary
Common Control Physical Channel (PCCPCH) and Secondary
Common Control Physical Channel (SCCPCH).
The Primary CCPCH is a fixed rate (30 kbps, SF=256)
downlink physical channels used to carry the BCH transport
channel.

Data
N
data1
=18 bits
T
slot
=2560 chips , 20 bits
1 radio frame: T
f
=10 ms
(Tx OFF)
256 chips

Figure shows the frame structure of the Primary CCPCH. The
frame structure differs from the downlink DPCH in that no
TPC commands, no TFCI and no pilot bits are transmitted.
The Primary CCPCH is not transmitted during the first 256
chips of each slot. Instead, Primary SCH and Secondary SCH
are transmitted during this period
econdary Commnon Control Physical Channel (P-CCPCH)
The Secondary CCPCH is used to carry the FACH and PCH.
There are two types of Secondary CCPCH: those that include
TFCI and those that do not include TFCI. It is the UTRAN that
determines if a TFCI should be transmitted, hence making it
mandatory for all UEs to support the use of TFCI. The set of
possible rates for the Secondary CCPCH is the same as for
the downlink DPCH. Fig show the frame structure of
S-CCPCH



T
slot
=2560 chips, 20*2
k
bits (k=0..6)
Pilot
N
pilot
bits
Data
N
data1
bits
1 radio frame: T
f
=10 ms
TFCI
N
TFCI
bits

The parameter k in figure 17 determines the total number of
bits per downlink Secondary CCPCH slot. It is related to the
spreading factor SF of the physical channel as SF = 256/2k.
The spreading factor range is from 256 down to 4. The FACH
and PCH can be mapped to the same or to separate
Secondary CCPCHs.
If FACH and PCH are mapped to the same Secondary CCPCH,
they can be mapped to the same frame. The main difference
between a CCPCH and a downlink dedicated physical channel
is that a CCPCH is not inner-loop power controlled. The main
difference between the Primary and Secondary CCPCH is that
the transport channel mapped to the Primary CCPCH (BCH)
can only have a fixed predefined transport format
combination, while the Secondary CCPCH support multiple
transport format combinations using TFCI.
Synchronous Channel (SCH)
The Synchronisation Channel (SCH) is a downlink signal used
for cell search. The SCH consists of two sub channels, the
Primary and Secondary SCH. The 10 ms radio frames of the
Primary and Secondary SCH are divided into 15 slots, each
of length 2560 chips. Figure illustrates the structure of the
SCH radio frame.

Primary
SCH
Secondary
SCH
256 chips
2560 chips
One 10 ms SCH radio frame
ac
s
i,0
ac
p
ac
s
i,1
ac
p
ac
s
i,14
ac
p
Slot #0 Slot #1 Slot #14
Th
e Primary SCH consists of a modulated code of length 256
chips, the Primary Synchronisation Code (PSC) denoted cp in
figure , transmitted once every slot. The PSC is the same for
every cell in the system.
The Secondary SCH consists of repeatedly transmitting a
length 15 sequence of modulated codes of length 256 chips,
the Secondary Synchronisation Codes (SSC), transmitted in


parallel with the Primary SCH. The SSC is denoted csi,k in
figure , where i = 0, 1, , 63 is the number of the
scrambling code group, and k = 0, 1, , 14 is the slot
number. Each SSC is chosen from a set of 16 different codes
of length 256. This sequence on the Secondary SCH indicates
which of the code groups the cell's downlink scrambling code
belongs to.
Physical Downlink Shared Channel (PDSCH)
The Physical Downlink Shared Channel (PDSCH) is used to
carry the Downlink Shared Channel (DSCH).
A PDSCH corresponds to a channelisation code below or at a
PDSCH root channelisation code. A PDSCH is allocated on a
radio frame basis to a single UE. Within one radio frame,
UTRAN may allocate different PDSCHs under the same
PDSCH root channelisation code to different UEs based on
code multiplexing. Within the same radio frame, multiple
parallel PDSCHs, with the same spreading factor, may be
allocated to a single UE. This is a special case of multicode
transmission. All the PDSCHs are operated with radio frame
synchronisation.
PDSCHs allocated to the same UE on different radio frames
may have different spreading factors.
The frame and slot structure of the PDSCH are shown on
figure .

T
slot
=2560 chips, 20*2
k
bits (k=0..6)
Data
N
data1
bits
1 radio frame: T
f
=10 ms

For each radio frame, each PDSCH is associated with one
downlink DPCH. The PDSCH and associated DPCH do not
necessarily have the same spreading factors and are not
necessarily frame aligned.
All relevant Layer 1 control information is transmitted on the
DPCCH part of the associated DPCH, i.e. the PDSCH does not
carry Layer 1 information. To indicate for UE that there is
data to decode on the DSCH, the TFCI field of the associated
DPCH shall be used.
The TFCI informs the UE of the instantaneous transport
format parameters related to the PDSCH as well as the
channelisation code of the PDSCH.
Acqusition Indication Channel(AICH)
The Acquisition Indicator channel (AICH) is a fixed rate
(SF=256) physical channel used to carry Acquisition



Indicators (AI). Acquisition Indicator AI
s
corresponds to
signature s on the PRACH.
Figure illustrates the structure of the AICH. The AICH
consists of a repeated sequence of 15 consecutive access
slots (AS), each of length 5120 chips. Each access slot
consists of two parts, an Acquisition-Indicator (AI) part
consisting of 32 real-valued symbols a
0
, , a
31
and a part of
duration 1024 chips with no transmission that is not formally
part of the AICH. The part of the slot with no transmission is
reserved for possible use by CSICH or possible future use by
other physical channels.
The spreading factor (SF) used for channelisation of the
AICH is 256.
The phase reference for the AICH is the Primary CPICH.
1024 chips
Transmission Off
AS #14 AS #0 AS #1 AS #i AS #14 AS #0
a1 a2 a0 a31 a30
AI part =4096chips, 32real-valuedsymbols
20 ms

CPCH Access Preamble Acqusition Indication Channel
(AP-AICH)
The Access Preamble Acquisition Indicator channel (AP-AICH)
is a fixed rate (SF=256) physical channel used to carry AP
acquisition indicators (API) of CPCH. AP acquisition indicator
API
s
corresponds to AP signature s transmitted by UE.
AP-AICH and AICH may use the same or different
channelisation codes. The phase reference for the AP-AICH is
the Primary CPICH. Figure illustrates the structure of
AP-AICH.
1024 chips
Transmission Off
AS #14 AS #0 AS #1 AS #i AS #14 AS #0
a1 a2 a0 a31 a30
API part =4096chips, 32real-valuedsymbols
20 ms

The AP-AICH has a part of duration 4096 chips where the AP
acquisition indicator (API) is transmitted, followed by a part
of duration 1024chips with no transmission that is not


formally part of the AP-AICH. The part of the slot with no
transmission is reserved for possible use by CSICH or
possible future use by other physical channels.
CPCH Collision Detection/Channel Assignment Indicator
Channel (CD/CA -ICH)
The Collision Detection Channel Assignment Indicator
channel (CD/CA-ICH) is a fixed rate (SF=256) physical
channel used to carry CD Indicator (CDI) only if the CA is not
active, or CD Indicator/CA Indicator (CDI/CAI) at the same
time if the CA is active. The structure of CD/CA-ICH is shown
in figure .
1024 chips
Transmission Off
AS #14 AS #0 AS #1 AS #i AS #14 AS #0
a1 a2 a0 a31 a30
CDI/CAI part =4096chips, 32real-valuedsymbols
20 ms

CD/CA-ICH and AP-AICH may use the same or different
channelisation codes.
The CD/CA-ICH has a part of duration of 4096chips where
the CDI/CAI is transmitted, followed by a part of duration
1024chips with no transmission that is not formally part of
the CD/CA-ICH. The part of the slot with no transmission is
reserved for possible use by CSICH or possible future use by
other physical channels.
The spreading factor (SF) used for channelisation of the
CD/CA-ICH is 256.
Paging Indication Channel (PICH)
The Paging Indicator Channel (PICH) is a fixed rate (SF=256)
physical channel used to carry the paging indicators. The
PICH is always associated with an S-CCPCH to which a PCH
transport channel is mapped.
Figure 24 illustrates the frame structure of the PICH. One
PICH radio frame of length 10 ms consists of 300 bits (b
0
,
b
1
, , b
299
). Of these, 288 bits (b
0
, b
1
, , b
287
) are used to
carry paging indicators. The remaining 12 bits are not
formally part of the PICH and shall not be transmitted (DTX).
The part of the frame with no transmission is reserved for
possible future use.



b
1
b
0
288 bits for paging indication
12 bits (transmission
off)
One radio frame (10 ms)
b
287
b
288
b
299
In each PICH frame, Np paging indicators {P
0
, , P
Np-1
} are
transmitted, where Np=18, 36, 72, or 144.
The PI calculated by higher layers for use for a certain UE, is
associated to the paging indicator P
q
, where q is computed as a
function of the PI computed by higher layers, the SFN of the
P-CCPCH radio frame during which the start of the PICH radio
frame occurs, and the number of paging indicators per frame
(Np):

( ) ( ) ( ) Np
Np
SFN SFN SFN SFN PI q mod
144
144 mod 512 / 64 / 8 / 18
|
|
.
|
\
|
(
+ + + + =

Further, the PI calculated by higher layers is associated with the
value of the paging indicator P
q
. If a paging indicator in a certain
frame is set to "1" it is an indication that UEs associated with
this paging indicator and PI should read the corresponding frame
of the associated S-CCPCH.
The PI bitmap in the PCH data frames over Iub contains
indication values for all higher layer PI values possible. Each bit
in the bitmap indicates if the paging indicator associated with
that particular PI shall be set to 0 or 1. Hence, the calculation in
the formula above is to be performed in Node B to make the
association between PI and P
q
..The mapping from {P
0
, , P
Np-1
}
to the PICH bits {b
0
, , b
287
} are according to Table 1.
TAB L E 1 MAPPI NG OF PAGI NG I NDI CATORS P
Q
TO PI CH B I TS
Number of paging
indicators per
frame (Np)
P
q
= 1 P
q
= 0
Np=18 {b
16q
, , b
16q+15
} =
{1, 1,, 1}
{b
16q
, , b
16q+15
} =
{0, 0,, 0}
Np=36 {b
8q
, , b
8q+7
} = {1,
1,, 1}
{b
8q
, , b
8q+7
} = {0,
0,, 0}
Np=72 {b
4q
, , b
4q+3
} = {1,
1,, 1}
{b
4q
, , b
4q+3
} = {0,
0,, 0}
Np=144 {b
2q
, b
2q+1
} = {1, 1} {b
2q
, b
2q+1
} = {0, 0}


Channel Mapping
Mapping between logical channels and transport
channels
Figure 10 illustrates the mapping between logical channel and
transport channels.
FI GURE 10 MAPPI NG B ETWEEN L OGI CAL CHANNEL AND TRANSPORT
CHANNEL S

Figure 11 illustrates the mapping between transport channel and
physical channels.
FI GURE 11 MAPPI NG B ETWEEN TRANSPORT CHANNEL AND PHYSI CAL
CHANNEL S.



In addition to the transport channels introduced earlier, there
exist physical channels tocarry only information relevant to
physical layer procedures. The Synchronisation Channel(SCH),
the Common Pilot Channel (CPICH) and the Acquisition
Indication Channel (AICH)are not directly visible to higher layers
and are mandatory from the system function point ofview, to be
transmitted from every base station. The CPCH Status Indication
Channel(CSICH) and the Collision Detection/Channel Assignment
Indication Channel (CD/CA-ICH) are needed if CPCH is used.
Physical layer porcedure
Cell search procedure
During the cell search, the UE searches for a cell and determines
the downlink scrambling code and frame synchronisation of that
cell. The cell search is typically carried out in three steps:
Step 1: slot synchronous
During the first step of the cell search procedure the UE uses the
SCHs primary synchronisation code to acquire slot
synchronisation to a cell. This is typically done with a single
matched filter (or any similar device) matched to the primary
synchronisation code which is common to all cells. The slot
timing of the cell can be obtained by detecting peaks in the
matched filter output.
Step 2: frame synchronous and code-group identification
During the second step of the cell search procedure, the UE uses
the SCHs secondary synchronisation code to find frame
synchronisation and identify the code group of the cell found in
the first step. This is done by correlating the received signal with
all possible secondary synchronisation code sequences, and
identifying the maximum correlation value. Since the cyclic shifts
of the sequences are unique the code group as well as the frame
synchronisation is determined.
Step 3: scrambling-code identification
During the third and last step of the cell search procedure, the
UE determines the exact primary scrambling code used by the
found cell. The primary scrambling code is typically identified
through symbol-by-symbol correlation over the CPICH with all
codes within the code group identified in the second step. After
the primary scrambling code has been identified, the Primary
CCPCH can be detected.And the system- and cell specific BCH
information can be read.


Random Access Procedure
At each initiation of the physical random access procedure,
Layer 1 shall receive the following information from the higher
layers (MAC):
The Transport Format to be used for the PRACH message part.
The ASC of the PRACH transmission.
The data to be transmitted (Transport Block Set).
The physical random-access procedure shall be performed as
follows:
1. Derive the available uplink access slots, in the next full
access slot set, for the set of available RACH sub-channels
within the ASC Randomly select one access slot among the
ones previously determined. If there is no access slot
available in the selected set, randomly select one uplink
access slot corresponding to the set of available RACH
sub-channels within the given ASC from the next access slot
set. The random function shall be such that each of the
allowed selections is chosen with equal probability.
2. Randomly select a signature from the set of available
signatures within the given ASC. The random function shall
be such that each of the allowed selections is chosen with
equal probability.
3. Set the Preamble Retransmission Counter to Preamble
Retrans Max.
4. Set the parameter Commanded Preamble Power to
Preamble_Initial_Power.
5. In the case that the Commanded Preamble Power exceeds
the maximum allowed value, set the preamble transmission
power to the maximum allowed power. In the case that the
Commanded Preamble Power is below the minimum level
required in [7], set the preamble transmission power to a
value, which shall be at or above the Commanded Preamble
Power and at or below the required minimum power specified
in [7]. Otherwise set the preamble transmission power to the
Commanded Preamble Power. Transmit a preamble using the
selected uplink access slot, signature, and preamble
transmission power.
6. If no positive or negative acquisition indicator (AI = +1
nor 1) corresponding to the selected signature is
detected in the downlink access slot corresponding to the
selected uplink access slot:
1) Select the next available access slot in the set of
available RACH sub-channels within the given ASC.
2) Randomly select a new signature from the set of
available signatures within the given ASC. The random



function shall be such that each of the allowed selections
is chosen with equal probability.
3) Increase the Commanded Preamble Power by AP
0
=
Power Ramp Step [dB]. If the Commanded Preamble
Power exceeds the maximum allowed power by 6dB, the
UE may pass L1 status ("No ack on AICH") to the higher
layers (MAC) and exit the physical random access
procedure.
4) Decrease the Preamble Retransmission Counter by one.
5) If the Preamble Retransmission Counter > 0 then repeat
from step 5. Otherwise pass L1 status ("No ack on AICH")
to the higher layers (MAC) and exit the physical random
access procedure.
7. If a negative acquisition indicator corresponding to the
selected signature is detected in the downlink access slot
corresponding to the selected uplink access slot, pass L1
status ("Nack on AICH received") to the higher layers (MAC)
and exit the physical random access procedure.
8. Transmit the random access message three or four uplink
access slots after the uplink access slot of the last
transmitted preamble depending on the AICH transmission
timing parameter. Transmission power of the control part of
the random access message should be P p-m [dB] higher
than the power of the last transmitted preamble.
9. Pass L1 status "RACH message transmitted" to the higher
layers and exit the physical random access procedure.

HSDPA Technology
ZTE CORPORATION
P. R. China
518057
Tel: (86) 755 26771900 800-9830-9830
Fax: (86) 755 26772236

Contents

Chapter 1 .......................................................................... 1
HSDPA Fundamental Principles and Key Technologies .. 1
HSDPA Summary ............................................................ 1
HSDPA Fundamental Principles .......................................... 2
HSDPA Channels ............................................................. 4
HS-DSCH ............................................................................. 4
HS-SCCH ............................................................................. 7
HS-DPCCH ........................................................................... 9
HSDPA Key Technologies .................................................. 9
AMC .................................................................................... 9
Fast HARQ ......................................................................... 11
Fast Scheduling .................................................................. 12
Mobility Management ..................................................... 12
HSDPA Terminal Requirements ....................................... 13
UE Type ............................................................................ 13
Working Mode .................................................................... 14
Frequency Band .................................................................. 14
Service Function Requirements ............................................. 14
Chapter 2 ........................................................................ 17
HSDPA Networking Policy and Network Planning ........ 17
HSDPA Technical Characteristics ...................................... 17
Co-CF/Hetero-CF Interference and Capacity Analysis .......... 19
HSDPA Solutions of ZTE ................................................. 23
HSDPA Introduction Policy .................................................... 23
HSDPA Networking Policy ..................................................... 24
Influence of HSDPA Introduction on Current Network Planning ... 28
Influence of HSDPA Introduction on Transmission Network ........ 36

HSDPA Network Upgrade Evolution Policy ......................... 37
Equipment Upgrade Influence and Requirements ..................... 37
Network Evolution Policy ...................................................... 39

Ch a p t e r 1
HSDPA Fundamental
Principles and Key
Technologies

Highlights
HSDPA fundamental principles
HSDPA channel overview
HSDPA key technologies
HSDPA Summary
In 3G times, the high-speed mobile data access services bring
huge profits to the operators. They have changed the decreasing
tendency of ARPU (Average Revenue Per user) in 2G times.
However, WCDMA, based on the R99/R4, can only provide the
transmission rate of maximum 2 Mbps, which can not meet the
increasing requirements for the high-speed mobile multimedia
services. At present, the appearance of CDMA 1x EV-DO
(Evolution, Data-Only), WLAN (Wireless Local Area Network) and
WiMAX technologies, WCDMA R99/R4 will drop behind as soon as
coming into commercial use.
Therefore, 3GPP decides to evolve the technology along with R5
HSDPA, which can meet the requirements of the transmission
bandwidth at suitable time. To meet the requirements of
uplink/downlink data service asymmetry, 3GPP has put forward
HSDPA (High Speed Downlink Packet Access) technology in R5.
HSDPA solves the contradiction between the system coverage and
the capacity, greatly increasing the system capacity and meeting
the requirements of uses for the high-speed services.
Meanwhile, one of the important characteristics of WCDMA is the
unbalance of uplink/downlink service traffic. Usually, the downlink
service traffic is larger the uplink service traffic. HSDPA can

HSDPA Technology

provide high-speed downlink access and low cost, which makes
HSDPA outstanding compared with other high-speed mobile data
access technologies and become the preferred technology when
WCDMA provides the high-speed mobile data access services.
HSDPA technology aims to providing high-speed downlink data
services for the multiple users. It is applicable to the services with
abundant download information, such as, multimedia and Internet.
HSDPA technology makes it possible for the data service to run on
the downlink at the rate of 10 Mbps. If MIMO (Multiple-Input
Multiple-Out-put) technology is adopted successfully, that is, the
multiple transmission antennas are used at the Node B side and
multiple reception antennas are use in UE, the rate can reach over
20 Mbps.
Through combining the fast physical layer (L1) re-transmission
and transmission that Node B control and the fast link adaptation
technology, HSDPA technology increases the throughput of the
packet data. The physical layer increases the rate through such
key technologies as AMC (Adaptive Modulation and Coding) and
HARQ (Hybrid Automatic Repeat reQuest). HS-DSCH (High Speed
Downlink Shared CHannel), HS-SCCH (High Speed Shared Control
CHannel) (the corresponding downlink control channel) and
HS-DPCCH (High Speed Dedicate Physical Control CHannel)
(uplink feedback channel) are added on L1. MAC-hs and the
corresponding HARQ protocol are added in Node B on L2 and L3.
HSDPA technology also asks for the modification on Iub/Iur
interface and adding the corresponding frame protocols.
Node B judges the channel quality according to the power control,
ACK/NACK proportion, QoS (Quality of Service) of each HSDPA
user, and the information that the dedicate HSDPA users feed back.
Besides, according to the current scheduling algorithm and user
PRI algorithm, Node B can perform the scheduling and the link
adaptation adjustment quickly.
HSDPA Fundamental
Principles
HSDPA technology is a new wireless network technology that 3GPP
WCDMA standards put forward in R5 protocol to meet the
requirements of the high-speed downlink data services. It does
not change the original WCDMA network structure in R99/R4, but
by introducing the short TTI (Transmission Time Interval = 2 ms),
AMC, HARQ of the multi-code transmission with fast physical layer
(L1), and the new MAC-hs entity, HSDPA technology can increase
the arte of the downlink data service to 100 Mbps. HSDPA is an
important technology that can increase the downlink capacity and
data service rate in the WCDMA network construction.
The basic principle of HSDPA technology is: When accessing into
the HSDPA wireless network and transmitting the downlink data,

Chapter 1 HSDPA Fundamental Principles and Key Technologies

UE periodically reports the CQI (Channel Quality Indicator) to
Node B, informing the data rate (the adopted AMC policy and the
code quantity) that the UE can support in the current wireless
environment. After receiving the data reported by UE, according to
the data QoS and the wireless environment, Node B selects the
suitable modulation mode (QPSK (Orthognal Phase Shift Keying)
or 16 QAM (Orthognal Amplitude Modulation) and TS (Time Slot)
to transmit the user downlink data on the HS-PDSCH (High Speed
Physical Downlink Shared CHannel) (dedicated for HSDPA). After
receiving the downlink data packet of Node B, the UE sends the
acknowledgement message (ACK/NACK) for each packet data
packet on HS-DPSCH, so that Node B knows whether the UE
receives the downlink data packet correctly, as shown in Figure 1.
Compared with R99 structure, HSDPA technology has introduced
the short TTI (2ms), AMC, and HARQ of the multi-code
transmission with fast physical layer (L1). In addition, HSDPA
technology moves the packet scheduler from RNC to Node B,
implementing the fast packet scheduling of the MAC-hs protocol
control in Node B.
After accessed to the HSDPA, UE can periodically reports the CQI
to Node B, informing the data rate (the adopted AMC policy and
the code quantity) that the UE can support in the current wireless
environment.
UE reports the acknowledgement information (ACK/NACK) of
every packer data, so that Node B can decide when and how to
re-transmit the data. If all UEs in the cell can measure the channel
quality, the fast packet scheduler can optimize the user scheduling.
For example, schedule the users according to the quality of
available channels, as shown in Figure 1.
FI GURE 1 HSDPA FUNDA MENTAL PRI NCI PL ES AND REL EVANT CHA NNEL S

HSDPA Technology

The above technologies work together, to increase the rate of the
downlink PS (Packet Service) data to 14.4 Mbps. This is
revolutionary in 3G network constriction to increase the downlink
capacity and to reduce the time delay. Besides, HSDPA can work
with OFDM and MIMO, providing the higher data traffic.
HSDPA Channels
Compared with R99 (and R4), HSDPA technology adds the
HS-PDSCH (compared with Node B), HS-SCCH (compared with
Node B), and HS-DPCCH (compared with UE) are newly added on
the physical layer. These physical channels are all 3-TS sub-frame
(2 ms) in the short frame format. They are all mapped onto the
HS-DSCH, as shown in Figure 2.
FI GURE 2 HSDPA RECEPTI ON AND FEEDB ACK
HS-PDSCH
HS-SCCH
HS-DPCCH (ACK/NACK and Feed back)
HS-SCCH
2 TS 7.5 TS +/- 128 Chip N TS
1 TS = 2560 Chip

HS-DSCH
HSDPA has introduces one new transmission channel, HS-DSCH,
to bear the user data. Its corresponding physical channel is
HS-PDSCH.
The structure of the HS-PDSCH sub-frame and TS is as shown in
Figure 3.


FI GURE 3 STRUCTURE OF HS- PDSCH SUB - FRAME AND TS

Slot #0 Slot#1 Slot #2
T
slot
=2560 chips, M*10*2
k
bits (k=4)
Data
N
data1
bits
1 subframe: T
f
=2 ms

HS-PDSCH can modulate in either the QPSK mode or 16QAM mode.
M is the bit that each modulation symbol represents. For example,
M = 2, it is QPSK mode. M = 4, it is 16QAM mode. The TS format
is as shown in Table 1.
TAB L E 1 HS- PDSCH TS FORMAT
Slot
format
Bit
Rate
(kbps)
Symbol
Rate
(ksps)
SF
Bits/
HS-DSCH
subframe
Bits/
Slot
Nda
ta
0(QPSK)) 480 240 16 960 320 320
1(16QAM) 960 240 16 1920 640 640
FI GURE 4 HSDPA PHYSI CAL CHANNEL AND TI ME SEQUENCE

HS-DSCH (High Speed Downlink Shared Channel) bears the
downlink user data. HS-DSCH TTI (interleave cycle) is 2 ms (three
TSs), so the round-trip delay between UE and Node B is smaller
during the re-transmission. The higher-phase modulation project
(such as, 16 QAM) is introduced to add the instantaneous peak
data rate. HS-DSCH supports the multi-code transmission and

HSDPA Technology

code multiplexing between different users. UE capability decides
the maximum bytes and it is 15 in the prescriptions. UE can select
5, 10, or 15 bytes. The users monitor the HS-SCCH signaling and
decide to use which HS-DSCH byte for de-spreading and receive
other necessary detection parameters.
HS-DSCH code resources include one or more channelized codes
whose SF (Spreading Factor) is 16 permanently. The available
code resources can be shared on the time domain. For example,
allocate one user with multiple channelized codes within one TTI
(2 ms). The code multiplexing technology makes it possible to
share the code resources in the code domain. Meanwhile, in the
same TTI, several users share the channelized code resources,
that is, the channelized codes fall into several subsets and then
are allocated to different users. Therefore, HS-DSCH supports not
only the transmission with smaller data traffic (only some
channelized codes on HS-DSCH are required for the data
transmission) but also the UE that can not de-spread all
channelized code sets.
HS-DSCH uses the shorter TTI (2 ms). The short TTI can decrease
the radio link adaptation time delay and increase the granularity of
the scheduling process. Therefore, it can trace the changeable
wireless environment better.
The structure of the HS-DSCH time division and code division is as
shown in Figure 5. The users share the downlink code resources
and power resources, performing the time division multiplexing.
This structure is applicable to the bursting packet data services.
FI GURE 5 MUL TI PL EXI NG STRUCTURE OF HS- DSCH TI ME DI VI SI ON AND
CODE DI VI SI ON

As shown in Figure 5, there are eight channelized codes
(corresponding to eight HS-PDSCH) allocated to HSDPA. UE


performs the time division multiplexing on these channelized
codes, and at the same time, performs the code division
multiplexing with the same TTI. For example, the first TTI in Figure
5, three channelized codes (three HS-PDSCH can be used to
receive the data) are allocated to UE3 and five are allocated to UE1.
In this mode, the transmission with smaller data traffic (not all
channelized codes are required) can be supported more flexible to
meet UEs with different capabilities. The protocol dives the UE
capability into twelve types, supporting five, ten, fifteen bytes.
Therefore, the channelized codes are required to be allocated to
different subsets, that is, code division multiplexed to different UE.
The number of subsets is variable, related to the Node B packet
scheduling policy. However, the code allocated to a specific UE
does not contain the capability beyond its receiving codes.
TAB L E 2 PACK ET DATA COMPARI SON OF HS- DSCH AND OTHER DOWNL I NK
CHANNEL
Channel
Type
HS-DSCH DSCH DCH FACH
SF 16 256 ~ 4 512 ~ 4 256 ~ 4
Modulation
Mode
GPSK/16QAM GPSK GPSK GPSK
Power
Control
Fixed/Slow
power control
Fast power
control
based on
DCH
Fast power
control at 1.5
khz
Fixed/Slo
w power
control
HARQ L1
implements
the packet
combination
RLC layer RLC layer RLC layer
Interleaving
Length
2 ms 10 ~ 80 ms 10 ~ 80 ms 10 ~ 80
ms
Channel
Coding
Project
Turrbo
coding
Turbo and
convolution
al coding
Turbo and
convolutional
coding
Turbo
and
convoluti
onal
coding
Transmission
Channel
Multiplexing
No Yes Yes Yes
Soft
Switchover
No No Yes No
Version R5 R99 R99 R99
HS-SCCH
Besides transmitting the user data, the Node B transmits the
control signaling to inform the next UE to wait for the scheduling.
The control signaling is borne on HS-SCCH. HS-SCCH can be

HSDPA Technology

time-division shared by all users. Compared with the
corresponding HS-DSCH TTI, HS-SCCH should be transmitted two
slots earlier compared with the corresponding HS-DSCH TTI.
FI GURE 6 REL ATI ON OF HS- SCCH AND HS- DSCH TI ME SEQUENCE

HS-SCCH bears the physical layer control signaling on HS-DSCH
for the decoding, such as, the adopted modulation coding policy,
channelized code, and HARQ. If the re-transmission or error
packet data occurs, this channel contains the possible physical
layer data combination. According to the maximum user number
of the multiplexing, UTRAN allocates the HS-SCCH. FI the
HS-DSCH does not bear the data,. Node B does not need to
transmit HS-SCCH. In addition, Node B can maintain the abundant
HS_SCCH but a single UE can support the reception of at most four
HS-SCCH. The network sends the signaling to inform the UE of the
HS-SCCH to receive.
Each HS-SCCH message module can hold three TSs, falling into
two parts. The first part (the first TS) bears the signaling sensitive
to the timing, which is to start the modulation process at the
regular time to avoid the buffering of the code-level data. The
second part (the rest two TSs) bear the signaling insensitive to the
timing, including the reliability CRC (Cycle Redundant Check)
codes of the HS-SCCH message and HARQ processing messages.
The two parts of HS-SCCH signaling perform the scrambling with
UE dedicate masks (in CRC), convenient for the UE to judge
whether the received HS-SCCH is sent to itself.
The first part parameters of HS-SCCH include:
The code bit number to de-spread the HS-DSCH; the channelized
code set allocation information 7 bit; the scheduling policy
information 1 bit; modulation mode; ID for HS-DSCH in QPSK or
16 QAM, 1bit.
Redundant information, used to perform the correct decoding and
data combination with the data received previously, 3 bit; ARQ
processing number, indicating the ARQ S.N. that the current data
belongs to; the first transmission ID or re-transmission ID,
indicating whether to combine the current data with the data in


the buffer (the data not correctly decoded previously) or to discard
the previous data and fill the buffer with the new data, 1 bit; the
length of the transmission module, 6 bit.
HS-DPCCH
HSDPA introduces one HS-DPCCH on the uplink, which bears the
uplink control signaling. That is, HS-DPCCH bears the CQI and
H-ARQ message, ACK/NACK. In Node B, the downlink quality is
fed back to decide which UE to transmit to and what rate to
transmit at. HS-DPCCH adds the new uplink channels on the
parallel code channel, because not all Nodes B can be upgraded to
support HSDPA.
FI GURE 7 STRUCTURE OF HS- DPCCH CHANNEL

HS-DPCCH falls into two parts, which bears the following
parameters.
ACK/NACK: Reflecting the results of the packet data decoding and
CRC result after the combination.
CQI of the downlink: Indicating the length of the transmission
module that can be received correctly, the length of the
transmission module that is estimated to receive the data on the
downlink, the modulation mode, and the number of the parallel
channels.
HSDPA Key Technologies
AMC
In the mobile wireless surroundings, the quality that UE receives
the signals depend on the distance between Node B and UE and
the channel attenuation (slow lognormal-fading and fast
Rayleigh-fading). According to the change on the time-varying
attenuation channel, the link adaptation technology adjusts the
transmission power, symbol rate, modulation orders, coding rate,
code plan, or combining the above factors to balance the link at
the real time. As a result, the system capacity is increased and the
communication quality is improved. The fast power control that

HSDPA Technology

the traditional WCDMA adopts is a kind of link adaptation
technology.
For the UE close to Node B, the power control can not decrease the
power to the largest degree. That is, when the power adjustment
reaches over 20 dB, the capacity is limited to increase. HSDPA link
adaptation function and AMC enabled the UE near Node B with
coding modulation combination (for user near the Node B, the
data throughput on the downlink can be increased to the largest
degree with the high-phase modulation and higher coding rate.
When the channel is in poor quality, AMC selects the low-phase
modulation and lower coding rate) of Ec/Ior (receiving signal
power/local cell interference). As a result, the user throughput is
increased unconditionally.
AMC is the typical link adaptation technology adopted in HSDPA.
Its core ides is to select the best downlink modulation and coding
mode according to the wireless channel quality (CQI repot) that
the current UE reports and the usage of the network resources, to
determine the rate of data transmission. Thus, the data
throughput of the terminal users in greatly increased and the
transmission delay is reduced. The data rate is adjusted by
changing the modulation policy, effective coding rate and the
number of HS-PDSCH codes. When the user is at the favorable
place (near the Node B), select the coding mode (such as, 16 QAM,
coding rate) with high-phase modulation and high rate to send
the user data, getting the higher transmission rate. When the user
is at unfavorable place, select the coding mode (such as, QPSK
modulation and coding rate) with low-phase modulation and
low rate, guaranteeing the communication quality. With HSDPA,
one user can use at most fifteen multi-code channels.
In R99, the physical channels use the variable SF and fast power
control technology. HSDPA uses the AMC (change the modulation
and coding format according to the temporary change on the
channel), multi-code transmission, fast and effective
re-transmission system.
HSDPA users can change the rate according to the modulation
mode, coding mode, and the number of the code channels, as
shown in Table 3.
TAB L E 3 HSDPA RATE
Modulation
Mode and
Coding Mode
Data rate
(One
Code)
Data rate
(Five
Codes)
Data rate
(10
Codes)
Data rate
(Fifteen
Codes)
QPSK 1/4 120kbps 600kbps 1.2Mbps 1.8Mbps
QPSK 1/2 240kbps 1.2Mbps 2.4Mbps 3.6Mbps
QPSK 3/4 360kbps 1.8Mbps 3.6Mbps 5.4Mbps
16QAM 1/2 480kbps 2.4Mbps 4.8 Mbps 7.2 Mbps
16QAM 3/4 720kbps 3.6Mbps 7.2 Mbps 10.8Mbps


Fast HARQ
ARQ is to ask for the re-transmission as soon as the data
transmission fails once. In the wireless transmission, the
transmission quality is poor because of the channel noise,
attenuation due to the mobility, and the interference due to other
users. To guarantee the communication quality, the data packet
must be protected. FEC (Forward Error Correction) codes are used
for the protection, that is, transmit the extra bit overhead in the
packet. Obviously, too many FEC codes make the transmission
efficiency become lower. Therefore, HARQ, a new hybrid error
correction method that is the combination of the ARQ and FEC, has
appeared. 3GPP introduces HARQ into HSDPA, reducing the time
delay and increasing the efficiency of the data transmission.
HARQ is a technology that combines FEC with the re-transmission.
It adjusts the transmission rate of the channels according to the
link situations, and combines FEC with the re-transmission. It is a
fast and effective re-transmission system.
In the case of decoding failure, the receiving party saves the
received data and requires the sending party to send the data
again. The receiving party combines the re-transmitted data with
the data received previously. HARQ technology can increase the
system performance and adjust the rate of the effective elements
flexibly.
If the data decode fails within one TTI, the data will be
re-transmitted in the following several TTIs. Once the data is
re-transmitted, UE combines the data of the current version with
the data re-transmitted currently, greatly increasing the
probability of the re-transmission success. The re-transmission
will cease until the data is decoded successfully or reaches the
pre-defined maximum attempt times. HARQ occurs in Node B, so
the re-transmission request can operate very fast.
HARQ can run in two modes. One is CC (Chase Combine) or Soft
Combining, that is, the re-transmission data is same as the initial
transmission. The other is IR (Incremental Redundancy), that is,
the retransmission data is different from the initial transmission.
The performance of IR is better than CC or Soft Combining,
because it can produce the diversity gains and increases the
coding efficiency. However, IR also requires larger memory at the
receiving end. The defaulted memory capacity at the terminal is
designed according to the maximum data rate that the terminal
can support and the Soft Combining mode. Therefore, only Soft
Combining mode can be used when the data is transmitted at the
largest rate. When the data is transmitted at lower rate, both
modes can be used. IR will increase the complexity, such as, the
UE memory. 3GPP does not prescribe which mode to use. CC can
be regarded as a special form of IR.
AMC is inconvenient when selecting a suitable modulation and
coding mode. The measurement for performing AMC may cause
the time delay. ARQ can automatically adapt to the temporary

HSDPA Technology

channel condition and is insensitive to the measurement delay and
errors. HSDPA combines AMC with HARQ nicely, achieving the
better link adaptation effect. AMC provides the rough data rate
selection while HARQ provides the precise rate adjustment based
on the channel condition.
In R99, the transmission channel ceases at RNC, so the packet
data is re-transmitted in SRNC. With the introduction of HS-DSCH,
Node B supports the MAC layer of HSDPA function. As a result,
Node B can directly control the re-transmission, increases the
re-transmission rate, and decreases the delay of the packet data.
In HSDPA, to lead in the traffic control system, the interface
between Node B and RNC should ensure that the Node B buffer can
be read and written normally, avoiding the data loss caused by the
buffer overflow.
Fast Scheduling
MAC-hs function is newly added in Node B, implementing the fast
packet scheduling of HSDPA and the real-time control of HS-DSCH.
The packet scheduling algorithm controls the fast allocation of the
shared resources, deciding the efficiency and performance of AMC
and HARQ to some degree. The packet scheduling algorithm can
best allocates the shared resources according to the quality of the
wireless channels, the data throughput waiting to transmit,
service PRI, UE capacity level, and the allocable resources. To suit
the fast change on the wireless channels, HSDPA puts the
scheduling functional unit in Node B instead of RNC. The
transmission intervals decrease to within 2 ms.
The typical packet scheduling is the Round-Robin realtime
scheduler. The users are served in order and by time, so the time
allocated to each user is same. However, the advanced scheduling
method is applicable because HSDPA has a high scheduling rate
and large AMC dynamic range. The channels are allocated
according to the current wireless surroundings. Proportional fair
packet scheduler is applicable because the order to provide service
to UE depends on the relative temporary channel quality, that is,
this scheduler tries to trace the fast attenuation feature of the
wireless channel. Since the UE gets the services according to the
relative surroundings, each user can get the similar time but the
system capacity increase by over 50%.
Mobility Management
The mobility procedures of HSDPA are different from the soft
switchover, where the UE keeps the HS-DSCH connection with
only one cell. In CELL_DCH status, UE moves from the Source
HS_DSCH Cell to the Target HS-DSCH Cell, depending on the
measurement report of UE and other information at the network
side. The network side controls the whole mobile process.


Updating the service cell is the soft-switchover process based on
CELL_DCH status, implemented by RRC signaling process.
FI GURE 8 HSDPA MOB I L I TY

Keep the configurations and the active sets of the dedicate
physical channels when updating the service cells and combine
them with the setup, release, and re-configuration of the dedicate
channels, or with the active set update during the soft switchover.
The service cell update falls into the synchronous update and
asynchronous update. The synchronous update is that the
network side controls the start and end of the sending and
receiving. The asynchronous update is that start and stop the data
sending and receiving at the UE and network side as quickly as
possible.
The service cells can be updated between cells in the same Node B
(intra Node B), or between cells in different Node B (inter Node B),
which is transparent for UE. In the inter Node B process, Node B
re-location of the service HS-DSCH may occur.
HSDPA Terminal
Requirements
UE Type
As a whole, HSDPA terminal (UE) includes the handy terminal (cell
phone), data card, and testing UE (tester). HSDPA UE falls into
twelve categories, Category 1 ~ Category 12, according to the
receiving capability. Different UE has different number of code
channels that UE supports, the minimum transmission intervals,

HSDPA Technology

the maximum transmission module size, the soft channel bit, and
the stimulated buffer. UE of Category has the poorest receiving
capability, Category 10 has the best, and Category 7 and 8 have
the medium.
Working Mode
HSDPA network supports both the UE with HSDPA function and
without HSDPA function.
To obtain the better and faster network services, HSDPA network
suggests using the dual-mode (GPRS/HSDPA) or multi-mode
(GSM/GPRS/HSDPA) terminal. The lowest requirements of the
terminal to access the HSDPA network is same as R99, that is, the
common WCDMA terminal. The future WCDMA terminal will
increase the uplink access bandwidth with HSUPA (High Speed
Uplink Packet Access) technology.
Frequency Band
The uplink is 1920 ~ 1980 MHz and the downlink is 2110 ~ 2170
MHz. The multi-mode terminal supporting GSM/GPRS supports
the requirements of the corresponding frequent band.
When accessing the HSDPA network, the interval of the
transmitting channel and its duplexing receiving channel is 134.8
MHz ~ 245.2 MHz.
The work mode, frequency band, and frequency intervals can be
selected manually. The default values are Auto.
Service Function Requirements
HSDPA terminal capacity levels is required to be CLASS A, which
supports the concurrence of the CS domain voice and PS domain
data, supporting the transmission power levels of Class 3 (24 dBm)
and Class 4 (21dBm).
Voice service: AMR12.2 K
Emergent call function: Supporting to make emergent calls no
matter the USIM/SIM card is inserted normally or is not inserted.
Supporting SMS (Short Message Service)
Supporting USSD (Unstructured Supplementary Service Data)
Data bearer service: All combined bearer services
Suggestions for other services:
When there is a new call accessing, the users should be
reminded with obvious visual symbols and ring tones.


When there is a call missed, the users should be reminded and
the related records should be saved.
The call records should contain the dialed, received, and
missed calls.
Numbers can be directly dialed from the contacts, records, and
short messages.

Ch a p t e r 2
HSDPA Networking Policy
and Network Planning

Highlights
HSDPA technical characteristics
HSDPA solutions of ZTE
HSDPA Technical
Characteristics
The comparison of R99 and HSDPA is as shown in Table 4.
TAB L E 4 R99/ HSDPA TECHNI CAL CHARACTERI STI CS
Item R99 HSDPA
System
Capacity (Mbps)
2.668 14.4 (physical layer)
Code Resource
Allocation
DPCH: Keep the high
usability rate of the
code table
HS-PDSCH: Allocate SF =
16 from right to left
HS-SCCH: Same as DPCH
System
Switchover
Hard switchover
Soft/Softer
switchover
Inter-system
switchover
Only hard switchover on
HS-PDSCH
Power control
Open-loop power
control and
closed-loop power
control
HS-PDSCH uses the slow
power control or does not
use the power control
Modulation
Mode
QPSK QPSK/16QAM
Link adaptation
technology
Fast power control
and soft switchover
AMC, HARQ, short frame
and fast channel feedback

HSDPA Technology

Item R99 HSDPA
Bit
Scrambling/De-
scrambling
N/A Only for HS-PDSCH
MAC-hs N/A For the fast scheduling
HSDPA network supports such high-speed PS (packet Service) as
HS-DSCH bearer streaming services (Type S), interactive services
(Type I), and background services (Type B), and DCH bearer data
and realtime voice. Type S includes the streaming services, such
as, IPTV, VOD (Video On Demand). Type I includes the services in
User Request Server Response mode, including the network
education, mobile bank, online game, location service. Type B
includes the services insensitive to the time delay but sensitive to
the errors, such as, data download, E-mail, SMS. The attributes of
the services are as shown in Table 5.
TAB L E 5 SERVI CE ATTRI B UTES
Servic
e
Bear
er
Mode
Requireme
nts for
Rate
Requirem
ents for
Time
Delay
Requirem
ents for
Data Error
Requireme
nts for
Network
Resources
Strea
ming
service

HS-D
SCH
High, and
the bit rate
should be
ensured
Low Very high More
Interac
tive
service
Low Low High Less
Backgr
ound
service
No
requirement
s
Very long Very high There are
idle
resources.
Realti
me
voice
DCH Very low Low Low More

WCDMA of ZTE supports the concurrence of the HSDPA services
and non-HSDPA services (such as, voice and videophone), and the
dynamic allocation of the code resources and power resources.
Usually, the spectrum frequency is indicated with the data
throughput per MHz and per cell of the transmitted services.
WCDMA spectrum frequency is high, as shown in Table 6.
TAB L E 6 SPECTRUM EFFI CI ENCY COMPARI SON
Item Spectrum Efficiency ( ) /( Cell MHz kbit
R99 537.6 7 PS384k
HSDPA 2795.2
Fifteen code channels
whose SF = 16

Chapter 2 HSDPA Networking Policy and Network Planning

Item Spectrum Efficiency ( ) /( Cell MHz kbit
CDMA2000 1x EV-DO 1920 TDMA
CDMA2000 1x EV-DV
2472 Release C
4000 Release D

HS-PDSCH uses the spreading factors whose SF = 16, so it has
smaller processing gains. Its coverage radius is smaller than that
of CS12.2 k voice (SF = 128) and larger than PS384 k (SF=8).
Usually, the coverage radius of the HSDPA is same as that of
PS128 k (1036m) in R99.
The cell radius of CDMA 2000 1x EV-DO is usually 907.5 m and it
is 643.7 m in dense areas.
Co-CF/Hetero-CF
Interference and Capacity
Analysis
When the basic R99 network construction completes and is ready
to update to HSDPA, the first thing to consider is the frequency
planning. In brief, HSDPA selects the R99 band or a new band to
be the working frequency. This section will analyze the WCDMA
inter-frequency interference.
When the operators complete the single-CF (Carrier Frequency)
coverage in the initial network construction, they will add CFs on
the network to absorb the traffic. The common method is to add
the power amplifier and to use the dual-CF. Using two
macro/micro cells with different CFs may result in the power
leaking between networks, and cause the interference finally.
WCDMA networking research of ZTE finds that R99 macro BS
constructs the hierarchical network with dual-CF, but the
inter-frequency isolation makes the interference smaller between
the two layers of the cells.

HSDPA Technology

FI GURE 9 HETERO- CF NETWORK I NG OF MACRO CEL L

According to TS25.942, the power attenuation difference between
the adjacent CF is 33 dB and that of the second neighboring CF is
43 dB. Considering the ACLR of the dynamic channel, the
difference may be larger. The difference between the CF channels
prevents the interference between dual-CF from affecting the
performance of the whole system.
The capacities of the macro cell on Layer A and that on Layer B
change when there is no inter-interference, when the first
neighboring CF is configured, and when the second neighboring CF
is configured. This change should be taken into consideration
when performing the emulation: The macro cell on Layer A share
the site with the macro cell on Layer B, so except the CF, all other
parameters are same.
There are interferences between cells of difference frequencies but
they interfere with each other very little. The first neighboring
frequency macro cell interferes the inter-system a little larger than
the second neighboring frequency macro cell does. The system
capacity that each carrier of the dual-carrier system corresponds
to is larger than the system overhead that the single-carrier
system results in. The emulation results show that the power that
the first neighboring frequency configuration results in is about 7
mw larger than the ideal BS power of the system without
inter-frequency interference, and the power that the second
neighboring frequency configuration results in is 6 mw larger than
the power that the first neighboring frequency configuration does.
Therefore, there is no need to calculate the inter-layer
interference caused by the hierarchical networking of the
hetero-frequency configuration.
To save the frequency resources, the operators usually adopt the
plan that HSDPA shares the power with R99 in the initial phase.
Therefore, the power resources reserved for HSDPA may have
effect on the previous R99 network capacity.


FI GURE 10 STI MUL ATI ON OF DOWNL I NK CAPACI TY AND TRANSMI SSI ON
POWER ( 12. 2 K VOI CE USER)

As shown in Figure 10, when the transmitting power of the system
is 10 W (4W is the common channel overhead), the system can
support about thirty-one users. When the transmitting power is 20
W, the system can support about forty users. That is, when R99
shares the same CF with HSDPA, the power resources for the
previous R99 users can be allocated to HSDPA users. When the
downlink power allocated to HSDPA users reaches 10 W, R99
network still supports thirty-one users, which is about 77% of the
power in the case of the full power configuration, 20 W. With the
increase of the actual power that the HSDPA users spend, the
interference in the same CF increases too. Therefore, the capacity
of R99 network decreases. Since WCDMA and HSDPA users are not
so many in the initial phase, when guaranteeing the certain
capacity oft the previous R99, preserve some power (such as, 30
~ 50%) to HSDPA users by configuring the power statically. The
appropriate power allocation method can both satisfy the R99
capacity and provide some HSDPA services.

HSDPA Technology

FI GURE 11 REL ATI ON OF R99 CAPACI TY AND DOWNL I NK TRANSMI SSI ON
POWER
55 60 65 70 75 80 85 90 95 100 105
0
10
20
30
40
50
60
70
80
90
100
B
S

D
L

T
x
P
w
r

A
s
s
i
g
n
R99 Capabilty Object %

FI GURE 12 I NFL UENCE OF HSDPA PRESERVED POWER ON R99 CAPA CI TY
-10 0 10 20 30 40 50 60 70
0
10
20
30
40
50
60
70
80
90
100
HSDPA Pwr Assign %
R
9
9

C
a
p
a
b
i
l
t
y

O
b
j
e
c
t

%

HSDPA is a downlink technology, sensitive to the downlink power
allocation. The power allocation directly influences the user
number threshold of the R99 downlink. As shown in Figure 11 and
Figure 12, when configuring the HSDPA in the same CF, to reach
100% of the R99 downlink capacity (forty-two users in emulation),
allocation all downlink power of the BS to R99 services. To reach
60%, allocate 35% of the power to the R99 services and the saved


power can be preserved for HSDPA services. The more power is
preserved for HSDPA services, the larger effect will be on R99
services. It is preferred to preserved 30 ~ 50 % of the power for
HSDPA services, and the loss of R99 capacity can be controlled
within 75 ~ 80%, which is tolerant.
HSDPA Solutions of ZTE
HSDPA Introduction Policy
The HSDPA technology provides the high-rate data service
downloading service and the first thing to research is the
application requirements for the data service model. In 3G times,
different user groups have different characteristics. The
middle-end and high-end user group ask for superior services,
such as, video stream services, high-speed file downloading
service. Domestic/Overseas data analysis prove that the
high-speed services occupy about 10 ~ 30%, which is low. The
number of the high-end users is one-fourth of the low-end users
but the throughput of a single high-end user is four times of the
low-user. The analysis can be the reference when introducing the
HSDPA policy. There are three methods to introduce HSDPA:
HSDPA independent networking
HSDPA global network coverage, based on WCDMA R99/R4
Build HSDPA network in key districts on the basis of WCDMA
R99/R4, to construct the quality network.
It costs a lot if building the HSDPA network independently and
CS/PS concurrence services can not be supported. Therefore, only
HSDPA sharing the network with R99/R4 can solve the problems.
ZTE introduces he HSDPA into R99/R4 network as the enhanced
technology for the network performances. HSDPA is introduced
into the current WCDMA R99/R4 is a good policy, which makes the
HSDPA smoothly updated on the current network and saves the
construction cost.
In planning, WCDMA covers from the dense cities and common
cities to suburbs and villages. The analysis on the service model
proves that the main services of the low-end user group are voice
and short message services and they ask for little for high-speed
data services. Most high-end users who are sensitive to the
high-speed services gather in the high-class office building, hotels,
and apartments in the dense cities, where the WCDMA network
emphasize on in the planning. HSDPA can be introduced into these
key places, satisfying the requirements of the high-end users and
setting up the surroundings for the high-speed service market.
When all conditions get mature, increase the market shares by
enlarging the HSDPA site range and adding CFs in the key districts.

HSDPA Technology

ZTE suggests setting up districted quality network of HSDPA in the
initial WCDMA network construction, or, if the cost allowing,
setting up WCDMA R99/R4 network that can be updated to HSDPA
in the whole network construction. The second plan is to enable
HSDPA services in the key districts and then with the maturity of
3G network, enlarging the HSDPA scale by enabling the HSDPA
function in the peripheral BSs step by step. Some districts without
a lot of people settled may become the hot spots of HSDPA
services, such as, high-class holiday inns and scenic areas. In this
case, the second plan has more advantages. Select the BS
equipment whose software supports the HSDPA upgrade, avoiding
the construction cost due to the hardware change in the latter
phase and avoiding the network operation risk.
ZTEs series BSs support upgrading to HSDPA in the software, so
during the network construction, the whole WCDMA network can
be set up to support software upgrading to HSDPA. After the
network comes into operation, HSDPA services can be enabled
step by step in where required according to the operators
requirements. The advanced protocol structure enables the
network to upgrade to HSDPA in the software and the HSDPA can
be deployed in the whole network rapidly and reliably.
HSDPA Networking Policy
HSDPA Sharing CF with R99/R4
There are two policies for HSDPA sharing the CF with R99/R4:
HSDPA shares the same CF with WCDMA R99/R4 and HSDPA sets
up the independent network with different CF with WCDMA
R99/R4.
1. HSDPA shares the CF with R99/R4
HSDPA and WCDMA work at the same frequency band and
share the same CF, BS power, channelized codes. They do
their own best in the uniform scheduling of the system,
enabling the users to enjoy the overall network services.
The advantages for the plan of HSDPA sharing the CF with
WCDMA:
WCDMA supports CS/PS services and HSDPA supports the
high-speed data service shared CF and power, which
makes the resources utilized to the most.
Flexible to select services and avoid such problems as UE
cell selection and residence due to different CF.
Supporting CS/PS concurrence services.
Guaranteeing the same coverage as WCDMA low-speed
data services. The link budget results indicate that when
the BS transmission power is 40 dBm, the CS64K video
phone service of the HSDPA and WCDMA have the same


maximum pathloss. That is, when the remaining
transmission power is abundant, the coverage range of the
HSDPA service and WCDMA traditional services are same.
The upgrade is easy and rapid. The cost is saved.
The problems for the plan of HSDPA sharing the CF with
WCDMA:
How to allocate the power appropriately for WCDMA
service and HSDPA service.
How to allocate the code resources and how to schedule.
The HSDPA users at the cell edge can perform the hard
switchover or switch by transferring from CELL_DCH
(HS-PDSCH) state to CELL_DCH (DCH) state. Which
switchover policy is better to guarantee the switchover
success ratio in different switchover cases?
Co-CF interference.
When HSDPA users and R99 users share the same CF, the
radio resources are shared, too. The configuration modes are
the static configuration and dynamic configuration.
The static configuration is that HSDPA and R99 allocate such
radio resources as the power and the codes at the fixed
proportion. The static configuration mode is simple and it can
keep the system stable and reliable. This mode is usually
applied in HSDPA Phase I.
The dynamic configuration is that the system allocates the
radio resources flexibly between HSDPA services and R99
service according to the load. The dynamic configuration mode
is complicated. In this mode, the utilization ratio of the system
resources is improved. This mode is usually applied in HSDPA
Phase II. The trigger system is applicable to the dynamic
configuration mode, such as, periodically or event triggering.
The dynamic configuration is to change the radio resources
proportion of R99 and HSDPA according to the actual cases.
2. HSDPA and R99/R4 use different CF
HSDPA performs the networking with WCDMA R99/R4 with
different CF and forms another layer of the network. To avoid
the switchover between different CF bands, HSDPA only
provides the high-speed data services. Usually, WCDMA data
card users are only provided with PS domain service and the
CS domain services are not supported.
R99/R4 network performs the bearer solution on the voice
service in CS domain independently.
The guidance policy in PS domain is that the low-speed PS
services go to R99/R4 first and the high-speed services
gather on HSDPA.
In this project, HSDPA and R99/R4 can perform the networking
independently and they will not influence with each other. The

HSDPA Technology

problems when HSDPA uses the different CF with R99/R4 in
the initial network construction phase:
HSDPA needs one carrier allocated independently, so the
cost for the network construction is high.
The users are not so many in the initial phase, so R99/R4
and HSDPA can occupy a respective carrier. The sharing
capacity is low and the overall network performance
decreases.
In the initial network construction, ZTE does not suggest this
hetero-CF plan. When the network develops to a certain degree
and the PS users increases step by step, a network can be set up
with a dependent frequency band to support only PS domain data
card. Meanwhile, attract more common mobile phone users with
the co-CF plan at other frequency bands.
Indoor HSDPA Planning
According to the experience of ZTE oversea WCDMA commercial
network, most 3G users and data traffic are indoors. Therefore,
the indoor system plays a very important part in the HSDPA
network.
In different kinds of large buildings, the building functions decide
the type of the people inside it. Their mobile possession ratio and
the utilization ratio are different. The objects of the indoor
distributed system are the large buildings where the indoor signals
have poor coverage, the traffic is large, and the requirements for
the communication quality are high. When analyzing the indoor
users, because of the difference of the user behaviors, make
different estimations for different areas in the building and then
sum them up. Decide whether to perform the HSDPA indoor
distributed system according to the services. The fields to perform
the HSDPA indoor distributed system include the civil aviation
airports, exhibition centers/conference centers/indoor
gymnasiums, commercial office buildings, hotels.
Combining with the experience of ZTE GSM and CDMA indoor
distributed system construction, the following problems should be
considered during the indoor HSDPA planning:
1. Extracting the signal source
The extraction principle: Calculate the capacity requirements
according to the user quantity and service requirements.
Combing with the capacity index of the signal source,
considering about the idle or occupied capacity of the
surrounding BSs, using the corresponding signal source as the
signal source of the indoor distributed system, the signal
source are reasonably make use of.
The indoor HSDPA can select the following signal sources:
Macro cell, micro cell, RF remote, and pico cell.


ZTE |suggests using Pico Cell and Pico RRU as HSDPA indoor
signal source. They can be placed independently in the key
districts indoors as HSDPA coverage, prevent from sharing the
indoor distributed antenna system with other indoor system
(WCDMA R99/GSM/PHS, etc.).
In the case of indoor and without obstacles, Pico Cell can cover
the area whose radius is 400 m. The indoor surroundings are
complicated. Usually, in the case of one layer of walls, each
Pico Cell covers the area whose radius is smaller than 40 m. In
this case, a single carrier provides eighty voice users or
sixty-four HSDPA users, which can satisfy the common indoor
applications.
Pico Node B is small and consumes low power. It is easy to
install indoors. Pico Node B is a WCDMA Node B product that
can satisfy the Pico Cell coverage. Pico Node B is equipped with
the complete Uu and Iub interfaces, meeting 3GPP
specifications. Pico RRU is a RF remote device that can be used
with the baseband pool. Uu interface can meet 3GPP
specifications. It does not have Iub interface. The data and
control interaction is performed through the fiber and
baseband pool connection, Pico Node B and Pico RRU belongs
to Pico Cell series products. The difference is that Pico RRU
does not have the baseband processing function and does not
have the Iub interface. Pico RRU is easy to operate on and
maintain. For other functions, Pico Node B and Pico Pico RRU
have the same performances.
Pico Cell and Pico RRU cover independently when there is
larger user traffic. In this case, each layer uses several Pico
Nodes B for the coverage. Each Pico Node B connects the RNC
in the equipment through the network cable or the fiber. The
advantages are:
Every Pico Node B can provide services, which is very
effective for the traffic in hot spots.
It avoids a lot of engineering jobs of the indoor distributed
system, such as, drilling holes on the wall, etc. It is most
suitable when the real estate does not allow the
construction.
Saving the cost of the passive apparatus, such as,
expensive RF feeders and leaked cables.
More Pico Nodes B are used in the project. With the same
cost, Pico Node B networking can obtain higher gains.
2. Indoor coverage system mode
R99 indoor coverage system falls into the distributed antenna
coverage system and the leaking cable coverage system. The
distributed antenna coverage system falls into the passive
coaxial distributed antenna system, active coaxial distributed
antenna system, and fiber distributed antenna system.
HSDPA has large transmitting power. Therefore, the power
increase of a single HSDPA distributed antenna will result in

HSDPA Technology

the overall power increase of other antennas. Take care when
using the indoor distributed antenna. In the initial phase of the
network construction, cover from he outdoor to indoor through
a certain sector signals of the outdoor macro cell. Set up the
indoor coverage through the micro cell, RF remote, and pico
cell in the buildings where abundant HSDPA are required. The
equipment can prevent the overall power of several distributed
antennas from ascending due to HSDPA. The cost is rather low.
Especially the pico cells can be used greatly indoors.
The above analysis proves that HSDPA indoor planning is
different from R99 indoor planning. ZTE suggests using the
micro cells and pico cells for HSDPA indoor planning and avoid
the indoor coverage with the distributed antenna. Determine
the key districts when making the HSDPA indoor planning,
such as, the VIP rooms, conference rooms. The pico cells and
micro cells with HSDPA can satisfy the requirements of the
high-speed data services. R99 coverage can satisfy the
common services. This plan is to configure HSDPA indoor
system flexibly without changing the R99 indoor planning.
Using the pico cells as HSDPA indoor system can save the cost
and construct the network rapidly.
Influence of HSDPA Introduction on
Current Network Planning
Influence on Code Resource Allocation
HSDPA is a high-speed shared downlink technology of WCDMA.
One of its characteristics is the high-speed spreading spectrum
code and multi-code transmission. Therefore, HSDPA asks for
large code resources. The three dedicate physical channels of
HSDPA are HS-PDSCH, HS-SCCH, and HS-DPCCH. Their
resources allocation has largest effect on the code resources of
the whole cell.
In R99, the downlink SF of CS12.2k service is 128. Excluding
the common channel overhead, the system supports about
123 ~ 124 users. When the R99 is upgraded to HSDPA, some
resources are allocated to HSDPA, so the previous R99
network will be affected.
HS-PDSCH code resources are allocated from right to left. The
occupies SF= 16 is 0 ~ 15. the cell common channels occupies
eight SF= 256. In one TTI, one UE performing the HSDPA
transmission needs at least one HS-SCCH (occupying SF = 128,
equal to two SF = 256). One HS-SCCH needs at least one
DPCCH (SF= 256, not supporting sharing, dedicate signaling).
That is, one HSDPA UE needs three SF = 256. The HSDPA user
occupying one code channel needs occupying the code
resources equal to nine SF = 128 (HS-PDSCH + HS-SCCH) and


one SF = 256 (DPCH). Calculating to 12.2 k voice user, it is
equal to 9.5 users.
FI GURE 13 I NFL UENCE OF SI NGL E HSDPA USER OCCUPYI NG HS- PDSCH
CODE RESOURCES ON R99
-2 0 2 4 6 8 10 12 14 16
0
50
100
150
200
250
Available Num of SF16 for HSDPA
A
v
a
i
l
a
b
l
e

N
u
m

o
f

S
F
2
5
6

f
o
r

R
9
9

Figure 13 indicates that a single HSDPA user consumes R99 code
resources in the case of multi-code concurrence, supposing that
this user has set up one HS-SCCH and one DPCCH. When the
system does not allocate the code resources to HSDPA services,
excluding the eight common channel code overhead whose SF is
256, there are 248 channelized codes whose SF = 16 for R99
services. More code tree resources whose SF is 16 are allocated to
this user, and the code resources for R99 decreases linearly. When
fifteen channelized codes whose SF is 16 are allocated to one
HSDPA user, the code tree resources of this cell are used up and
only five channelized codes whose SF is 256 are remained.
Seeing the allocation of the code resources, R99 and HSDPA are in
the linear relation in the code resource and they can interchange.
In the initial phase of WCDMA network construction, R99 voice
service and HSDPA data service do not have very strong
requirements, so the static configuration mode is applicable to
code resource allocation. Or, the foreground can perform the
dynamic adjustment according to the change of the cell data
throughput and user quantity.
Usually, there are two dynamic allocation methods for the
HS-PDSCH quantity. One is to estimate the HS-PDSCH quantity
according to the users requirements beforehand and adjust
according to OMC configuration. This method is called as periodical
adjustment. The other is to adjust the HS-PDSCH quantity
dynamically by the foreground algorithm according to the
proportion of the cell data throughput and voice services (co-CF
supports HS-DSCH) or the packet data service user quantity (only

HSDPA Technology

supporting HS-DSCH). This method is called as the data
throughput or user quantity adjustment.
Analysis on HSDPA Power Sharing
After the HSDPA function is introduced, two aspects of the power
control methods are added:
Allocation of HSDPA total power resources: Including HSDPA
operation power allocation and dynamic allocation.
The power control of HSDPA-related physical channels: The
power control method of HS-PDSCH, HS-SCCH, and
HS-DPCCH.
The location method of the HSDPA total power:
RNC fixed allocation method: calculate the average data
throughput in the corresponding areas beforehand. Estimate
the quantity of HS-PDSCH to configure and the corresponding
power resources (the code resource should match the power
resource). Configure the power percentage that HSDPA
occupies in OMC.
RNC dynamic allocation method: RNC allocates the initial
HS-PDSCH And HS-SCCH Total Power. When the system is
running, RNC updates the HS-PDSCH And HS-SCCH Total
Power dynamically because of the system load and the
periodically triggering.
Node B dynamic use: RNC does not allocate Node B with
HS-PDSCH And HS-SCCH Total Power. Node B allocates the
appropriate power to HSDPA according to the power that DPCH
occupies.
No matter which method is adopted to allocate the service power
between HSDPA and R99, the final facto to affect the network
traffic and user quantity is the power allocation proportion. In the
initial phase of the network construction, the typical service of R99
is CS12.2 k voice service and that of the HSDPA is PS domain data
services. The following will analyze the effect of HSDPA on R99
capacity from the power allocation aspect.


FI GURE 14 I NFL UENCE OF HSDPA POWER AL L OCATI ON ON R99 CAPACI TY
-13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3
30
40
50
60
70
80
90
100
R99 Capability Loss
R
9
9

U
s
e
r

N
u
m

P
e
r
c
e
n
t

%
Total HSDPA Power offset to BsTxPwer (dB)
HSDPA heavy load
HSDPA light load

Figure 14 indicates the influence that the different power allocated
to HSDPA has on the R99 capacity. Supposing that R99 capacity
uses the quantity of CS12.2 k voice users for the calculation, when
the HSDPA power allocation is 0, the RR capacity reaches the
maximum value, 100%. The more power is allocated to HSDPA,
the smaller R99 capacity will grow. In Figure 14, the blue curve
indicates the power resources preserved for HSDPA but the
system does not bear the HSDPA services. The red curve indicates
that the network bears the HSDPA services and the located power
are all used. The two curves indicate two extreme cases. The
space between the two curves indicates the R99 capacity that can
be kept after the HSDPA power is allocated. For example, when
the HSDPA power allocation offset is -5 dB, that is, the power
allocated to HSDPA service is 38 dBm (43 5 = 38), R99 capacity
can be kept at 60 ~ 85% of the maximum capacity. Thus, HSDPA
has the effect on R99 and this effect is related to HSDPA power.
In R99 network planning, it is usually planned to 50 ~ 75% of the
load. When R99 network reaches the maximum capacity, the BS
downlink power only occupies about 50%. That is, the remaining
50% BS power is preserved for the HSDPA service and R99
capacity will not decrease. However, if all the preserved power is
sent for HSDPA service bearer, R99 capacity will get lost due to the
co-CF interference.

HSDPA Technology

FI GURE 15 I NFL UENCE OF HSDPA POWER AL L OCATI ON ON R99 CAPACI TY
( R99 L OAD 50%)
-13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2
30
40
50
60
70
80
90
100
R99 Capability Loss
R
9
9

U
s
e
r

N
u
m

P
e
r
c
e
n
t

%
Total HSDPA Power offset to BsTxPwer (dB)
HSDPA heavy load
HSDPA light load

Figure 15 indicates the influence of HSDPA power resource
allocation proportion on R99 capacity when R99 network downlink
power load is at about 50%. R99 may contains more power
redundancies, so more power resources can be preserved for
HSDPA and R99 capacity will not decrease suddenly. In Figure 15,
the blue curve indicates when the preserved power resources are
used for the HSDPA service transmission, the co-CF interference
will occur on R99 users. The red curve shows the worse case that
R99 capacity gets lost. Same as Figure 14, the space between the
two curves indicates the R99 capacity loss that HSDPA services
may bring. It is related to HSDPA service power. The lower R99
load is planned in the initial planning, the less R99 capacity will
lose after HSDPA is introduced.
Influence of Channel Condition on HSDPA
Network Performance
Theatrically, different load of the surrounding cells and different
surroundings of the radio channels will have different influence on
the HSDPA service throughput. The coverage of the interference
factors in different throughput rate is as shown in Figure 16. The
coverage of the orthognal factors in different throughput rate is as
shown in Figure 17. When the HSDPA users are at the edge of the
cell, the external interference becomes larger. You can study the
system performances when the HSDPA users are at the edge of
the cell by observing the change of Ior/Ioc.


FI GURE 16 I NFL UENCE OF I NTERFERENCE FACTOR ON COVERAGE
PROPORTI ON I N SPECI FI C THROUGHPUT
0 10 20 30 40 50 60 70 80 90 100
0
1000
2000
3000
4000
5000
6000
C
e
l
l

T
h
r
o
u
g
h
p
u
t
/
(
k
b
i
t
/
s
)
CoverageRat e (% )
Ior/ Ioc = 3
Ior/ Ioc = 1
Ior/ Ioc = 0
Ior/ Ioc = -1
Ior/ Ioc = -3

In Figure 16 the five curves (from up down, they are blue, red,
green, light blue, pink) indicate the relation of the GHSDPA
coverage rate and its throughput in the case of different
interferences. When the radio surroundings are better, like the
blue curve (Ior/Ioc = 3 dB), the throughput can support the
whole-network coverage. When the radio surroundings are not
good, like the pink curve (Ior/Ioc = -3 dB), the throughput is not
within 0 ~ 100% of the effective coverage. Pay attention to the
near end, such as, 0 ~ 30%.
As shown in Figure 16, when the external interference that the
users meet is little, HSDPA can support larger throughput when
guaranteeing the certain coverage rate. To ensure the HSDPA
resources made full use of, enable the UE in the favorable place to
obtain the HSDPA resources through the admission control. For
the UE far away from the BS or in the channel whose surrounding
are bad, reduce the HSDPA information rate or switch to PS
service for bearer on other dedicate channel.

HSDPA Technology

FI GURE 17 I NFL UENCE OF ORTHOGNAL FACTOR ON COVERAGE PROPORTI ON
I N SPECI FI C THROUGHPUT
0 10 20 30 40 50 60 70 80 90 100
0
1000
2000
3000
4000
5000
6000
7000
C
e
l
l

T
h
r
o
u
g
h
p
u
t
/
(
k
b
i
t
/
s
)
CoverageRat e (%)
Ort hognalFac t or = 0. 4

In Figure 17, the five curves (from up down, they are blue, red,
green, light blue, pink) indicate that the influence of the orthognal
factors on the HSDPA is very obvious. Figure 17 indicates the
curve of changing the orthognal factor when Ior/Ioc = 1 dB.
In the above analysis, we have not given the absolute value of the
HSDPA coverage radius, because this value gives limited reference
when the channel surroundings, bearer services, and the antenna
parameters change. However, to see the coverage of the
low-speed HSDPA link, compare the emulate surroundings with
the city surroundings, and compare the BS spaces and site
distribution dense in the case of continuous coverage with R99
voice network.
From all above analysis, we can see that HSDPA has high
requirements for the channel condition. The HSDPA throughput
may change impetuously because of the change on the external
interference factor and orthognal factor. Therefore, when the
power allocation and the code resource allocation complete, the
advantages of the HSDPA performances can develop best through
the high-quality scheduling algorithm and switchover algorithm
during the system operation.
Switchover Policy
HSDPA uses the service cell to upgrade, that is, the hard
switchover. HS-PDSCH does not support the soft switchover, so
there are no switchover gains. The HSDPA users at the cell edge
can perform the hard switchover or switch from the CELL_DCH
(HS-PDSCH) state to CELL_DCH (DCH) state.


FI GURE 18 COMPARI SON OF HSDPA AND R99 TRAFFI C/ COVERAGE
0 10 20 30 40 50 60 70 80 90 100
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
D
L

C
a
p
a
b
i
l
i
t
y

(
k
b
p
s
)
distance/cell_radius %
R99 PS
HSDPA

In Figure 18 we can see the difference of R99 (384k) and HSDPA in
the coverage and throughput. The red real line indicates the R99
throughput. Because of the limited code resources, when UE is
close to the BS, R99 throughput reaches the upper limit (7 384 K)
and the throughout will not go up more. The red broken line
indicates the R99 throughput without considering the limited code
resources.
In Figure 18, at the near end, HSDPA throughput is far better than
R99 throughput. In the place of about 45% radius, HSDPA
performance decreases to the level of R99. When the performance
reaches 70%, it enters the R99 switchover area, so R99 obtains
the switchover gains and the throughput gets the compensation.
Therefore, the system can hold at a stable level. HSDPA does not
have the switchover, so its throughput decreases sharp at the far
end.
A reasonable switchover policy can solve this problem. At the near
end, users use HSDPA to increase the user data traffic. When the
user moves away from the BS, switch the HSDPA service into DCH
PS services. When the user moves to HSDPA service coverage
area of another BS, switch the DCH PS service into HS-DSCH and
switch back to HSDPA service. With the new switchover policy of
HSDPA, the limited resources can be made best use of and the
advantaged of HSDPA performance can be developed. The new
switchover policy is the supplementary and extension of the
previous R99 switchover policy, which bring as little performance
loss to R99 network as possible.

HSDPA Technology

Influence of HSDPA Introduction on
Transmission Network
To introduce the HSDPA into WCDMA network construction in the
initial phase, the requirements for the transmission resources of
Iub interface increase a lot. In most places, the idle resources of
the previous 2G network can not satisfy the requirements. The
idea that 3G transmission borrows 2G transmission in Phase I and
2G transmission borrows 3G transmission in Phase II can not be
set up. The transmission resources at Iub interface must be
re-planned, for the transmission expansion or re-construction.
No matter in the case of R99 co-CF or hetero-CF, some
transmission resources should be preserved. HSDPA asks for more
transmission resources, because Node B capacity increases and
the user traffic model changes after the HSDPA is introduced.
Cell throughput:
R99 cell peak rate: 7 384 Kbps = 2.688 Mbps
HSDPA cell peak rate: 15 960 Kbps = 14.4 Mbps
After the HSDPA is introduced, the cell peak throughput increases
to five times of the previous throughput.
HSDPA cell peak rate of 3-carrier BS:
14.4 3 1.3 80% = 43.68 M
In the formula, 1.3 is the factor after the transmission protocol
overhead and 80% is the system load.
The transmission bandwidth of Iub interface:
Considering about the traffic model, the quantity of the numbers
distributed in the cell, the quantity of the Node B cells, and the
traffic model are important in calculating the transmission
bandwidth.
A brief formula to calculate the Iub interface transmission
bandwidth:
Iub interface bandwidth (uni-directional) = Ncell (Nuser Ev
Vv R + Nuser Ecs Vcs + Nuser Vps) (1 + Q)/{1/Y}
Thereinto:
Ncell is the quantity of the cells.
Nuser is the quantity of the numbers distributed in the cell.
Ev is the quantity of voice Erl of each user.
Vv is the voice rate and R is the voice active factor.
Ecs is the quantity of the video phone Erl of each user.
Vcs is the rate of the video phones.
Vps is the average data rate of each user.


Q is the A transmission overhead.
Y is the load factor.
Conclusion:
After the HSDPA is introduced, the cell peak throughput increases
to five times of the previous throughput and the average
throughput increases to twice. Considering about the peak rate
and the traffic model, actual Iub transmission resources should
take the value between theses two. In Phase I, it is preferred to
preserve about five times of the transmission resources and
preserve at least twice.
Lastly, with the increase of HSDPA users and the decrease of the
charge, more users will change the habits of using the service
(including the voice services). Therefore, the traffic model will
change and the transmission resources at all interfaces will be
affected. Such effects we can not estimate now, so preserve
abundant capacity when planning the transmission network and
keep the well expansible capacity.
HSDPA Network Upgrade
Evolution Policy
Equipment Upgrade Influence and
Requirements
In 3GPP specifications, the introduction of the HSDPA is to add
MAC-hs entity on the MAC layer. The MAC-hs entity locates on
Node B, responsible for the HARQ operation and the corresponding
scheduling. This change has the following effects on the network
equipment.
The following software functions are added onto Node B:
HSDPA MAC functional module
HSDPA FP functional module
HSDPA control functional module
HSDPA downlink processing module
HSDPA uplink processing module
The following software functions are added onto RNC:
Modules related to HSDPA radio resource management
HSDPA control plane processing module
HSDPA user plane processing module

HSDPA Technology

Therefore, the equipment in the newly-constructed network
should be prepared for updating to the HSDPA at any time. If
the equipment hardware supports the smooth upgrade, the
investment will not increase.
After the HSDPA is introduces, the capacity requirements will
increase. BS should support over two CFs and the
mainstreaming BS should be expanded to support three CFs.
HSDPA must consume the BS transmission power and channel
processing resources. In Phase I, HSDPA services cover the
hot spots. It shares the CF with R99 and then use the
independent CF later. The following aspects should be taken
into consideration for sharing the power amplifier resources
and processing resources.
Sharing the channel processing board resources:
In one BS, the baseband resources are shared between
different sectors of same CS, to increase the system soft
capacity.
When using the baseband pool + RF remote, the baseband
resources are shared between different BSs in the unit of
sector.
The macro cell can share the baseband pool with the RRU
attached to the macro cell.
Sharing the power amplifier resources:
Perform the multi-carrier power amplifier sharing in each
sector
Perform the multi-sector power amplifier sharing with the
power amplifier resource pool
The first method is to perform the multi-carrier combination in the
digital intermediate frequency and to use only one transmit link to
deal with the multi-carrier. Several carriers input the power
amplifiers at the same time and share the power resources of the
power amplifier. The total power of all carriers is the maximum
rated power of the power amplifier. The power of each carrier can
be set within this range randomly.
The second method is to use all the power amplifiers of the three
sectors as one logic power amplifier. All CFs allocate the power to
each power amplifier evenly through the power amplifier resource
pool, so the power is equal on any power amplifier and the
resources on any power amplifier can be taken full use of. This
method brings some negative effects as well. The power amplifier
resource pool uses the emulate apparatus, so it has larger effect
on the carrier performance due to the large performance
discretion, unbalanced range, and unbalanced phase. As a result,
the EVM degrades because the due to the orthogonal of each CF
goes bad. The interference between sectors increases. The
performance of the whole network is affected.
Lastly, the introduction of HSDPA makes the 3G network more
attractive to the users and the number of users increases rapidly.


Thus, the requirements of RNC/CN capacity and integration
increase too. RNC/CN should support the smooth expansion and
the effect on the network should be decreased to the best.
Network Evolution Policy
In Phase I of WCDMA network construction, R99 and R4 has no
obvious differences at the radio side. With the increasing
requirements of the users for the high-speed data services and the
maturity of HSDPA technology, the radio network evolutes to
HSDPA naturally.
HSDPA technology evolutes too, so the following policies are
applicable for the HSDPA introduction and evolution:
1. In Phase I, the HSDPA is introduced. Most data users are
indoors and indoor data services consume a lot of outdoor BS
power. Therefore, set up the HSDPA coverage indoors first and
then connect all HSDPA coverage spots into pieces with the
outdoor macro BS. Finally, all the city is covered.
2. The hardware has existed and the investment on the software
upgrade is less. Therefore, in Phase II, medium/large cities
and traffic trunks can be covered with HSDPA.
3. HSDPA evolutes phase by phase. It takes time for the network
equipment and terminal to improve their supports of HSDPA.
According to the requirements of the high-speed data services
and operation, HSDPA evolution falls into two phases:
Phase I: The network side supports ten code bits and the
cell downlink throughput is 7.2 Mbps. In Phase I, HSDPA
services concentrate in the isolated islands, such as, indoor
and hot spots. However, the cell traffic can reach the
higher throughput. In Phase I, the terminal supports five
code bits. The terminal develops faster than the network,
so the frequent upgrading of the network should be taken
into consideration. In Phase I, the network can satisfy the
requirements of the operation and services within some
time, which is more competitive compared with other
technologies.
Phase II: The network side supports fifteen code bits and
the cell downlink throughput is 10.8 Mbps. In Phase II, the
network can satisfy the requirements of the data service
hot spots where the traffic is increasing further.

HSUPA Technology
ZTE CORPORATION
P. R. China
518057
Tel: (86) 755 26771900 800-9830-9830
Fax: (86) 755 26772236

Contents

Chapter 1 .......................................................................... 1
HSUPA Overview .............................................................. 1
General ......................................................................... 1
Protocol Architecture ....................................................... 2
Key Technologies of the HSUPA ......................................... 3
Transmission Time Interval (TTI) ............................................ 3
HARQ Multiprocess Processing ................................................ 4
Scheduling of Node B Control.................................................. 5
Channel Structure of HSUPA ............................................. 9

Ch a p t e r 1
HSUPA Overview

Key points
Basic concepts of HSUPA
Changes of protocol architecture
Key technologies of HSUPA
Comparison between HSUPA and HSDPA
New channels introduced by HSUPA
General
The Release 6 of the 3GPP standards focuses on the stipulation
of HSUPA standards. High Speed Uplink Packet Access (HSUPA,
or more formally, E-DCH or EUL, enhanced uplink technology) is
a type of enhanced uplink transmission technology. Compared
with the DCH of Release 99 of the 3GPP standards, the HSUPA is
primarily used to enhance the data transmission rate of the
uplink packet switched domain by means of multi-code
transmission, spreading factor 2 or 4, HARQ technology, fast
data scheduling and 10ms TTI / 2ms TTI. Theoretically speaking,
the uplink peak transmission rate can reach 5.76Mbit/s.
Tip:
The standard name of HSUPA in the current protocol is
Enhanced Uplink. However, to simplify the description, this
document still uses the term HSUPA in contrast to the term
HSDPA.
By putting the scheduler in the Node B, the HSUPA greatly
reduces the delay of both scheduling control signaling and UE
response so that it can control cell load in a faster, more
accurate and effective way, keeping the cell load always close to
the preset load threshold. Together with other key technologies
such as the HARQ and shorter TTI, the HSUPA enables the UE to
allocate as much power as possible to the E-DCH, thus

WCDMA-RNS HSUPA Overview

increasing the uplink throughput and fully utilizing the limited
power and bandwidth resources.
As enhanced uplink technology of the WCDMA system, the
HSUPA technology is applicable to wireless games, interactive
services, streaming media-based video services, data upload
services, and other services.
Table 1 shows the comparison between the HSUPA and the
HSDPA.
TAB L E 1 COMPARI SON B ETWEEN THE HSUPA AND THE HSDPA
HSUPA HSDPA
Objective
Both are introduced so as to optimize and
evolve the packet switched services.
Channel
Uplink dedicated
channel
Downlink shared
channel
Key technologies
HARQ, uplink
SoftSwitch, NODE B
scheduling and
power control
HARQ, NODE B
scheduling, AMC
technology and high
order modulation
Rate 5.76Mbps 10.8Mbps
TTI 2ms, 10ms 2ms
Protocol Architecture
The protocol architecture of HSUPA is as shown in Figure 1

Chapter 1 HSUPA Overview

FI GURE 1 PROTOCOL ARCHITECTURE OF HSUPA

PHY PHY
EDCH FP EDCH FP
Iub UE NodeB
Uu
DCCH DTCH
TNL TNL
DTCH DCCH
MAC-e
SRNC
MAC-d
MAC-e
MAC-d
MAC-es /
MAC-e
MAC-es
Iur
TNL TNL
DRNC

The changes of protocol architecture are as follows:
UE: A new MAC entity, MAC-es/MAC-e, is introduced and
placed right below MAC-d. It is responsible for HARQ
retransmission, scheduling, MAC-e multiplexing and E-TFC
selection.
Node B: A new MAC entity, MAC-e is introduced to take care
of HARQ retransmission, scheduling and MAC-e
demultiplexing.
SRNC: A new MAC entity, MAC-es is introduced to take care
of re-sequencing and macro diversity combination.
Key Technologies of the
HSUPA
Compared with the Release 99 of the 3GPP standards, the
HSUPA brings in multi-code transmission, spreading factor 2 or 4,
HARQ technology, fast data scheduling and 10ms TTI / 2ms TTI,
and moves the scheduler from the RNC to the Node B so as to
realize E-DCH scheduling of MAC-e protocol control in the Node
B.
The following describes the TTI, the HARQ multiprocess and the
fast scheduling of Node B in detail.
Transmission Time Interval (TTI)
An optional TTI of 2ms is introduced to the HSUPA uplink.
The reason to introduce the 2ms TTI to the HSDPA downlink is
that there is no power control on HS-PDSCH transmission.


However, having no power control does not present any problem
for the uplink, because the uplink with no power control will
reduce the system capacity. The objective to adopt short TTI on
the uplink is to reduce the HARQ retransmission delay. The 2ms
TTI faces the uplink coverage problem due to limited power
resources in the terminal. When each TTI contains data of the
same volume, the power transmitted within 2ms may be less
than that transmitted within 10ms. Apart from that, when the
TTI is reduced to 2ms, the interleaving gain is also reduced.
Therefore, 10ms TTI is adopted in the standard so as to ensure
normal operation at the cell edge.
HSUPA UE types are as shown in Figure 2
FI GURE 2 HSUPA UE TYPES

As stipulated in the UE Category of 3GPP, 10ms TTI is
mandatory for all types of applications, while 2ms TTI is optional
for some applications. Of course, 2ms TTI can increase the
system capacity when there is no other restrictions like link
coverage, and can even result in higher peak rate in fine
wireless environments.
HARQ Multiprocess Processing
The Hybrid Automatic Repeat reQuest (HARQ) is a type of
technology combining Forward Error Correction (FEC) and
retransmission. It can rapidly adjust the channel transmission
rate and combine the FEC and retransmission based on the link
status. The physical layer HARQ is controlled by higher layers.
Similar to the HSDPA, 3GPP introduces the HARQ to the HSUPA
to improve system performance by reducing delay and
increasing data retransmission efficiency. The HARQ can flexibly
adjust the rate of active elements and compensate for the error
codes brought by link adaptation. The HARQ of HSUPA is
implemented on the MAC-e and physical layer of the Node B.
Similar to the HSDPA HARQ, the N-channel SAW HARQ
mechanism is adopted for all the HSUPA HARQ multiprocesses,
as shown in Figure 3.


FI GURE 3 HARQ MULTIPROCESS PROCESSING
HSPDSCH
1
UE
1
Packet1UE
1
packet2
Node B
1
UE
1
Packet3UE
1
Packet4UE
2
Packet1UE
1
Packet5UE
1
packet2UE
1
Packet6UE
1
Packet4
UL DPCCH UE
1
UL DPCCH UE
2
A N A N A A A A A
A
UE
1
Packet7UE
2
Packet2
UE1HARQ channel 1 UE1HARQ channel 2 UE1HARQ channel 3 UE1HARQ channel 4
UE2HARQ channel 1

The HARQ has two operation modes: Chase Combine (CC) or
soft combining and Incremental Redundancy (IR). In CC mode,
the retransmitted data are the same as those initially
transmitted. In IR mode, the retransmitted data differ from the
data initially transmitted (for example, only the check bits are
retransmitted). The IR mode has better performance than the
CC mode. However, it requires more memories at the terminal
side. The default memory capacity of the terminal is designed
according to both the maximum data rate supported by the
terminal and the soft combining mode. Therefore, only soft
combining mode can be used when data transmission is at the
maximum data rate. However, both modes can be used in the
case of low data transmission rates. The IR mode makes the UE
memory more complex. However, 3GPP does not have any
restrictions on the modes. The CC mode can be viewed as a
special form of the IR mode.
Scheduling of Node B Control
The basic scheduling structure of HSUPA Node B control is as
shown in Figure 4.


FI GURE 4 BASI C SCHEDUL I NG STRUCTURE OF HSUPA NODE B CONTROL

As shown in Figure 4, the UE first sends to the serving Node B a
scheduling request asking the serving Node B to allocate
resources. This request contains scheduling information (inband
signaling borne by the E-DPDCH, including a 4-bit high priority
logical channel ID, a 9-bit UE cache utilization status (a 5-bit
Total E-DCH Buffer Status, or simply, TEBS and a 4-bit Highest
priority Logical channel Buffer Status, or simply, HLBS) and a
5-bit UE power status) as well as the Happy Bit borne by the
E-DPCCH. The serving Node B then decides the scheduling grant
according to the QoS information and the scheduling request of
the UE. The scheduling grant has the following properties: It is
only applicable to the E-DCH TFC but not the DCH TFC, and it
controls the maximum E-DPDCH/DPCCH power ratio allowed for
the activated process. For a non-activated process, the power
ratio is 0, and the UE is prohibited from transmitting power. All
the grants are definitive. And the scheduling grant can be
transmitted at each TTI or at a lower frequency.
The Node B sends two types of scheduling grants, namely,
Absolute Grant and Relative Grant. The Absolute Grant gives an
absolute limit on the maximum UL resources that can be used
by the UE, while the Relative Grant can increase or decrease the
granted value used last time. The Absolute Grant is sent by the
serving cell of the serving E-DCH and can be effective for a UE, a
group of UEs or all UEs. It contains the E-RNTI (16-bit mask
processing) of the UE (or the UE group) of the grant, a 5-bit
maximum power ratio that can be used by the UE, and a 1-bit
HARQ process activation identifier. Therefore, the Absolute
Grant information consists of 6 bits, namely, xag, 1, xag, 2, ...,
xag, 6 and it can be sent via the E-DCH Absolute Grant Channel
(E-AGCH). The Relative Grant (update) can be sent by the
serving or non-serving Node B as a supplement to the Absolute
Grant. The Relative Grant from the Serving E-DCH RLS can have
one of the following three values: UP, DOWN or HOLD,
while the Relative Grant from the non-serving E-DCH RL can
have either of the following values: HOLD or DOWN. The


HOLD command is sent via Discontinuous Transmission (DTX),
while the DOWN command corresponds to overload indicator.
To sum up, the scheduling of Node B control can fast control the
Raise Over Thermal (RoT), and the granted value is primarily the
power or the bit rate.
To transmit data, the scheduled UE can select the
Retransmission Sequence Number (RSN) including the HARQ RV
version selection and a proper power offset of the physical layer
E-DPDCH and E-DPCCH according to the received scheduling
information that combines the Absolute Grant and the Relative
Grant as well as the ACK/NACK message sent previously.
In view of the entire UTRAN protocol, the basic functional blocks
of the HSUPA are as shown in Figure 5
FI GURE 5 OVERALL STRUCTURE OF HSUPA

Figure 5 shows both the connection between the URTAN and a
UE that uses the E-DCH and that is in the SoftSwitch status and
the protocol entities related to the HSUPA at the UE side and the


network side. It also shows the basic concept and operating
mechanism of the HSUPA.
E-DCH active set: The set of cells that have the E-DCH
bearer with the UE. It can be a sub set of the DCH active set.
E-DCH serving cell: The cell from which the UE receives
Absolute Grants. The UE only has one E-DCH serving cell.
E-DCH serving RLS: A set of RLs of the E-DCH serving cell,
generally, a group of E-DCH active set cells in the Node B
where the E-DCH serving cell is.
Non-Serving E-DCH RLS: A set of E-DCH cells of all
non-E-DCH serving RLSs.
The key feature of the HSUPA is Node B control scheduling. Its
process is as follows:
The UE has an E-DCH serving cell whose Node B is in charge
of the E-DCH scheduling. The E-DCH serving cell sends the
scheduling command, that is, Absolute Grant to the UE via
the downlink E-AGCH. The Absolute Grant specifies the
absolute value of the maximum resources available to the UE.
It contains information such as the E-RNTI of the UE and the
maximum transmit power allowed for the UE.
The E-DCH serving cell and the non-E-DCH serving cell send
the Relative Grant to the UE via the downlink E-RGCH. The
Relative Grant is an offset (or fine tuning) of the Absolute
Grant. It can be UP, HOLD or DOWN. UP means
tuning up, DOWN means tuning down, and HOLD means
no change. Only the Serving E-DCH RLS can send the UP
Relative Grant. The Non-Serving E-DCH RLS can send the
HOLD or DOWN Relative Grant only. Normally, the
Non-Serving E-DCH RLS sends the DOWN Relative Grant
because of uplink overload.
The UE selects the E-TFC based on the grant information
received. It sends data (including the retransmitted data) via
the E-DPDCH and the E-TFC information, the HARQ RV
information (RSN) and a Happy Bit via the E-DPCCH. The
Happy bit is used to advise the Node B whether the UE is
satisfied with the resources (grant) currently allocated, that
is, whether the UE requires higher grant.
The MRC combination is conducted on the E-DCH data
received by various cells in the same Node B of the E-DCH
Set and then the data are sent to the Mac-e for processing.
There is a Mac-e in each Node B for each UE. It
demultiplexes the Mac-e PDU into MAC-es PDU to send to the
RNC. It also sends the E-DCH scheduling information and the
HARQ ACK/NACK message.
There is a Mac-es entity in the SRNC for each UE. The
Mac-es entity conducts macro diversity combination of
MAC-es PDUs of different Node Bs, and then re-sequences
and splits them into Mac-d PDUs to send to the Mac-d.


The HSUPA uses the Scheduling Grant on the downlink to
indicate the maximum uplink resources available to the UE. The
Node B determines the grant scheduling according to the cell
load, the service QoS and the UE scheduling request. On the
uplink, the UE determines the service grant to be used for the
transmission according to the grant scheduling of the Node B
and the current remaining power.
Channel Structure of HSUPA
To suit the above operating mechanism, a new uplink transport
channel, that is, Enhanced Dedicated Channel (E-DCH) is
introduced to the HSUPA to bear the user data. The
corresponding physical channel is E-DCH Dedicated Physical
Data Channel (E-DPDCH), and the SF ranges from 2 to 64.
To bear the control information associated with the E-DCH, the
E-DCH Dedicated Physical Control Channel (E-DPCCH) with SF
being 256 is introduced to the HSUPA.
To bear the E-DCH absolute grant, the E-DCH Absolute Grant
Channel (E-AGCH) with SF being 256 is introduced to the HSUPA.
To bear the E-DCH relative grant, the E-DCH Relative Grant
Channel (E-RGCH) with SF being 128 is introduced. And to bear
the E-DCH HARQ acknowledgement indicator of the HARQ, the
E-DCH HARQ Acknowledgement Indicator Channel (E-HICH) with
SF being 128 is introduced.
The channel structure and timing relation of HSUPA is as shown
in Figure 6.


FI GURE 6 CHANNEL STRUCTURE AND TI MI NG REL ATI ON OF HSUPA


AG Absolute Grant
E-AGCH E-DCH Absolute Grant ChannelE-DCH
E-DCH Enhanced Dedicated Channel
E-DPCCH E-DCH Dedicated Physical Control Channel E-DCH

E-DPDCH E-DCH Dedicated Physical Data ChannelE-DCH

E-HICH E-DCH HARQ Acknowledgement Indicator Channel
E-DCH HARQ
E-RGCH E-DCH Relative Grant ChannelE-DCH
E-RNTI E-DCH Radio Network Temporary IdentifierE-DCH

E-TFC E-DCH Transport Format CombinationE-DCH

HARQ Hybrid Automatic Repeat Request
HSUPA High Speed Uplink Packet Access
RG Relative Grant
RLS Radio Link Set
RSN Retransmission Sequence Number
SG Serving Grant
TSN Transmission Sequence Number
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
II
III
IV
VI
VII
VIII
IX
XI
XII
XIII
XIV
XV
XVI
XVII
XVIII
l
l
II
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
II
III
IV
VI
VII
VIII
IX
XI
XII
XIII
XV
XVI
XVII
XVIII
XIX

TJCWR1207067E0 WCDMA RNS Product Training Manual I

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

TJCWR1207067E0 WCDMA RNS Product Training Manual I

Uploaded by

Copyright:

Available Formats

WCDMA RNS Product Training Manual

You might also like