HSDPA Overview

15 Code 16QAM/64QAM TTI = 2 ms Hybrid ARQ Fast Link Advanced

Shared Modulation with incr. redundancy Adaptation Scheduling
transmission

Benefit
Higher Downlink Peak rates: 21 Mbps. Can be 42 Mbps using DC HSDPA feature
Higher Capacity: +100-200%
Reduced Latency: ~75 ms
HSUPA Overview

Fast NodeB
1-4 Code TTI = 2 / 10 ms Hybrid ARQ
Power Control ControlledSc
Multi-Code with incr. redundancy
heduling
transmission

Benefit
Higher Uplink Peak rates: 2~5.76 Mbps
Higher Capacity: +50-100%
Reduced Latency: ~50-75 ms
HSxPA Motivation and General Principle
• Improved performance and spectral efficiency in DL and UL by introducing a shared channel principle:
• Significant enhancement with peak rates up to 42 Mbps (DC HSDPA Feature) in DL, and 2 Mbps (11.5 Mbps
with 16QAM) in UL
• Huge capacity increase per site; no site pre-planning necessary
• Improved end user experience: reduced delay/latency, high response time

Rel. 99
Dedicated pipe for every UE

HSUPA (3GPP Rel6)

Dedicated pipe for every UE in UL

Pipe (codes and grants) changing
with time
E-DCH scheduling
HSDPA (3GPP Rel5)
Fast pipe is shared among UEs
 UL DCH vs HSDPA vs HSUPA Concepts
HSDPA HSUPA
Efficient UE power
BPSK and Dual- amplifier
Modulation QPSK/16-QAM/64QAM
BPSK
Soft handover No Yes
Required for near-far
Fast power avoidance
No Yes
control
Point to Multipoint Scheduling cannot be as
Scheduling
multipoint to point fast as in HSDPA
Non-scheduled Yes, for minimum/
No
transmission guaranteed bit rate Similar to R99 DCH but
with HARQ
HSUPA could be better described as Enhanced DCH in the uplink

HSUPA (E-DCH) is an uplink DCH with BTS-based HARQ and scheduling and true multicode support

Feature DCH HSUPA HSDPA

Variable spreading factor Yes Yes No

Multicode transmission Yes Yes Yes

(No in practice)
Fast power control Yes Yes No

Soft handover Yes Yes No

(associated DCH only)
Adaptive modulation No No Yes

BTS based scheduling No Yes Yes

Fast L1 HARQ No Yes Yes

Hybrid – Automatic Repeat ReQuest Concept
• ARQ in 3GPP R99:
• Re-transmission process when UE will send a “Request” for “Retransmission”, if it cannot decode error data on first transmission
• Acknowledgment in “RLC Layer” is terminated at RNC
• The “Retransmitted data” is the copy of “First Transmitted data”
• UE will discard the “Error information” that it cannot decode to have space to receive “Retransmitted data”, and will decode only the “Retransmitted data”

• Limitation:
• Slow acknowledgement in RNC not suitable for High-speed communication like HSxPA
• There is no guarantee that “Retransmitted data” can be decoded

• Improvement of H-ARQ:
• Physical layer (Layer 1) acknowledgements in NodeB are used for retransmission decisions (Low delay)
• Provides gain by soft combining from the original and subsequent packets prior to the decoding

Procedure of H-ARQ:
• “UE category” information from “UL control channel” is used
• Combining type according to “UE capability” is selected
• “Retransmission Scheme” is sent to UE via “DL control channel” (Includes “Type of combining” , ”First or Retransmitted data”)
• UE use sinformation as a “KEY” to handle the data. “New data”  UE will discard the buffer and decode the “New data”. “Retransmitted data”  UE will
combine this data with the “first transmitted data” stored in its buffer before decoding. The combination method will be selected according to combining
type
• UE responses decoding result to NodeB via “UL control channel”. If decoding is successful – “ACK” response will be sent, NodeB will transmit the next
packet. If not – “NACK” response will be sent, NodeB will transmit the retransmitted packet
H-ARQ • Chase combining:
Incremental Redundancy

• Re-transmitted blocks are identical to the original block

• An erroneous block is not discarded, but is combined with
subsequent re-transmissions
• Low complexity at the UE

• Incremental redundancy:
• Each retransmission contains different redundancy
version from the first transmitted data (constructed with
lower coding rates with more redundancy)
• Less re-transmission amount due to more redundancy bit
• High complexity at the UE
DL
send data UE
Rel 99 RNC RLC ACK/NACK
re-transmission NB
DCH UL
send data
RLC ACK/NACK
re-transmission

DL
send data
Rel 6 RNC L1 ACK/NACK
DL: HS-DSCH
HS-DSCH UL: EDCH NB re-transmission UE
E-DCH UL send data
L1 ACK/NACK

re-transmission
HSDPA
HSDPA – New Channels for HSDPA per UE
NB • High Speed Dedicated Physical Control
HS-DPCCH
HS-DPCCH Channel
HS-DPCCH
• SF=256
HS-DPCCH • Carries H-ARQ ACK/NACK, Channel Quality
Information (CQI)
• It can only exist with UL DPCH

Uplink

Down
link High Speed Physical Downlink Shared Channel
– Data bearer: Peak data rate 21 Mbit/s
– QPSK, 16 QAM and 64QAM can be applied
HS-PDSCH
– SF=16
– Up to 15, always associated with a DCH
HS-PDSCH
NB HS-PDSCH
HS-PDSCH

High Speed Shared Control Channel

– SF=128
HS-SCCH – Carries H-ARQ information, channelization code set,
modulation scheme. Up to 4 logical channels per UE
HSDPA – Protocol Stack
PS-RAB PS-RAB

PDCP PDCP Iu-UP Iu-UP

RLC RLC GTP-U GTP-U

MAC-d MAC-d UDP UDP

HS-DSCH HS-DSCH HS-DSCH HS-DSCH
-FP -FP -FP -FP
IP IP
MAC-hs MAC-hs
AAL2 AAL2 AAL2 AAL2 AAL5 AAL5

ATM ATM ATM ATM ATM ATM

PHY PHY
PHY PHY PHY PHY PHY PHY

UE Uu Node B Iub CRNC Iur SRNC Iu SGSN

• HS-DSCH is terminated by MAC-hs in the Node B in contrast to all other UTRAN Transport Channels (terminated in
the RNC)
HSDPA – Adaptive Modulation and Coding
• Concept
• Process to adapt Modulation and Coding Scheme to be used in each transmission time interval
• Ensure “Maximum data speed” possible in each radio quality and UE capability
• Procedure
• Information of CQI, UE category from “UL Control Channel” is used
• Modulation Scheme according to CQI and UE is chosen: QPSK ,16QAM or 64QAM
• Coding scheme according to CQI and UE is selected:
• Number of Channelization Codes (1 – 15 Multiple codes)
• Code rate ( ¼ , ¾)
• “Modulation & Coding scheme” information is sent to the UE via “DL Control Channel”
• UE uses that information as a “KEY” to decode HSDPA Data Channel
Higher order Modulation will be used (QPSK => 64 QAM)

Use of higher order modulation and high-rate error-correction coding for good radio
channel conditions
 Improved user data rate, higher overall throughput
HSDPA – Scheduling Concept
• General Concept:
• Scheduling is the process of distributing resources to each UE in the system

R99 Scheduling:
• Process when RNC assigns “Transport CH” or “Data pipe” to each UE
• Process on Call establishment
• Speed of data link depends on UE request, Radio resource and service requirement
• Target is to maintain the speed of link from call establishment to call release by power control

• Fast scheduling Concept:

• Process when “Node B” assigns “Shared” resource to each UE in every TTI (2ms)

Not only UE scheduling but includes:

• “Fast link adaptation” to control speed of link by AMC
• “Control retransmission process” by H-ARQ

“3 processes to dynamically assign the shared radio resources to all user in the most effective way”
HSDPA – Scheduling Procedure
• Collect information
• From UL Control Channel (CQI, UE category)
• From NodeB (Queue information, Buffer status)
• Take into account by “Scheduling Algorithm”
• Choose “Characteristic of HSDPA DL Data Channel”
• Modulation and Coding scheme to be used
• Retransmission scheme to be used
• Send the “Characteristic of HSDPA DL Data Channel” information to UE via DL HSDPA DL Control Channel, follow with HSDPA
DL Data channel
• Target UE uses information in DL control channel as a “Key” to decode information

UE Scheduling Algorithms
Round Robin
• Use only “Queue information” to select UE in round robin basis
• Most fair algorithm but provides lowest cell throughput
• All of the UEs in the cell will receive the same possibility to get the resource
Max CIR
• Use only “CQI information” to serve best radio condition UE
• High peak data rates and maximum NodeB cell throughput
• Limited fairness of selecting and serving HSDPA users
Proportional Fair
• Use many factors to consider in the algorithm
• Good balance between throughput maximization and fairness for all HSDPA users in cell
Round Robin Scheduler
• Objective of the round robin scheduler is to
• assign resources on an equal time basis, i.e. different throughputs if UE experience different channel
conditions
• A round robin scheduler assigns the time slots in rotation without accounting for
the channel conditions experienced by each UE
• If a connection does not have data to transfer from the Node B then the time slot is
assigned to the next connection
• A proportional fair scheduler accounts for the channel conditions and assigns an
increased proportion of time slots to UE experiencing good channel conditions. A
proportional fair scheduler increases the overall cell throughput

Bad CQI
Good CQI
UE1 UE5 TTI UE No. Data volume
1 UE1 1.8 Mbyte
2 UE2 1.8 Mbyte
UE2 3 UE3 1.8 Mbyte
16 QAM
QPSK 4 UE4 1.8 Mbyte
5 UE5 3.6 Mbyte
UE3 6 UE1 1.8 Mbyte
Total 12.6 Mbyte

UE4
Based on assumption: all UEs are UE category 6 and use 5 code at full coding rate
Proportional Fair Scheduler
• Provides a multi-user diversity gain of 30%-60% in terms of improved HSDPA cell
throughput for low to medium UE speeds, depending on the environment
• Improved end user experience is acquired by increased HSDPA cell throughput
• R-FR algorithm is used for the MAC-hs packet scheduling in the BTS
• It utilises the radio channel state information from the UEs in its scheduling decisions
• The relative instantaneous channel quality is calculated in every 2ms TTI
HSDPA - UE Categories
• QPSK till 64 QAM modulation with multi code transmission used to achieve high data
rates
• 12 different UE categories defined, categories are characterized by
• Number of parallel codes supported
• Minimum inter-TTI interval
• Theoretical peak bit rate up to 21 Mbps for category 13 onward UE using 15 codes and
64QAM
 HS-SCCH
HS-SCCH

HSDPA Code Multiplexing HS-SCCH

• With Code Multiplexing, maximum of 15 UEs

can be scheduled during one TTI from single HS-PDSCH
cell HS-PDSCH
HS-PDSCH
• Multiple HS-SCCH channels (max 15) HS-PDSCH
• One for each simultaneously receiving HS-PDSCH
UE HS-PDSCH
HS-PDSCH
• Available HS-PDSCH codes and HS-PDSCH HS-PDSCH
power of cell are divided between UEs HS-PDSCH
HS-PDSCH
• HS-PDSCH codes actually used depends on HS-PDSCH
the channel conditions of a UE HS-PDSCH
HS-PDSCH
• Important when cell supports more codes HS-PDSCH
than UEs do
HS-PDSCH
• Cell supports 15 HS-PDSCH codes, Cat6
and Cat8 UEs => 3 users can be
scheduled on TTI cat 8 cat 6 cat 6 cat 6 cat 6

• BTS must also be capable of 10/15 codes in

order to dynamically adjust HS-PDSCH codes
Maximum code allocation for HSDPA
SF=1

SF=2

SF=4

SF=8

SF=16 15 HS-PDSCH codes Case1:

SF=32 Used by 2 HSDPA UEs  no

SF256 available for the 3rd UE for
Codes for common associated DCH
SF=64 channels in the cell Up to 15 HS-SCCH codes

SF=128 Case2:

Used by AMR user  only one

SF=256 SF128 code remains for
Codes for associated DCHs and associated DCH
non-HSDPA users
Case1+2:

Used by HSDPA UE as associated DCH and HS-SCCH

HSUPA
HSUPA – New Channels for HSUPA per UE
• E-DPDCH (E-DCH Dedicated Physical Data Channel)
• Data channel, for transmitting MAC-PDUs:
• Total E-DCH Buffer Status (TEBS) NB
• UE Power Headroom (UPH)
• E-DPCCH (E-DCH Dedicated Physical Control Channel)
• Transmitted in parallel with E-DPDCH – contains L1 control information for HARQ
and scheduling:
• E-DCH Transport Format Combination Indicator (E-TFCI)
• Retransmission Sequence Number (RSN)  control HARQ
• Happy Bit  UE could use more resources or not Uplink

E-AGCH (E-DCH Absolute Grant Channel) Down

– Time multiplexed shared channel using explicit addressing (E-RNTI) of UEs link
– Primary E-RNTI for single UE
– Secondary E-RNTI for group of UEs
– Grants E-DPDCH/DPCCH power ratio
E-RGCH (E-DCH Relative Grant Channel) – (Common or dedicated) NB
– Implicit addressing by assignment of physical channel (DL code + signature)
– Common channel by assignment of same physical channel to UEs for E-RGCH
purposes
– Up, Down, Hold
E-HICH (E-DCH Hybrid ARQ Indication Channel) E-RNTI: E-DCH RNTI
• Used for signalling ACK/NACK (Dedicated) RNTI: Radio Network Temporary Identity
HSUPA – Protocol Stack
DTCH DCCH DCCH DTCH

MAC-d MAC-d

MAC-es
MAC-es /
MAC-e
MAC-e MAC-e EDCH FP EDCH FP

PHY PHY TNL TNL TNL TNL

Uu
UE NodeB Iub DRNC Iur SRNC

• E-DCH is terminated by MAC-e in the Node B, while other transport channels are terminated at RNC level
HSUPA – Node B controlled Scheduling
The UE provides the BTS scheduler with (in MAC-e header):
• UE buffer occupancy: how much data is in RLC buffers
• Information about the priority of the data in the buffer NB
• Available transmission power resource

In physical layer (E-DPCCH) the UE provides to BTS:

MAC-e PDU
• E-TFI, indicating what is transmitted in the E-DPDCH E-DPCCH on E-DPDCH
(L1) (L2)
• Information of the HARQ redundancy version for the packet
• Timing is known, thus BTS knows which ARQ channel to expect
• Happy bit: Is the current data rate satisfactory
• UE would not be happy of the data rate if it could transmit with higher rate
• I.e. have enough data in its buffers and would have sufficient power resource to transmit with a higher
power than currently

Faster scheduling reduces noise rise

variations
Operation point can be
 Less headroom needed increased because
variance is reduced
 Cell capacity and user data rates
are increased

A first estimated increase in the cell

capacity is 15-20% for the same noise
rise outage
HSUPA – Node B controlled Scheduling
• Node B scheduler shares resources between UEs with HSUPA connections
• RNC scheduler continues to manage R99 DCH connections
• Similar to HSDPA scheduler in MAC-hs, HSUPA scheduler in MAC-e is faster than an RNC scheduler
• Both absolute and relative grants are used (E-AGCH and E-RGCH)
• Scheduling period is 10ms

RNC Node B UE
384 1. 384 384 1. RNC limits the E-TFCI based upon UE
256 capability and QoS profile
256 256
12 12 12 2. NodeB limits E-TFCI based upon packet
2. scheduling principles
8
64 8
64 8
64 3.
3. UE limits E-TFCI based upon transmit
32 32 32 power capability
16 16 16 4. 4. UE selects E-TFCI based upon data to
8 8 8 be transferred
Zero Grant Zero Grant Zero Grant
HSUPA – Fast Scheduling Reduces Noise Variance
• Scheduler uses absolute and relative grants to maximise the utilisation of every user and minimise
the difference between the requested and allocated bit rates
• Scheduling decisions are based upon the
– Uplink interference margin
– Physical layer feedback (happy bit)
– Iub capacity allocation
– Available baseband processing (CE) capacity

• HSUPA scheduler combines Increase in

throughput and load based interference
algorithms Power based
floor (dB)
scheduling
• Throughput based scheduling is
applied for lower loads
• Power based scheduling is applied Throughput
for higher loads based
scheduling

Cell Load (%)

HSUPA - UE Categories
• BPSK modulation with multicode transmission used to achieve high data rates
• 6 different UE categories defined, categories are characterised by
• Number of parallel codes supported
• Support of 2ms TTI - 10ms TTI supported by all the HSUPA UEs
• Theoretical peak bit rate up to 5.74 Mbps for category 6 UE using 2 ms TTI
• No coding and no retransmissions - all bits must be delivered correctly over the air…

HSUPA Transport
Codes x Spreading TTI Data rate
Category Block size

1 1 x SF4 10 7110 0.71 Mbps

2 2 x SF4 10 14484 1.45 Mbps
2 2 x SF4 2 2798 1.40 Mbps
3 2 x SF4 10 14484 1.45 Mbps
4 2 x SF2 10 20000 2 Mbps
4 2 x SF2 2 5772 2.89 Mbps
5 2 x SF2 10 20000 2 Mbps
6 2xSF2 + 2xSF4 10 20000 2 Mbps
6 2xSF2 + 2xSF4 2 11484 5.74 Mbps
Maximum code allocation with HSUPA
• Allocation of 15 is not possible when HSUPA is enabled in the cell

SF=1

SF=2

SF=4 Branch needed to

allocate E-AGCH
SF=8

SF=16 14 HS-PDSCH codes

SF=32
Codes for common
SF=64 channels in the cell Codes for associated DCHs and
non-HSDPA users
SF=128

SF=256 Up to 15HS-SCCH
codes E-AGCH (256)

E-RGCH/E-HICH (128)
Dual Carrier HSDPA
 What is DC HSDPA?
Dual-Carrier HSDPA, also known as Dual-Cell HSDPA, is part of 3GPP Release 8 specification. It is the natural evolution
of HSPA by means of carrier aggregation in the downlink.

The basic idea of the multicarrier feature is to achieve better resource utilization and spectrum efficiency by means of joint
resource allocation and load balancing across the downlink carriers.
New HSDPA User Equipment categories 21-24 have been introduced that support DC-HSDPA. DC-HSDPA can support up to
42.2 Mbit/s, but unlike HSPA, it does not need to rely on MIMO transmission.

The support of MIMO in combination with DC-HSDPA will allow operators deploying Release 7 MIMO to benefit from the DC-
HSDPA functionality as defined in Release 8. While in Release 8 DC-HSDPA can only operate on adjacent carriers, Release
9 also allows that the paired cells can operate on two different frequency bands. Later releases allow the use of up to four
carriers simultaneously.
From Release 9 onwards it will be possible to use DC-HSDPA in combination with MIMO being used on both carriers. The
support of MIMO in combination with DC-HSDPA will allow operators even more capacity improvements within their network.
This will allow theoretical speed of up to 84.4 Mbit/s.
DC HSDPA

2 Collocated cells (Different UARFCN) are

serving 1 UE in same time. It is called
Dual carrier HSDPA.

1 cell theoretical can reach 21 Mbps on

15 code (64Qam). With Dual Carrier
HSDPA scheme, it can reach double
42Mbps in Downlink (21*2).
Parameter To activate DC HSDPA

We also need to add nodeb Multi group for HSDPA and add DC cells into.
Don’t Forget Nodeb license check for 64QAM/DC HSDPA number.
HSUPA 2ms
What is HSUPA 2ms?
E-DCH Transport Block Size Tables for FDD": 10 ms TTI and 2 ms TTI. The 10 ms TTI is mandatory for all HSUPA UEs and
the 2 ms TTI is optional.
The HSUPA 2ms TTI feature enables HSUPA services to use the 2 ms TTI on the Uu interface. Compared with using the 10
ms TTI, using this feature provides the following benefits:
• Higher uplink peak rate for a single UE
• Faster data scheduling
• Shorter data transmission delay over the Uu interface
• This feature is introduced in 3GPP Release 6 and is supported by RAN10.0 and later versions.
Radio Bearers
HSUPA services use the 2 ms TTI when the following conditions are met:
• The cell and UE support this feature.
• The MAP_HSUPA_TTI_2MS_SWITCH check box under the MapSwitch parameter in the SET UCORRMALGOSWITCH
command is selected.
If HSUPA PS services use the 2 ms TTI, the following conditions must be met:
• For HSUPA PS streaming services, the maximum uplink rate is greater than or equal to the value of the
StreamHsupa2msTtiRateThs parameter.
• For HSUPA PS BE services, the maximum uplink rate is greater than or equal to the value of the
BeHsupa2msTtiRateThs parameter.
You can run the SET UFRC command to set the StreamHsupa2msTtiRateThs and BeHsupa2msTtiRateThs parameters.
HSUPA 2ms (5,76Mbps) support on UE cat6
onward