HSDPA-HSPA+ Evolution

Wulan Dwi Anggraini

Faculty of Electro Engineering
Telkom University, Bandung
Email: wulandwiangg@gmail.com

Abstract3GPP Releases 5 and 6 defined the baseline for

mobile broadband access. HSPA evolution in Releases 7, 8 and 9
has further boosted the HSPA capability. The downlink and
uplink data rates improve with higher order modulation 64QAM
downlink and 16QAM uplink. The downlink data rate can also
be increased by a multi-antenna solution (MIMO, Multiple Input
Multiple Output). The terminal consumption is reduced
considerably with HSPA evolution by using discontinuous
transmission and reception. The 3G network capability has
improved enormously from release 99 to Release 9.
KeywordsHSDPA, HSUPA, HSPA evolution, HSPA+

1.

INTRODUCTION

GSM allowed voice go to wireless with more than 4.5

billion subscribers globally. HSPA allowed data go to wireless
with 1.5 billion subscribers globally. At the same time the
amount of data consumed by each subscriber has increased
rapidly leading to a fast increase in the mobile data traffic: the
traffic growth has been 100% per year in many markets. More
than 90% of bits in mobile networks are caused by data
connections and less than 10% by voice calls. Mobile
networks have turned from voice network into data networks.
Mobile operators need to enhance network capabilities to
carry more data traffic with better performance. Smartphone
users expect higher data rates, more extensive coverage, better
voice quality, and longer battery life.
3GPP specified important evolution steps on top of
WCDMA: HSPA for downlink in Release 5 and for uplink
Release 6. The downlink solution, High Speed Downlink
Packet Access (HSDPA) was commercially deployed in 2005
and the uplink counterpart, High Speed Uplink Packet Access
(HSUPA), during 2007. Further HSPA evolution is specified in
3GPP Release 7, and its commercial deployment is expected
by 2009. HSPA evolution is also known as HSPA+.
Most of the current mobile data traffic is carried by HSPA
networks. HSPA+ is expected to be dominant mobile
broadband technology for many years to come due to
attractive data rates and high system efficiency combined with
low cost devices and simple upgrade on top of WCDMA and
HSPA network.
More than 550 operators have deployed the HSPA network
in more than 200 countries by 2014. All WCDMA networks
have been upgraded to support HSPA and many networks also
support HSPA+ with 21 Mbps and 42 Mbps. HSPA technology
has become the main mobile broadband solution globally[1].

2.

HIGH SPEED DOWNLINK PACKET ACCESS

A. Introduction
High Speed Downlink Packet Access (HSDPA) is an
extension of capabilities of UMTS included in 3GPP Release
5, with the target of providing higher bit rates and capacity. It
is also called 3.5G, with transmission rates up to 14.4 Mbps
and 20 Mbps (for MIMO systems) over a 5 MHz bandwidth.
HSDPA can significantly enhance downlink speeds, with
average realistic throughputs of 400-700 kbps and bursts at
over 1 Mbps, even in initial stage. This dramatically improves
the user experience of different applications such as web
browsing, streaming or intranet access. Also, in combination
with HSUPA, it can be driver for advance services like VoIP.
HSDPA shares the spectrum and codes from WCDMA and,
most of the time, only requires a software upgrade of existing
UMTS R99 base stations[2].
B. HSDPA vs Release 99 DCH
In Release 99 there basically exists in the specifications
three different methods for downlink packet data operation:
dedicated channel (DCH), forward access channel (FACH)
and downlink shared channel (DSCH). Since the DSCH has
been de facto replaced with the high-speed DSCH of HSDPA,
it is not covered in more depth here. This has been recognized
in the Third Generation Partnership Project (3GPP) as well
and the DSCH has been removed from the specifications from
Release 5 onwards, simply due to the lack of interest for actual
implementation.
The FACH is used either for small data volumes or when
setting up the connection and during state transfers. In
connection with HSDPA, the FACH is used to carry the
signaling when the terminal has moved, due to inactivity from
Cell_DCH state to Cell_FACH, Cell_PCH or URA_PCH
state. The FACH is operated on its own and, depending on the
state the terminal is in, FACH is decoded either continuously
(Cell_FACH) or based on the paging message. For the FACH
there is neither fast power control nor soft handover.
The Release 99 based DCH is the key part of the system
despite the introduction of HSDPA and Release 5 HSDPA is
always operated with the DCH running in parallel. If the
service is only for packet data, then at least the signaling radio
bearer (SRB) is carried on the DCH. In case the service is
circuit-switched like AMR speech call or video call parallel
to PS data then the service always runs on the DCH. With
Release 6 signaling can also be carried without the DCH, as
explained in connection with the fractional DCH (F-DCH). In
Release 5, uplink user data always go on the DCH (when

HSDPA is active), whereas in Release 6 an alternative is

provided by the Enhanced DCH (E-DCH) with the
introduction of high-speed uplink packet access (HSUPA).
C. Key technologies with HSDPA
Several new channels have been introduced for HSDPA
operation. For user data there is the high-speed downlink
shared channel (HS-DSCH) and the corresponding physical
channel. For the associated signaling needs there are two
channels: high speed shared control channel (HS-SCCH) in
the downlink and high-speed dedicated physical control
channel (HS-DPCCH) in the uplink direction. In addition to
the basic HSDPA channel covered in Release 5 specifications,
there is now a new channel in Release 6 specifications the
fractional dedicated physical channel (F-DPCH) to cover for
operation when all downlink traffic is carried on the HSDSCH.
The general HSDPA operation principle is shown in Figure
1.0, where the Node B estimates the channel quality of each

active HSDPA user on the basis of the physical layer

Figure 1.0 HSDPA Node B scheduling principle
feedback received in the uplink. Scheduling and link
adaptation are then conducted at a fast pace depending on the
scheduling algorithm and the user prioritization scheme.
The other key new technology is physical layer
retransmission. Whereas in Release 99 once data are not
received correctly, there is a need for retransmission to be sent
again from the RNC. In Release 99 there is no difference in
physical layer operation, regardless of whether the packet is a
retransmission or a new packet. With HSDPA the packet is
first received in the buffer in the BTS (as illustrated in Figure
1.1). The BTS keeps the packet in the buffer even if has sent it
to the user and, in case of packet decoding failure,
retransmission automatically takes places from the base station
without RNC involvement. So, the terminal has respectively
combined the transmissions, capturing the energy of both[3].
The adaptive modulation and coding (AMC) technique is
used in order to compensate for variations in radio
transmission conditions, while the transmission power remain
constant. HSDPA-enabled user equipment send channel
quality reports to the base station at 2 ms intervals, which are
used to adapt the modulation or resource accordingly. The fast
scheduling is implemented in Node B, compared to UMTS
R99, where the scheduler is located in RNC. The scheduler
determines to which terminal the transmission on HS-DSCH
Figure 1.1 BTS retransmission handling

will be directed and, depending on the AMC, at what data

rate[2].
B.1 HSDPA Channels
The HSDPA is operated similar to DSCH
together with DCH, which carries the services
with tighter delay constraints, such as AMR
speech. To implement the HSDPA feature, three
new channels are introduced in the physical layer
specifications.

High Speed Downlink Shared Channel (HSDSCH) carries the user data in the downlink
direction, with the peak rate reaching up to
10-Mbps range with 16 QAM (quadrature
amplitude modulation). This channel allows
several users to be time-multiplexed so that
during silent periods the resource are
available to other users.

High Speed Shared Control Channel (HSSCCH) carries the necessary physical layer
control information to enable decoding of the
data on HS-DSCH and to perform the possible
physical layer combining of the data sent on
HS-DSCH in case of retransmission or an
erroneous packet.

Uplink High Speed Dedicated Physical Control

Channel (HS-DPCCH) carries the necessary
control information in the uplink, namely,
ARQ acknowledgements (both positive and
negative
ones)
and
downlink
quality
feedback information[3].
3.

HIGH SPEED UPLINK PACKET ACCESS

A. Introduction
High-Speed Uplink Packet Access (HSUPA) principles for
wideband code division multiple access (WCDMA) for
Release 6 specifications. The Third Generation Partnership
Project (3GPP) term was not HSUPA, but enhanced dedicated
channel (E-DCH). Usage of the term HSUPA instead of EDCH follows the trend that began with employment of the
term HSDPA; and using the term HSUPA synonymously for
corresponding uplink improvement has been widely adopted
in the wireless industry, though not officially covered in 3GPP
specifications[3].

HSUPA is the complementary bearer for HSDPA in the

uplink, which may allow very high upload throughputs (up to
5.8 Mbps).[2]
B. HSUPA vs Release 99 DCH
HSUPA is not a standalone feature, but uses most of the
basic features of the WCDMA Release 99 in order to work.
The only change is a new way of delivering user data from the
user equipment (UE) to the Node B, all other parts of the
specifications remain untouched. For example, the basic
power control loop functions in Release 99 are just as essential
for HSUPA operation. HSUPA provides a flexible path beyond
the 384-kbps uplink which can be seen as the realistic
maximum for WCDMA before HSUPA. A similar technology
to that of HSDPA is being used by introducing fast uplink
hybrid-ARQ (HARQ), Node B based uplink scheduling and
easier multicode transmission than with Release 99. Anyway,
it is worth keeping in mind that none of the old features were
replaced by HSUPA and that it was more of an add-on than a
replacement.
C. Key technologies with HSUPA
The HSUPA feature of the 3GPP WCDMA system is in fact
a new uplink transport channel E-DCH that brought some
of the same features to the uplink as the HSDPA with its new
transport channel high-speed downlink shared channel (HSDSCH) provided for the downlink. The E-DCH transport
channel supports fast Node B based scheduling, fast physical
layer HARQ with incremental redundancy and, optionally, a
shorter 2-ms transmission time interval (TTI). Though unlike
HSDPA HSUPA is not a shared channel, but a dedicated one,
by structure the E-DCH is more like the DCH of Release 99
but with fast scheduling and HARQ than an uplink HSDPA:
that is, each UE has its own dedicated E-DCH data path to the
Node B that is continuous and independent from the DCHs and
E-DCHs of other UEs.
C.1 Fast L1 HARQ for HSUPA
The basic principle behind HARQ for HSUPA is the same
as that for HSDPA. After each transmitted TTI the Node B
indicates to the transmitting UE whether the packet was
received correctly or not. In the event of incorrect reception
the UE will retransmit the packet. The Node B tries to recover
the packet by combining the energy of the retransmission with
previous transmissions until the packet is received correctly or
the maximum number of retransmissions is reached. The
HSUPA HARQ may either use Chase combining where each
retransmission is an exact copy of the initial transmission, or
incremental redundancy where retransmissions contain
additional redundancy bits for the initially transmitted bits.

C.2 Scheduling for HSUPA

In Release 5, HSDPA moved downlink scheduling from
the RNC to the Node B in order to be able to make scheduling
decisions with minimum latency as close to the radio interface

as possible. HSUPA scheduling does the same thing for the

uplink and moves the scheduling to the Node B, but the
similarities between HSDPA and HSUPA scheduling end
there. With HSDPA all the cell power can be directed to a
single user for a short period of time, and in this way can
reach very high peak data rates for that particular UE, but
simultaneously leaving all the others with a zero data rate. In
the next time instant the Node B resources are used to serve
some other UE and so on. Obviously, with HSUPA this is not
possible, because when HSDPA is a one-to-many type of
scheduling HSUPA is a many-to-one scheduling. The uplink
transmission power resources of a cell are distributed evenly
to the users or, to put it simply, each UE has its own
transmitter and can only transmit data from that particular UE.
So, obviously, in the uplink the cells transmission power
resources just cannot be given to a single UE at one time and
to another UE at some other time, but users have their own
transmitter power resource which clearly cannot be shared.
This fact alone leads to a need to have a great level of
parallelism in uplink scheduling and, thus, the dedicated
channel approach was seen as the only feasible one with
HSUPA, contrary to the shared channel approach of HSDPA.
B.2 Motivation and impact of two TTI lengths
While HSDPA only supports a single TTI (2 ms), with
HSUPA there are two TTI lengths 2 and 10 ms that can be
chosen. The motivation for the 2-ms length was the potential
delay benefit while 10 ms was needed for range purposes to
ensure cell edge operation.
A potential delay benefit could be obtained if there are not
too many retransmissions using a 2-ms TTI, as the delay
between retransmissions is shorter compared with the 10- ms
case. A problem occurs when approaching an area of low
geometry (closer to the cell edge) where signaling using a 2ms period starts to consume a lot of transmission power,
especially at the BTS end. This is illustrated in Figure 2.0. The
difference from HSDPA is that now potentially a much larger
number of users are expected to be active simultaneously and,
thus, aiming to also provide downlink signaling to such a large
number of users using a 2-ms period would become
impossible.

Figure 2.0 2-ms and 10-ms TTI applicability in a cell

With data rates below 2 Mbps there are no major
differences from the capacity point of view regardless of the
TTI used. When going above 2 Mbps per user, then the block
size using 10 ms would get too big and, thus, data rates above
2 Mbps are only provided using a 2-ms TTI. As with macrocells, practical data rates in the uplink have limitations due to

transmission power limitations. This means the 10-ms TTI is

expected as the starting value for system deployment[3].
4.

HSPA EVOLUTION

A. Introduction
High Speed Packet Access (HSPA) was included in the
Third Generation Partnership Project (3GPP) Release 5 and 6
for downlink and for uplink. The 3GPP Release 7, 8 and 9 have
brought a number of HSPA enhancements providing major
improvements to the end user performance and to network
efficiency. HSPA evolution is optimized for co-existence with
WCDMA/HSPA supporting legacy Release 99 UEs on the
same carrier and designed for simple upgrade on top of HSPA.
The HSPA evolution aims to improve the end user performance
by low latency, lower power consumption and higher data
rates. The HSPA evolution is also known as HSPA+.
B. Higher Order Modulation
Higher order modulation allow higher peak but rate
without increasing the transmission bandwidth. Release 6
supported QPSK (Quadrature Phase Shift Keying) and
16QAM (Quadrature Amplitude Modulation) transmission in
the downlink and dual-BPSK (Binary Phase Shift Keying) in
the uplink. The release 7 introduced 64QAM transmission for
the downlink and 16QAM for the uplink. Higher order
modulation required signal-to-noise ratio for correct reception
is higher.
C. MIMO
MIMO is a technique that employs multiple transmit
antennas and multiple receive antennas, often in combination
with multiple radios and multiple parallel data streams.
MIMO transmission is shown in Figure 3.0. MIMO allows
double data rate by transmitting dual data streams in good

Figure 3.0 2 x 2 MIMO transmission concept

channel conditions Whereas multipath is an impediment for
other radio systems. MIMO exploits multipath, relying on
signals to travel across different uncorrelated communications
paths. This results in multiple data paths effectively operating
somewhat in parallel and, through appropriate decoding, in a
multiplicative gain in throughput.
MIMO was first standardized in 3GPP TS 25.308 Release
6, and was further developed in Release 7 with spatial
multiplexing for HSPA+ using Double Transmit Adaptive
Array (D-TxAA). In 3GPP Release 7 MIMO cannot be used in
combination with 64 QAM, but this feature is available in
Release 8. Subsequently, in Releases 9, 10 and 11, 2x2 MIMO

is combined with carrier aggregation to double the peak data

rate.
D. Enhanced FACH and Enhanced RACH
WCDMA network data rate and latency are improved with
the introduction of Release 5 HSDPA and Release 6 HSUPA.
Once the packet call has been established, user data can flow
on HSDPA/HSUPA in the Cell_DCH (Dedicated Channel)
state. When the data transmission is inactive for a few
seconds, the UE is moved to Cell_PCH (Paging Channel) state
to minimize the UE consumption. When there is more data to
be sent or receive, the UE is moved from Cell PCH to
Cell_FACH (Forward Access Channel) and to Cell_DCH
state.
Release 99 RACH and FACH can be used for signaling
and for small amounts of user data. The RACH data rate is
very slow, typically below 10 kbps, limiting the use of the
common channels. Release 5 or Release 6 do not provide any
improvements in RACH or FACH performance. The ide in
Release 7 Enhance FACH and Release 8 Enhanced RACH is
to utilize the Release 5 and Release 6 HSPA transport and
physical channels.
Enhance FACH and RACH bring a few performance
benefits:
RACH and FACH data rates can be increased beyond 1
Mbps. The end user can get immediate access to
relatively high data rates without latency of channel
allocation
The state transition from Cell_FACH to Cell_DCH will
be practically seamless. Once the network resources for
the channel allocation are available, a fasttransition can
take place to Cell_DCH.
Discontinuous reception can be used in the Cell_FACH
to reduce the power consumption. The discontinuous
reception can be implemented since Enhance FACH uses
short 2 ms TTI of HSDPA.
E. Achitecture Evolution
3GPP network increasingly be used for IP-based packet
services. 3GPP Release 6 has four network elements in the user
and control plane: base station, RNC (Radio Network
Controller), SGSN (Serving GPRS Support Node) and GGSN
(Gateway GPRS Support Node). The architecture in Release 8
LTE will have only two network elements: A base station in the
radio network and an Access Gateway (a-GW) in the core
network. The a-GW consists of control plane MME (Mobility
Management Entity) and user plane SAE GW (System
Architecture Evolution Gateway). The flat network architecture
reduces the network latency and thus improves the overall
performance of IP based services. The flat model also improves
both user and control plane efficiency. The architecture
evolution is illustrated in Figure 4.0.
With the flat architecture with all RNC functionality in the
base station and using direct tunnel solution, only two nodes
Figure 4.0 Architecture Evolution

are needed for user data operation. This achieves flexible

scalability and introduces the higher date rates with HSPA
evolution with minimum impact on the other nodes in the
network. This is important to achieve low cost per bit and
enable competitive flat rate data charging offerings[4].
5.

CONCLUTION

HSPA evolution has brought a number of important

enhancements on the top of HSPA Release 5 and 6. HSPA
enhancement provide major improvements to the end user
performance in increasing peak bit rates, reducing mobile
terminal power consumption and reducing latency. The peak
bit rates can be tripled per 5 MHz carrier with 64QAM and
2x2 MIMO from 14 Mbps to 42 Mbps. The UE power
consumption is considerably reduced in HSPA evolution due
to discontinuous transmission and reception.
The set-up times are reduced by mapping also the
RACH/FACH common channels on top of HSDPA/HSUPA.
Actually all the services including signaling can be mapped on
the top of HSPA in HPSA evolution. There are only a few
physical layer channels left from Release 99 specifications in
Release 8 otherwise, everything is running on top of HSPA.
That explains also why the end user performance and the
network efficiency have improved considerably compared to
Release 99. 3GPP Release 7 simplifies the network
architecture. The number of network elements for the user
plane can be reduced from four in Release 6 down to two in
Release 7.
REFERENCES
[1]
[2]
[3]
[4]

Holma. H, Toskala. A, and Tapia. Pablo HSPA+ Evolution to Release

12: Performance and Optimization John Wiley and Sons, Ltd., 2014.
Mishra. Ajay, Advance Celluler Network Planning and optimisation
John Wiley and Sons, 2007.
Holma. H, Toskala. A, HSDPA/HSUPA for UMTS John Wiley and
Sons, Ltd., 2006.
Holma. H, Toskala. A, WCDMA for UMTS: HSPA Evolution and LTE
John Wiley and Sons, Ltd., 2010.