HSPA+ is an advanced version of the 3G cellular radio technology known as wideband code division multiple access (W-CDMA). HSPA refers to high-speed packet access with an enhancement to W-CDMA that combines high-speed downlink packet access (HSDPA) and high-speed uplink packet access (HSUPA). HSPA+ is a further enhancement that introduces higher-level modulation techniques such as 16-state quadrature amplitude modulation (16QAM) and 64-state QAM (64QAM) in the uplink and downlink respectively to replace the standard quadrature phase-shift keying (QPSK) modulation of the base W-CDMA standard.
Whereas W-CDMA could achieve data rates in the 384-kbit/s to 2-Mbit/s range, HSPA+ can achieve rates of 14, 21, and even 42 Mbits/s under some conditions. These rates compare favorably with the newer Long-Term Evolution (LTE) fourth-generation technology and extend a cellular carrier’s ability to adequately serve data-hungry customers and maximize its investment in 3G networks before upgrading to LTE.
Table Of Contents
- Carrier Investment Choices
- HSPA+ Explained
- HSPA+ Technical Details
- Dual-Carrier HSPA+
- New Test Requirements For HSPA+
- Glossary of HSPA+ Terms
For most of the 20th century, operators used the work of Danish mathematician and engineer A.K. Erlang as the basis for network planning: essentially predicting the number of simultaneous users a telecommunications network would have to support. As long as networks were used mainly for voice calls, the same broad principles applied to mobile networks, with the added flexibility of using a smaller cell size in geographic “hot spots” where more users could be expected and cell capacity exceeded.
But the coming of the home PC in the 1990s, particularly in its laptop form, meant a big change in demand. Fixed-line data modems delivering up to 56 kbits/s of data and General Packet Radio Service (GPRS) cellular modems at up to 28 kbits/s offered a less than acceptable user experience and gave operators a new challenge.
Three main solutions emerged: Data Over Cable Service Infrastructure Specifications (DOCSIS) modems using existing cable TV infrastructure, Asynchronous Digital Subscriber Line (ADSL) modems using the copper of fixed-line telephony, and third-generation cellular networks with higher cell capacities (also known as “mobile broadband”). Today, as the take-up of data services on mobile networks continues to rocket, the rules of network provision need to be rewritten.
First, data services are by their nature discontinuous. Moving to packet rather than circuit service delivery allows more users to share the same resource (though at the same time making directing the data more complex).
Second, the progressively smaller cell sizes needed to fully cover the needs of ubiquitous mobile phone ownership provides additional bandwidth for both voice and data.
Third, successive advances in technology and system specifications developed by the Third Generation Partnership Project (3GPP) have provided higher cell capacity and improvements in single-user data rate from the 384 kbits/s of original W-CDMA (specification release 99) through HSDPA and HSUPA—collectively HSPA—to evolved HSPA (HSPA+), dual-carrier HSDPA (DC-HSDPA), and LTE (Fig. 1).
The increases in data rate came courtesy of increased modulation density made possible by better components, particularly in digital receivers. Current HSPA networks deliver data rates up to 14 Mbits/s downlink and 5.5 Mbits/s uplink. HSPA+ takes this to 21 Mbits/s downlink and 11 Mbits/s uplink. DC-HSDPA doubles the downlink speed, and first-generation LTE starts out at 100 Mbits/s downlink and 50 Mbits/s uplink.
HSPA+ is an advanced 3G technology (3.9G) that some carriers such as T-Mobile call a 4G technology. While most cell-phone operators are transitioning to LTE, many are still expanding their HSPA capability because of its high speed and low cost.
While the industry hype is all about LTE, many operators have chosen HSPA+ as a more cost-effective short-term upgrade strategy. For those whose networks are based on 3GPP specifications, most of which have already deployed HSPA, HSPA+ and DC-HSDPA are software upgrades—ideal in these days of tight budgets. HSPA and its evolutions deliver high enough speeds to compare with most home broadband systems, so the user experience will adequately meet customer expectations.
Figure 2 shows how device manufacturers view the opportunity to provide end users with equipment for high-speed services, with HSPA and derivatives dominating for the foreseeable future. The Global Suppliers Association (GSA) predicts the following challenge for operators:
- By 2015, there will be 3.5 billion mobile broadband users worldwide.
- The total data traffic volume in 2015 will be more than 30 times that of 2010.
- Shipments of mobile broadband-enabled consumer devices will increase 55-fold between 2008 and 2014 with total shipments reaching 58 million per year in 2014, including new product categories such as cameras and internet radios.
The major goals of HSPA+, as defined by the 3GPP standards organization, are:
- To exploit the full potential of the CDMA physical layer before moving to the orthogonal frequency division multiplex (OFDM) physical layer (PHY) of LTE
- To achieve performance comparable with LTE in a 5-MHz channel bandwidth
- To provide smooth interworking between HSPA+ and LTE
- To achieve co-existence of both technologies in one network
- To allow operation in a packet-only mode for both voice and data
- To be backward compatible with earlier user devices
Current W-CDMA systems are all based on a 5-MHz channel bandwidth, of which 3.84 MHz is used and the remainder acts as a guard band between channels. New in Release 8 is the option for dual-carrier HSDPA, which enables the system to aggregate the content of two contiguous carriers—doubling downlink data rates at a stroke and further enabling HSPA to maintain its place in the high-speed world.
It’s important to recognize that improving uplink performance also helps the downlink. By providing faster acknowledgement, downlink capacity and latency both benefit. The option to have the HSPA+ network operate fully in packet mode for both voice and data updates the backhaul network to make future LTE deployment simpler: only the PHY (basestation radio) would need a major upgrade.
Important features of HSPA+ are:
3GPP Release 7
- Downlink multiple-input multiple output (MIMO)
- Higher-order modulation for uplink (16QAM) and downlink (64QAM)
- Continuous packet connectivity (CPC)
3GPP Release 8
- Combined downlink MIMO and 64QAM: peak rate can be up to 42 Mbits/s
- Circuit-switched (CS) over HSPA
- Dual-carrier HSDPA (though this cannot be combined with MIMO)
Releases 9 and beyond add further multi-carrier capability, including non-contiguous channels, adding supplemental downlink capacity using unpaired spectrum, enhanced MIMO, and 256QAM modulation. Current visions show “HSPA+ Advanced,” supporting more than 300 Mbits/s downlink and almost 70 Mbits/s uplink, in Release 11. How the tradeoffs between the further developments of HSPA+ and LTE will change with time remains to be seen.
With the possibility to use 16QAM on the E-DCH (Enhanced Dedicated Channel) in the uplink, HSPA+ can achieve uplink peak data rates of 11.5 Mbits/s.
With the possibility to use 64QAM in the downlink, HSPA+ can achieve downlink data rates of 21 Mbits/s. 64QAM is a user equipment (UE) capability, i.e., not all UE will be able to support it.
CPC is a collection of enhancements that allow more users to be continuously connected to the network and, at the same time, increase UE battery life and improve the link quality, especially for low-data-rate services like Voice over Internet Protocol (VoIP). To support CPC, the following uplink and downlink improvements had to be made:
- New uplink DPCCH slot format (no FBI or TFCI overhead and longer TPC field)
- New uplink DPCCH gating/discontinuous transmission (DTX) UE to BTS
- New downlink discontinuous reception (DRX) at the UE
- New downlink: so called HS-SCCH-less operation (helps reduce signalling overhead)
These features are attractive to service providers because they can increase capacity and are relatively simple upgrades to the network and terminals.
Figure 3 illustrates the CPC concept with a real-world example. In this case, the user is downloading a Web page. After the Web page has been downloaded, the user stops browsing and reads the page. During this reading time, the user does not require any data exchange between the mobile and the base transceiver station (BTS). In the upper graphic (HSPA release 6), the uplink DPCCH is continuously transmitted and the downlink channels are continuously received by the UE during the reading phase by the user.
The lower graphic illustrates the application of the CPC mode. In this case, after the Web page is downloaded, the UE quickly goes into the uplink DPCCH gating mode (or DTx mode). Following this, the UE receiver goes into discontinuous reception (or DRx).
The scheduling of these two events is managed by a series of rules to maximize overlap so DTx and DRx will happen at approximately the same time. The UE then can go into a “micro sleep mode,” which significantly helps battery life. It also reduces the signal to interference ratio (SIR) generated by all these channels including the HS-SCCH, which in turn allows more users to be connected at the same time (CELL_DCH state).
HS-SCCH-less operation is used to reduce the signaling overhead, especially for services using relatively small packets, e.g., VoIP. The first HS-DSCH transmission of small transport blocks on predefined HS-PDSCHs is performed without the accompanying HS-SCCH. Consequently, the UE has to blind-detect the transport format used on HS-DSCH. To help the UE identify whether an HS-DSCH transmission is addressed to it, CRC attachment method 2 is defined using a 24-bit CRC on the HS-DSCH masked with the UE ID (or H-RNTI). Finally, the first transmission always uses QPSK and redundancy version 0.
HARQ-ACK is sent on the HS-DPCCH if the UE successfully blind-detects the first transmission. If the blind detection is not successful, the UE buffers the received data for potential later retransmission and soft combining without transmitting an explicit NACK. If the first transmission fails, HARQ retransmission is used for HSDPA operation.
However, at most two retransmissions are allowed, the redundancy version is preconfigured to 3 and 4 respectively, and QPSK is always used. Unlike the first transmission, the HARQ retransmissions include the HS-SCCH configured as HS-SCCH type 2. CRC attachment method 2 is still used with retransmissions.
3GPP Release 8 defines dual-carrier or dual-cell high-speed downlink packet access (DC-HSDPA) to allow the network to transmit HSDPA data to a mobile device from two cells simultaneously, doubling achievable downlink data rate to 42 Mbits/s. Dual-carrier operation is characterized as simultaneous reception of more than one HS-DSCH transport channel. Dual-cell operation may be activated and deactivated using HS-SCCH orders.
While operating in DC-HSDPA mode, the UE receives HSDPA transmissions from two cells. The first cell is known as the serving cell, and the second is the secondary serving cell. The two cells transmit on adjacent carrier frequencies and potentially two different cell powers.
While the serving cell has a full set of common channels (SCH, P-CCPCH, CPICH, PICH, etc.), the UE must assume the secondary serving cell will only transmit a CPICH. Both cells can transmit HS-PDSCH and HS-SCCH to the UE simultaneously.
The data content of each cell’s HS-PDSCH is different. In each transmission time interval (TTI), the UE reads the configuration of the HS-PDSCH transmitted by each cell from an HS-SCCH channel also transmitted by each cell with an independently assigned H-RNTI for each cell. This means the network has the flexibility to configure each cell’s HS-PDSCHs the same or differently.
On the uplink, the UE transmits a single HS-DPCCH to the serving cell. This HS-DPCCH carries either 1 or 2 Ack/Nack bits depending on how many HS-PDSCH transmissions the UE attempted to decode. The HS-DPCCH also carries two CQI reports, one for each cell. While the secondary serving cell is active, the UE uses deltaAck+1, deltaNack+1, or deltaCQI+1 to determine the HSDPA gain factor (Ahs), and Ahs = 38/15 with index 9 being added to the existing capabilities.
To the MAC-ehs layer, the two cells essentially look like two HS-DSCH transport channels. Each of these HS-DSCH channels is controlled by its own independent HARQ process entity, with each entity containing a unique set of HARQ processes. Each of the two HARQ process entities is fed by a common priority queue, which means the rest of the stack from MAC-d upwards is unaware that two carriers are being used to transmit data to the UE.
With signalling, the UE indicates whether it supports DC-HSDPA in the RRC Connection Setup Request message and then signals its DC-HSDPA category in the RRC Connection Setup Complete message. The network enables and activates DC-HSDPA at call setup in the RRC Connection Setup or RB Setup message.
Once on a connection, DC-HSDPA can be enabled or disabled by all the reconfiguration messages or by using the RB Release or Active Set Update message. The Release 8 versions of all these messages contain a new information element (IE), Downlink Secondary Cell Info FDD, to signal the configuration of the secondary serving cell in terms of its downlink UARFCN, primary scrambling code, HS-SCCH channelization codes, 64QAM support, and other parameters.
Today’s devices, whether they’re low-cost feature phones, smart phones, tablets, or laptop data cards, typically already support GSM, GPRS, EGPRS, WCDMA, and HSPA. In adding Release 7 and 8 HSPA+ capabilities, developers must ensure they correctly interpret and implement the required new features while ensuring they don’t cause failures in the existing base product.
Agilent’s new products for Release 7 and 8 HSPA+ include extended SystemVue design libraries including features specific to DC-HSDPA, enhanced Signal Studio waveform creation software, new vector signal analysis software that supports MIMO and the analysis of the uplink transmission to the serving cell of the dual-carrier HS-DPCCH Ack/Nack and CQI report decodes, and an updated 8960 (E5515E) wireless communications test set that supports DC-HSDPA connections for all defined HS-DSCH categories that support DC-HSDPA: 21, 22, 23 and 24.
HSPA+ is implemented on most smart phones today and will continue as a mainstream high-speed cellular data service for the time being. It meets video’s speed needs. While LTE will continue to replace older technologies, HSPA+ will continue to be a fallback mode.
CELL-DCH: cell dedicated channel
C-PICH: common pilot channel
CQI: channel quality indicator
CRC: cyclic redundancy check
CS: circuit switched
DPCCH: dedicated physical control channel
FBI: feedback information
HARQ: hybrid automatic repeat request
H-RNTI: high-speed radio network temporary identifier
HS-DSCH: high-speed downlink shared channel
HS-PDSCH: high-speed physical downlink shared channel
HS-SCCH: high-speed synchronization control channel
LTE: Long-Term Evolution
MAC: media access control
MAC-ehs: medium access control—evolved high speed
P-CCPCH: physical common control packet channel
PICH: page indicator channel
RB: radio bearer
RRC: radio resource control
SCH: synchronization channel
SIR: signal-to-interference ratio
TFCI: transport format combination indicator
UARFCN: UTRA absolute radio frequency channel number
UE: user equipment or user experience
UTRA: universal terrestrial radio access