Home > HSPA+ > T-Mobile ‘4G’- like HSPA+

T-Mobile ‘4G’- like HSPA+

So the word on the street according to T-Mobile USA is HSPA+ is like or better than ‘4G’. Well there is some strength in the argument if not all – let’s explore and see if the litmus test proves to be true. Interestingly enough, HSPA+ offers a better and cleaner solution at the moment for circuit switch calls and IRAT handover to GSM then LTE or WiMAX!

T-Mobile USA announced last week that it will increase the maximum possible data speeds offered on its upgraded 3G network to 42 Mbit/s in 2011. The fourth-ranked mobile operator in the US has been upgrading its GSM-based 3G network with a high-speed packet access plus (HSPA+) software update for months now. Upgraded markets offer a theoretical peak download speed of 21 Mbit/s. The operator has previously said that this translates into average download speeds in the 5-to-8-Mbit/s range using its HSPA+ webConnect Rocket laptop stick with peaks at 10 Mbit/s and over.

This already puts T-Mobile in the lane with Clearwire LLC average mobile download speeds of 3 to 6 Mbit/s and within spitting distance of Verizon Wireless ‘s promised average downloads of 5 to 12 Mbit/s for its upcoming Long-Term Evolution (LTE) network. After being late to 3G, T-Mobile clearly doesn’t intend to lose the marketing battle to offer “4G speeds” even if no carrier actually provides a real 4G service yet.

“Our new network offers today’s available 4G speeds to more people than any other wireless network in the country and we’re not done yet,” says Neville Ray, chief network officer for T-Mobile USA, in a statement. “We are now on pace to more than double our HSPA+ footprint — reaching more than 200 million people by this year — with plans to offer 42-Mbit/s theoretical speeds in 2011.”

The operator may also be able to make the 42-Mbit/s move without significant hardware upgrades on the infrastructure side. Other operators like Telus Corp. have started to work with HSPA+ “dual-carrier” technology to get a speed boost without multiple extra antennas.

FACT Check #1 – HSPA+ Data Rates in 3GPP Rel.8, 9

HSPA+ makes all-IP architecture come to reality. Base stations connect to the network via standard gigabit Ethernet to the ISP’s edge routers connected to the internet or other ISP via peering arrangements. This makes the network faster, cheaper to deploy and operate.

However the legacy architecture is still possible with the Evolved HSPA. This ‘flat architecture’ communicates ‘user plane’ IP directly from the base station to the GGSN IP router system, using any available link technology. User IP data bypasses the Radio Network Controller (RNC) and the SGSN of the previous 3GPP UMTS architecture versions. This is a major step towards the 3GPP Long Term Evolution (LTE) flat architecture as defined in the 3GPP standard Rel-8. In essence the flat architecture turns the cellular base station into an IP router. It connects to the Internet with cost effective modern IP link layer technologies like Ethernet, and for user plane data it is not tied to the SONET/SDH infrastructure or T1/E1 lines any more. It is defined in 3GPP TR25.999.

In UMTS, in each slot the maximum number of bits transmitted is 2560, or rather chips.  If you want to know where this 2560 comes from or why chips then please refer 3GPP TS 25.211.

HSPA+ Calculations

There are 15 slots per 10ms frame so since the TTI for HSDPA is 2ms, there will be 3 slots. So there will be a total of 7680 chips.

QPSK has 2 bits per symbol = 7680 * 2 chips for 2ms = 15360 chips/2ms = 15360 * 1000 /2 chips per second.

Now the SF is fixed at 16 = (15360 * 1000) / (2 * 16) = 480 Kbps

Terminal that uses 15 QPSK codes will get 480 * 15 = 7.2Mbps. On other hand 16 QAM will have 4 bits per symbol so the rate would be 7.2 * 2 = 14.4Mbps

Multi-carrier HSPA+ /Dual Cell HSPA+

Rel-8 introduced dual-carrier HSDPA operation in the downlink while Rel-9 similarly introduced dual-carrier HSUPA operation in the uplink and also enhanced the dual-carrier HSDPA operation by combining it with MIMO. The maximum data rate in Release-8 HSDPA is 42Mbps. With Dual-carrier operation, this could be doubled to 84Mbps and with 4 carriers, this will become 168Mbps!

Further enhanced multi-carrier HSDPA operation is being specified for Rel-10, where the base station will be able to schedule HSDPA transmissions over three or four carriers simultaneously to a single user with the carriers are spread over one or two frequency bands. Solutions specified in earlier releases can be reused to a large extent. The difference is that now it is possible to configure a UE with one primary serving cell and up to three secondary serving cells. As in earlier releases, the secondary serving cells can be activated and deactivated dynamically by the base station using so-called “HS-SCCH orders.” With MIMO transmission on all four carriers, the peak rate would be doubled to 168 Mbps compared to Rel-9 and for typical bursty traffic the average user throughput would also experience a substantial increase.

Dual carrier operation has been promoted in the 3GPP standardization by Qualcomm and Ericsson, and a RAN1/4 Study Item (SI) entitled “Feasibility Study on Dual Cell HSDPA Operation” was proposed and approved in TSG RAN #39. In TSG RAN #40 the SI was further maintained considering the good progress status, and a Work Item (WI) entitled “Dual-Cell HSDPA Operation on Adjacent Carriers” was approved. The WI was completed in December 2008 TSG RAN#42 with the final status report and with a complete set of change requests to the specification. The work item “Conformance Test Aspects – Dual-Cell HSDPA operation on adjacent carriers” was started to allow for conformance testing for the new specification.

Legacy UEs (HSDPA category 1-20) do not support dual carrier operation. Hence, new HSDPA UE categories 21-24 have been introduced with the following operational considerations:

  • Co-existence with legacy UEs.
  • Capability to be served dynamically (on per TTI granularity) on either or both of the allocated carriers at the same time.
  • Capability to feedback ACK/NACK and CQI for both the carriers simultaneously.
  • HS-SCCH less operation and uplink power control to be carried on anchor carrier.
  • The Node-B may dynamically enable and disable the supplementary carrier to save UE battery power depending upon the downlink traffic and channel considerations by means of HS-SCCH orders.
  • No support of MIMO and dual cell simultaneously.
  • Capability to perform measurements on the supplementary carrier without compressed mode.
  • Mobility procedures are supported based on serving HS-DSCH cell. In a baseline DC-HSPA configuration Transmit Diversity (STTD) and Receive Diversity (2 Rx-Antennas) are used on each carrier.

FACT Check #2 – Spectral Efficiency/Shannon Bound

To better understand HSPA+ and reasons for deploying the different technologies and to better predict the evolution of capability, it is useful to examine spectral efficiency. The evolution of data services will be characterized by an increasing number of users with ever-higher bandwidth demands. As the wireless-data market grows, deploying wireless technologies with high spectral efficiency will be of paramount importance. Keeping all other things equal such as frequency band, amount of spectrum, and cell site spacing, an increase in spectral efficiency translates to a proportional increase in the number of users supported at the same load per user—or, for the same number of users, an increase in throughput available to each user. Delivering broadband services to large numbers of users can best be achieved with high spectral-efficiency systems, especially because the only other alternatives are using more spectrum or deploying more cell sites. Increased spectral efficiency, however, comes at a price. It generally implies greater complexity for both user and base station equipment.


C is the channel capacity in bits per second;

B is the bandwidth of the channel in hertz (passband bandwidth in case of a modulated signal);

S is the total received signal power over the bandwidth (in case of a modulated signal, often denoted C, i.e. modulated carrier), measured in watt or volt2;

N is the total noise or interference power over the bandwidth, measured in watt or volt2; and

S/N is the signal-to-noise ratio (SNR) or the carrier-to-noise ratio (CNR) of the communication signal to the Gaussian noise interference expressed as a linear power ratio (not as logarithmic decibels).

Complexity can arise from the increased number of calculations performed to process signals or from additional radio components. Hence, operators and vendors must balance market needs against network and equipment costs. One core aspect of evolving wireless technology is managing the complexity associated with achieving higher spectral efficiency. The reason technologies such as OFDMA are attractive is that they allow higher spectral efficiency with lower overall complexity; thus their use in technologies such as LTE and WiMAX.

The roadmap for the EDGE/HSPA/LTE family of technologies provides a wide portfolio of options to increase spectral efficiency. When determining the best area on which to focus future technology enhancements, it is interesting to note that HSDPA, 1xEV-DO, and IEEE 802.16e-2005 all have highly optimized links—that is, physical layers. In fact, as shown in Figure above, the link layer performance of these technologies is approaching the theoretical limits as defined by the Shannon bound. (The Shannon bound is a theoretical limit to the information transfer rate [per unit bandwidth] that can be supported by any communications link. The bound is a function of the Signal to Noise Ratio [SNR] of the communications link.) The Figure also shows that HSDPA, 1xEV-DO, and IEEE 802.16e-2005 are all within 2 to 3 decibels (dB) of the Shannon bound, indicating that there is not much room for improvement from a link layer perspective. Note that differences do exist in the design of the MAC layer (layer 2), and this may result in lower than expected performance in some cases as described previously.

The curves in the figure above apply to an Additive White Gaussian Noise Channel (AWGN). If the channel is slowly varying and the effect of frequency selectivity can be overcome through an equalizer in either HSDPA or OFDM, then the channel can be known almost perfectly and the effects of fading and non-AWGN interference can be ignored—thus justifying the AWGN assumption. For instance, at 3 km per hour, and fading at 2 GHz, the Doppler spread is about 5.5 Hz. The coherence time of the channel is thus 1 sec/5.5 or 180 msec. Frames are well within the coherence time of the channel, because they are typically 20 msec or less. As such, the channel appears “constant” over a frame and the Shannon bound applies. Much more of the traffic in a cellular system is at slow speeds (for example, 3 km/hr) rather than at higher speeds. Thus, the Shannon bound is relevant for a realistic deployment environment. As the speed of the mobile station increases and the channel estimation becomes less accurate, additional margin is needed. This additional margin, however, would impact the different standards fairly equally. The Shannon bound only applies to a single user; it does not attempt to indicate aggregate channel throughput with multiple users. It does indicate, however, that link layer performance is reaching theoretical limits. As such, the focus of future technology enhancements should be on improving system performance aspects that maximize the experienced SNRs in the system rather than on investigating new air interfaces that attempt to improve the link layer performance.

FACT Check #3 – WiMAX

Although WiMAX has emerged as a potential alternative to 3GPP technologies for wide-area wireless networks. Based on OFDMA and accepted by the International Telecommunications Union (ITU) as an IMT-2000 (3G technology) under the name OFDMA TDD WMAN (Wireless Metropolitan Area Network), WiMAX is challenging existing wireless technologies—promising greater capabilities and greater efficiencies than alternative approaches such as HSPA. But as WiMAX, particularly mobile WiMAX has come closer to reality, vendors have continued to enhance HSPA, and perceived WiMAX advantages are no longer apparent. Here is one of my earlier studies for a comparison of T-Mobile HSPA+ vs. Sprint WiMAX, though Sprint has latency due to Mobile IP (using Clearwire network over a Foreign Agent).

It should be noted, however, that IEEE 802.16e-2005 contains some aspects that may limit its performance, particularly in scenarios in which a sector contains a large number of mobile users. The performance of the MAC layer is inefficient when scheduling large numbers of users, and some aspects—such as power control of the mobile station—are provided using MAC signaling messages rather than the fast power control used in WCDMA and other technologies. Thus, while WiMAX uses OFDMA, the performance will likely be somewhat less than HSPA due to increased overhead and other design issues.

Relative to LTE, WiMAX has the following technical disadvantages: 5 msec frames instead of 1 msec frames, Chase combining instead of incremental redundancy, coarser granularity for modulation and coding schemes and vertical coding instead of horizontal coding.42 One deployment consideration is that TDD requires network synchronization. It is not possible for one cell site to be transmitting and an adjacent cell site to be receiving at the same time. Different operators in the same band must either coordinate their networks or have guard bands to ensure that they don’t interfere with each other.

This may introduce problems as more operators introduce networks in the same spectrum band; for example, the 2.5 GHz band in the United States may be used for both TDD and FDD operation.

Although IEEE 802.16e exploits significant radio innovations similar to HSPA+ and LTE, it faces challenges such as economies of scale and technology maturity. Very few operators today have access to spectrum for WiMAX that would permit them to provide widespread coverage.

In reference to economies of scale, GSM-HSPA subscribers number in the billions. Even over the next five years, the number of WiMAX subscribers is likely to be quite low. For example, Informa projects 82.1 million by 201343 while Maravedis predicts a lower 75 million WiMAX subscribers by the end of 2014. Finally, from a technology standpoint, mobile WiMAX on paper may be slightly more capable than today’s available versions of HSPA. Further LTE is not very far from deployment and it shall bring in significant changes to latency, throughput for mobile devices.

Here is one table that helps for comparison of 4G technologies with HSPA+


Much of the debate in the wireless industry is on the merits of different radio technologies, yet other factors are equally important in determining the services and capabilities of a wireless network. These factors include the amount of spectrum available, backhaul, and network topology. Spectrum has always been a major consideration for deploying any wireless network, but it is particularly important when looking at high-performance broadband systems. HSPA and HSPA+ can deliver high throughput rates on the downlink and uplink with low latency in 5 MHz channels when deployed in single frequency (1/1) reuse. By this, we mean that every cell sector (typically three per cell) in every cell uses the same radio channel(s). As previously discussed, an OFDMA approach in a 5 MHz radio channel yields only a small performance advantage. To achieve higher data rates requires wider radio channels, such as 10 or 20 MHz wide channels, in combination with emerging OFDMA radio technologies. Very few operators today, however, have access to this much contiguous spectrum. It was challenging enough for GSM operators to obtain UMTS spectrum. If delivering very high data rates are the objective, then the system must minimize interference. This result is best achieved by employing looser reuse, such as having every sector use only one-third of the available radio channels (1/3 reuse). The 10 MHz radio channel could now demand as much as 30 MHz of available spectrum.

Finally, evolved HSPA+ systems, with peak rates of 42 Mbps, will largely match the throughput and capacity of OFDMA-based approaches in 5 MHz, 3GPP adopted OFDMA with 3GPP LTE, which will provide a growth platform for the next decade. The overall network topology also plays an important role, especially with respect to latency. Low latency is critical to achieving very high data rates, because of the way it affects TCP/IP traffic. How traffic routes through the core network—how many hops and nodes it must pass through—can influence the overall performance of the network. Other innovations, such as MIMO and higher order modulation are now being deployed.The core EPC/SAE network for 3GPP LTE emphasizes just such a flatter architecture. In summary, it can be misleading to say that one wireless technology outperforms another without a full understanding of how that technology will be deployed in a complete system that also takes spectrum and handset ecosystem.

Resources: Rysavy Research & Nomor Research GmbH

Categories: HSPA+ Tags: ,
  1. No comments yet.
  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: