The speed challenge

There are some questions that come up time and time again when talking about WiMAX. One of those is how will 802.16e differentiate from 3.5G cellular networks in terms of downlink and uplink performance? Is there really such a big difference, particularly as 3G moves towards HSPA and HSPA+?

As you would expect, the 3GPP camp, which represents GSM, EDGE and WCDMA operators, says there is not much difference at all. The idea put forward by the WIMAX community that 802.16e is somehow complementary to HSPA, because it offers a much better data and mobile internet experience than voice-centric cellular, is one it unequivocally rejects.

A spokesperson for the 3G Americas lobby group, which represents 3GPP operators, sent me his views on this subject after I asked for his assessment of Clearwire’s chances of making a significant impact in the US. Below is part of his email response:

3G Americas and some independent technical research analysts have studied the performance metrics of WiMAX versus HSPA and HSPA+. WiMAX has technical performance characteristics that are similar to HSPA and HSPA+. HSPA is currently providing typical throughput speeds of 700-1700Kbps for downloads and 500-1200Kbps for uploads. To put the evolution of HSPA into further perspective, HSPA+, already deployed in many parts of the world, is delivering theoretical peak throughput performance of 21Mbps for downloads today, with up to 42Mbps on the downlink and 11.5Mbps on the uplink peak theoretical throughput speeds available in the future. Typical peak user rates for HSPA+ have the possibility of reaching 5Mbps on the downlink and 3Mbps on the uplink.

Of course, any claims made on speed – particularly when the word ‘theoretical’ is used and a broad ‘typical’ speed-range is cited – tend to generate heated responses from the opposing camp. Arguments about spectral efficiency; questions about the number of base stations required; and enquiries into what assumptions are made about the number of users per base station sector are just some of the approaches invariably used to dig beneath and undermine the headline claims. It makes the argument more complex to follow but is necessary nonetheless. Unfortunately (or fortunately, depending on your point of view) complexity suits only the side with the weakest argument.

What does appear to be going in mobile WiMAX’s favour is that it’s doing a pretty good job – in a commercial environment – on delivering on what it claims it would deliver in terms of speed. In the run-up to the launch of the Clear service in Portland, Oregon, Clearwire was saying it was aiming to offer between 3Mbps and 5Mbps on the downlink, and between 1Mbps and 2Mbps on the uplink. And according to independent reports, Clearwire is indeed achieving this.

Worldmax, an 802.16e operator in the 3.5GHz band in the Netherlands, announced last month a package comprising 8Mbps on the downlink and 800Kbps on the uplink. This is much faster than the speeds offered by the country’s cellular operators, although it has to be said that Worldmax is initially focused on fixed and nomadic services rather than true mobility. Even so, high-speed data sessions tend to be done when stationary.

Moving forward, the key differentiator between WiMAX and HSPA on speed might be the uplink. The WiMAX camp typically claims that it can provide three to four times more uplink throttle than HSPA, which will be significant if more data-hungry interactive applications – which require a faster uplink – become popular.