Long Term Evolution is the means by which the mobile industry hopes to dominate the broadband industry in the future. There has been rapid progress, yet LTE and 4G will raise questions for the industry beyond the technology. Investment cases, regulation and the business model have all yet to be fully addressed
When it comes to mobile broad-band, the only definition that really makes sense is being able to offer a user experience similar to a typical home, DSL connection. There's now wide agreement that in the 3GPP path, higher throughput versions of HSPA can now provide that. In some cases, operators may want to go beyond that, and in others, where capacity is not an issue, there will be no driver to get to those levels. But there can be little doubting now that HSPA is the benchmark for mobile broadband.
The GSMA says that the number of worldwide subscribers using HSPA networks has topped the 50 million mark, a significant rise from 11 million one year ago. Its research arm, Wireless Intelligence, expects the number of HSPA connections to grow by four million per month by the end of 2008.
As this growth continues, the industry will benefit from economies of scale – although the market is already widespread. The number of operators with commercial HSPA networks has reached 191 and there are now over 740 HSPA-enabled devices available from 116 manufacturers, according to GSMA figures. The most widely-deployed peak data speeds over HSPA are currently between 3.6Mbps and 7.2Mbps. This translates to an end user speed of more than 1Mbps, comparable to many of today's fixed line broadband services.
"These figures highlight the global success of Mobile Broadband and the fact that we are continuing to see greater and greater economies of scale," said Rob Conway, CEO and Member of the Board of the GSMA. "This is driving down the cost of devices and equipment and enabling more and more users across the world to enjoy easy access to media-rich services anywhere at anytime.
Certainly, operators are increasingly committing publicly to continued HSPA investment, rather than to firm LTE rollout dates.
"These figures pay testament to the growing power and appeal of Mobile Broadband, which is enabling customers to work and socialise in a range of innovative new ways," said Steve Pusey, Global CTO of Vodafone. "Vodafone continues to upgrade its 3G/HSPA networks to offer its customers an unparalleled mobile experience by increasing average network speeds and in turn deliver even greater flexibility and convenience."
Juniper Research senior analyst Howard Wilcox also interprets the numbers as evidence that HSPA will be the preferred mobile broadband platform for some time to come.
"This is a significant achievement for HSPA and clear confirmation that it will be the dominant mobile broadband technology for the foreseeable future. Juniper Research can only see this growth increasing in the years to come, achieving an estimated 1 billion HSPA subscribers globally in 5 years' time. We see HSPA as a real service enabler, essentially replicating the desktop web experience on handheld devices," Wilcox says.
Juniper Research expects services such as social networking, music and video downloads and uploads direct to and from devices, and location-based services to really flourish with HSPA. It also sees HSPA as being a substitute for fixed broadband access in some areas.
"With the speeds being achieved we also see HSPA as a realistic alternative for fixed broadband access particularly in rural areas and amongst users who re-locate frequently. In addition we anticipate that growth will be driven by the sub 35 age range – 15 to 20 year olds now who have been brought up with a mobile for as long they can remember, broadband at home and digital multichannel TV – the "digital generation" – will expect mobile broadband as a given in future, when they enter world of work," Wilcox says.
Some analysts have been saying lately that the last thing the industry needs is another investment round in another air interface, with all the challenges that presents to upgrades of network and handset hardware and software. Those chasing the LTE spend have been urged not to overstate the market opportunity. Although rollouts are planned, they are shifting back in dates, and are unlikely to widespread, or reach large percentages of the population, for many year.
So where does that leave LTE/ SAE, the 4G standard designed to take mobile broadband from being merely an equal player with basic DSL, to something that can emulate fibre connectivity, with multi-Mbps connectivity?
Well, don't get the impression that there is in some way a lack of industry momentum behind LTE. The NGMN Alliance, an industry group led by the operators themselves, has made rapid progress in the two years since it was established. The operators are driven by a need to make this next rollout as smooth, and as cost-efficient, as possible. That means solving IPR issues before they become issues. It means ensuring interoperability and common platforms. It means making sure the technology does what it says it can do on the label, rather than promise much and deliver less.
So what progress has been made? The first field trial results have shown performance, on 20 MHz spectrum, of 170 Mbps down link speed. The demo presented by the suppliers revealed average rate per sector between 50 to 90 Mbps (for a 2-5Gbytes/ sector and user per month) with latencies below 10ms. Joachim Horn, CTO of T-Mobile, speaking on behalf of NGMN, expressed strong satisfaction with the results at this stage, saying they are "better than expected" and being confident about the LTE roadmap.
These results gave the NGMN confidence to predict that standardisation would be finalized by the end of the year with first trials in 2009, and larger deployment starting from 2010 onwards.
Karim Taga, of A D Little, said the NGMN has made better progress than any similar industry body he can think of, and he praises the efforts made to sort out interoperability and IPR issues.
Yet although investment decision have already been made and announced in Asia (NTT) and the USA (Verizon), he sees Europe presenting more of a hurdle for operators.
"Currently operators are facing an explosion of the traffic in their network, extending their geo-reach, migrating their 2G customer base to 3G etc. This means they are reviewing their access strategies -whether to refarm and deploy UMTS 900 or slow down their deployment and wait for LTE (among many options). So if LTE doesn't deliver on its promise – a 10x better performance than HSPA – or if the delivery is delayed, the risk is substantial that operators will continue to deploy on the HSPA road map and consider LTE in their next investment life cycle.
"This is particularly critical if regulators do not consider the urgency of freeing up spectrum on time (at least 20 MHz per operator to fully leverage the performance of LTE). And this risk is particularly eminent for Europe, where operators may lose a competitive edge due to the delay in spectrum allocation. The US and Asia have already conducted their tenders and network deployment in LTE/SAE are already planned, Europe is still very fragmented," Karim cautions.
The NGMN Alliance argues that operators will need access to lower frequency bands, especially to those within the UHF band (i.e. 470 – 806/862 MHz), in order to ensure that ubiquitous coverage can be provided in an economically viable fashion. Without this coverage, the operators argue, the digital divide will continue – giving cities and high value areas top level coverage, with other areas lagging behind. Of course, those UHF bands are those currently targeted for release with the switchover from analogue to digital television. It also wants access to sufficient frequency bands above 1GHz, in order to meet capacity requirements. The NGMN wish list includes, sufficient spectrum allocated within the 470 to 862 MHz band to allow multiple full deployment of next generation of mobile networks. More than 120 MHz of harmonised spectrum would be needed to accommodate commonly envisaged deployment scenarios (involving parameters such as number of channels, number of operators, and choice of FDD or TDD technology). The channelling arrangement within the band needs to be defined taking into account the possible asymmetry of traffic due to services such as mobile TV. Sufficient guard-band will be needed to reduce the threat of interference between the digital dividend services and DVB-T. The spectrum for IMT should be harmonised globally, if possible and at least on a regional basis. As an example of regional harmonization, Europe needs two contiguous sub bands, to minimise guard-band requirements and it needs to overcome limitations on specific channels. In Europe, as we have seen, such frequency allocation is rarely smooth, or, more importantly, speedy.
So the situation can be summed up like this. On the one hand, HSPA has been and continues to be great success for mobile operators. But faced with competitive pressures from rival wireless access technologies, and a fixed broadband market that will be offering ever-higher speeds, as well as a challenge to drive new service revenues, the industry knows that many of its operators will need to play right at the top of the value chain, where the ability to offer converged, broadband services will be critical. HSPA will not offer enough capacity and coverage to do that – whereas LTE could. To introduce LTE, though,the industry needs to cut its capital and operating costs far below what it has been used to for introducing a new technology. And it needs to solve a whole range of regulatory and licensing issues. Last, but not least, the industry will need to address a whole new business model, as by its very nature 4G will open up opportunities for a range of players (consumer electronics manufacturers, content providers and owners) who have until recently been at the mercy of the mobile operators. Long Term Evolution may be more Mid -Term, and it may be more revolutionary than evolutionary after all.