Mallinson: What is holding the UK back on LTE?

Competition and choice are supposed to be good for consumers. Instead, market conditions and conflicts are causing the United Kingdom to suffer from one of the most lacklustre mobile broadband progressions in the developed world. 

UK authorities pride themselves on having a large number of national competitors in mobile communications with Vodafone, O2, and 3 operating entirely independently, T-Mobile and Orange a merged company called Everything Everywhere retaining separate brands and stores, and a variety of mobile virtual network operators, notably including Virgin Mobile. This encourages big marketing spends, large handset subsidies and forces tight profit margins, but network investment is poor and rather limited to where demand is highest.  Squabbling over the rules for spectrum allocation is repeatedly delaying the "4G spectrum" auctions including urgently needed 800MHz and 2.6GHz bands.

Freeing up new and old spectrum for LTE

There is no prospect for LTE services launching in the UK on new spectrum until the latter part of 2013. This means the country is lagging leaders in Scandinavia and the United States by more than three years. For example, Verizon Wireless' LTE network now covers more than half of the United States--160 million POPs--and the company is on track to cover its existing 3G CDMA2000 EV-DO footprint with LTE by the end of 2013. Allocating additional spectrum in Europe for mobile broadband has been problematic and matters are getting worse in the UK.  The UK's telecommunications regulator, Ofcom, held a first consultation on its proposed auction this year, but it was deluged with feedback and complaints from most mobile operators. Ofcom had initially been hoping to hold an auction for the new spectrum in the first quarter of 2012, but the schedule was delayed to the first half of 2012. Now, due to the second consultation, the auction will not be held until toward the end of 2012.

Spectrum currently licensed to mobile operators for GSM and will eventually be refarmed for LTE. With 900MHz earmarked for UMTS and many devices on sale already supporting that possibility, it seems most likely LTE will have to go into the 1800MHz band. However, regulatory and practical constraints make it unlikely the latter will occur any sooner than putting LTE into new spectrum.

The path to 4G is potholed

Current mobile broadband capabilities are poor in many places due to deficiencies in coverage and capacity. Initially, users were delighted to obtain fast data services even on a limited basis, but expectations have risen substantially with widespread adoption of smartphones, following more modest uptake of mobile broadband on PCs. Although mobile broadband coverage was poor at the outset, at least networks were not capacity constrained then.

Growth in mobile broadband, with demand doubling each year, requires corresponding increases in network capacity including access, backhaul and core networks. Many more frequencies are urgently needed in addition to spectrum conservation techniques already being implemented including bandwidth compression, advanced modulation (e.g., 64 QAM), sophisticated antennas (e.g., with MIMO), cell-splitting and load balancing where traffic flows are highest. The 2.6GHz spectrum is ideally suited to increasing radio access capacity in traffic demand hotspots.

Coverage is also still particularly lacking. Mobile broadband was initially focused on densely-populated areas including airports, railway stations, coffee shops and hotels. With increasing demand, network performance has deteriorated due to congestion as users contend for the limited capacity, and those located at 3G cell edges fare particularly poorly with interference effects.  This may even give city users the impression there is a coverage problem--ask any frustrated urban iPhone user--when, in fact, the difficulty is that their transmissions are drowning in those from other users nearby.

Outside built-up areas matters can be even worse. During my regular rail journeys on a very busy route between London and the UK's south coast, my smartphone and laptop's embedded modem spend most of their time with EDGE, GPRS or nothing, rather than a 3G connection, because Vodafone's UMTS was deployed to boost capacity where traffic levels are high, rather than to complete coverage.  It is quite typical, and thrifty, to lay new radio access technologies on the same network grid, including masts, as for the old technologies. The problem with that approach is that radio propagation tends to get worse at higher frequencies: adding GSM 1800 to a GSM 900 grid, or adding UMTS 2100 to either a GSM 900 or GSM1800 grid creates a Swiss-cheese effect of coverage holes.  The 800MHz band would be particularly useful in creating an umbrella of coverage for mobile broadband with LTE because this can be provided using a grid of relatively widely-spaced masts, such as those implemented in 900MHz deployments. Germany wisely prioritised deployment of rural LTE in its spectrum licensing to provide broadband service to homes in "white spots" where DSL service was inadequate or too costly to operate.

Be careful what you bid for

The UK needs a massive new infrastructure upgrade with LTE, but the market is too weak financially to do the job properly. The 3G auctions in 2000 were cunningly constructed to raise a staggering £22.5 billion from five national licensees. The government found plenty of worthy causes for this money, but not penny of it was spent on building mobile networks. The train commuting 3G coverage problems described above result from skimping on 3G network upgrades in the aftermath of those 3G auction excesses that left little in the kitty for network investment by operators. Even the €3.9 billion Italy just raised in auction for 4G spectrum is a major financial burden.  Network sharing in the UK is providing some cost savings that are helping boost economic efficiencies, but with political resistance to full-blown consolidation and the ongoing regulatory squeeze on termination charges and roaming fees, UK operator EBITDA margins for major operators including Vodafone (29 per cent) and Everything Everywhere (20 per cent) are a long way below those in nations where retail and network market share is more concentrated - including the United States (45 per cent for Verizon) and Italy (46 per cent for Vodafone).

Weak financial performance makes it impossible to justify strong investments. One way or another, UK consumers are most likely doomed to have inferior 3G and 4G networks for the rest of the decade. However, if the 4G auction is not delayed indefinitely, if the field of desperate bidders is kept as broad as it is today and if maximising auction receipts continues to trump network development policy objectives, citizens might benefit eventually instead, with proceeds protecting welfare pay-outs, public sector jobs or saving some QUANGOs from the axe.

Keith Mallinson is a leading industry expert, analyst and consultant. Solving business problems in wireless and mobile communications, he founded consulting firm WiseHarbor in 2007. WiseHarbor is publishing an annual update to its Extended Mobile Broadband Forecast in May 2011. The new forecast will include network equipment, devices and carrier services to 2025. Further details are available at: Find WiseHarbor on Twitter @WiseWarbor.