Planned 4G mobile services may struggle to deliver the promised higher data rates but some smart linking of existing Wi-Fi node networks might act as effective support.
Mobile operators increasingly assume that they will sell 4G to consumers on the basis of higher data rates, and then use the combination of a new technology and newly available spectrum to increase the capacity of their network. It's true that 4G will indeed provide some additional capacity, but operators would be well advised to prepare for it falling significantly short of what is required. Mobile capacity is determined not by one single factor, but by three: the amount of spectrum; the efficiency of the technology; and the number of cells deployed.
Unlike spectrum and technology, the number of cells is limited only by the economics of deploying more. Over the last few decades most of the gains in capacity have come from more cell sites, but cellular operators have broadly now stopped deploying them.
True, there are city-centre additions to meet local demand, but the likelihood that they would each have the inclination and the resources to increase their network ten-fold, from around 15,000 cells in the UK to around 150,000, is remote. If the will exists on the part of mobile operators and other initiative leaders, however, there are alternative strategies that, arguably, could take advantage of an existing wireless infrastructure; one that would also make the most of the opportunities that access convergence can offer.
There are already millions of 'cells-in-waiting' around the country in the form of the Wi-Fi routers installed in homes, offices, coffee shops, fast-food restaurants, pubs, libraries, and myriad places elsewhere. Many already make use of these as low-cost yet high-speed data download points. In Q4/2011 Nokia started trialling a free Wi-Fi service in central London: 26 hot-spots were set up to support the service, largely concentrated around West End shopping areas; reportedly users would not need to register or sign in to gain access.
If everyone was allowed to access any Wi-Fi hot-spot without the need to enter passwords and click through log-in screens then immediately a wireless network with 100-fold more capacity than we have now, 10-fold higher data rates, would be available ' and yet with no significant cost increase.
Cellular communications would then be used for voice and a back-up for data when out of range of Wi-Fi. This model would be predicated on the near-seamless interoperability between the different networks, a phenomenon sometimes referred to as the 'telecoms cloud', but which could be called 'access convergence'.
There are already some networks that show the way toward access convergence. Users of BT home broadband can opt into its 'FON' network giving them access to any other BT home Wi-Fi cells that have similarly opted in as well as free access to BT's OpenZone network of commercial Wi-Fi hotspots. Together this numbers many millions of cells in the UK. It is not perfect because it can be tricky to get smartphones to roam seamlessly from cellular to the FON network ' but this is more because it is not in the interests of the cellular operators to enable this than any technical problems.
Sooner or later BT, one of the mobile operators, or a 'virtual' operator such as Tesco Mobile will put together a compelling package of fixed and mobile broadband with access to most Wi-Fi nodes. It will not be quite as simple as just using existing Wi-Fi nodes, of course: there will be areas where targeted coverage is still needed, achieved by a combination of operators deploying femtocells both indoors and outdoors, and by further Wi-Fi deployments.
Wireless grid to go?
A further improved approach would be a mechanism that allowed all Wi-Fi owners, whether residential, commercial or otherwise, to offer their capacity into a 'wireless grid' that would tie into the access convergence ethos. This grid would then link into existing mobile networks such that roaming from cellular to Wi-Fi and from hotspot to hotspot was seamless for the user - in the same way that cellular phones can already be programmed to roam seamlessly onto home Wi-Fi networks.
Phones would use the cellular network for voice and the Wi-Fi network where available for data, resorting to cellular data only where there was no Wi-Fi coverage. The benefits offered in return for provision of capacity into the 'grid' might vary. Domestic users offering their own capacity would in return be allowed to use that of others - just as FON works today. For branded commercial owners (nationwide coffee cafe chain Starbucks, say, or retailer Tesco) the benefit offered might be to enable some advertising. For the owners of Wi-Fi networks there might be some repayment from the mobile operators depending on the amount of data they carried, which, in turn, the mobile operators would collect through the mobile subscription.
The access convergence model does not require any new technology - but it does need someone to coordinate the 'wireless grid', allowing Wi-Fi owners to sign up and providing the necessary software to configure routers, downloads to cell phones and working with mobile operators to build the necessary relationship and contracts.
There are some players already in this space including BT FON and iPass, but a balance needs to be struck between competitive provision of small cell resources and the risk that the market becomes fragmented.
It may be that, just as governments administer feed-in to the national electricity grid, there is a role for government to facilitate the emergence of the owner of the wireless grid or at least the rules for enabling multiple operators to access the same resources. Left alone it is hard to see the wireless network providers having the right commercial incentives to get to such an end-point any time soon.
Commercial incentives form the crux of the problem. Looked at from the position of a cellular operator, wishing to maximise their revenue by keeping valuable data traffic on their network, the orthodox deployment of 4G looks entirely logical. Looked at from the position of the communications network as a whole a Wi-Fi solution appears superior.
Few people are looking from the position of the best future for a country's wireless communications infrastructure. There is no organisation in the UK charged with doing this - the regulator Ofcom is the closest the industry has to a national custodian but has no mandate for policy change whereas the government can change policy, but has put responsibility for mobile infrastructure in the UK Department for Culture, Media and Sport, a body that arguably has no history of handling issues of this kind. It seems likely that we will fail to take the turn in the road at this junction. As a result, the journey will take longer and be more expensive. Small-cell solutions will result ' there are no other options for meeting capacity requirements. *
Professor William Webb, FREng, FIET, FIEEE is a director of Neul (www.neul.com), and is an IET vice president.