June 2006

Senior product director of VoIP discusses the relative merits of wireless and fibre connectivity.

When Ofcom announced that it is preparing to initiate another round of spectrum sales, the industry sat up and paid attention. The reason for this was that the proposed spectrum - in the 71-76 GHz and 81-86 GHz bands - can offer unusually fast broadband speeds for fixed wireless point-to-point communications. If the appropriate technology is developed, this spectrum will be able to support speeds of up to 10 Gbps, leading to speculation that a true competitor to fibre-accessed broadband could emerge. 

The concept of wireless access was first raised during the early days of the Internet revolution. Since then what began as an exchange mechanism for electronic data has sparked a need for on-demand, 24/7 computing and communications. The recent advent of WiFi technology, WiMAX and wireless hotspots is only just beginning to address this need. However, despite wireless hotspots extending the reach of the Internet, users are still constrained to a geographically limited location. Meanwhile, many users want mobile access as they move about their lives, putting truly mobile wireless access and the higher frequency spectrum to the forefront of debate. 

Until now, the highest band available in the U.K. for commercial broadband point-to-point FWS is the 64-66 GHz band. Access to anything above this for FWS is currently only permitted through the provision of a non-operational licence for test and development purposes. Out of the 16 open bands in the U.K., the maximum amount of bandwidth available is small (up to around 2 GHz) when compared to the 10GHz (2 x 5GHz) of spectrum, which is the subject of this consultation. Bands below 64 GHz are also rigidly structured and defined by set channel plans, which, although providing a clear structure for deployment and co-ordination of FWS, can constrain the type and capacity of application deployed.

The proposed higher spectrum promises a range of potential applications, including very high capacity, point-to-point wireless networks, which could be used as an alternative to traditional fibre optic cables. Expected data speeds range from 1Gb per second to 10Gbps over distances of between one and two kilometres. As mentioned, the high speeds that will be made available have already led commentators to declare that the new spectrum could enable a fixed wireless competitor to take on the traditional fibre broadband connections. This does, at first sight, seem a fair proposition - a wireless network could be much less expensive and potentially faster to deliver than a fibre one. However, there are a number of important factors that need to be addressed before a true wireless competitor to fibre can emerge.

The primary consideration is financial. Considerable investment into the provision of fibre has proven that the technology can deliver high-grade business-quality bandwidth and capacity. In the event of a mass move to wireless access, both operators and businesses alike will feel strongly that, having already invested a significant amount in this technology, they may not wish to abandon it for a new raft of investments in an unproven or immature wireless technology. This point will prove to be even more relevant in the next few years as a nationwide fibre network is rolled out. Super fast and secure broadband will be available to businesses without any further investment.

We may, however, see a combination of fixed and wireless access technologies complementing each other in order to deliver on promised services. For example, many high-capacity, home-based applications such as Digital TV will need broadband fixed access which may not be feasible to deliver across a wireless network. In years to come we'll see gateways that will allow fixed/mobile convergence of services including voice, data and TV using either wireless access or fixed access. The role of wireless in this case is to expand access coverage in rural areas, which will give an extended network with high capacity transmission facilities. Due to the reach of the spectrum in question it remains to be seen if it will be viable to deploy it in rural areas where the density of potential customers tends to be small.

Additionally, a wireless network within the proposed spectrum will still require backhaul services to knit various base stations together into a network. This backhaul is likely to be fibre based and considering the bandwidths potentially being delivered to the end users, will also be significant in size.

It must also be noted that holding a spectrum sale is very different to having a fully viable technology. Before the 71-76 GHz and 81-86 GHz bands can be put to use, there will be a long process to set official standards, develop the technology and then prove it works. In the years it takes for a true alternative to emerge, fibre will have established itself as the dominant means of broadband connectivity. Fibre can already deliver speeds matching or exceeding those available in this wireless spectrum. Furthermore, fibre networks can offer a more resilient and robust connectivity than wireless. Wireless is vulnerable to environmental factors in a way fibre is not, with users of some frequencies even complaining of signal loss in heavy rain or snow. Fibre cables buried deep underground are protected from external elements and are therefore much more reliable for important communications. 

Perhaps the most salient point, however, lies in discussing who would actually want to exploit the new spectrum. Mobile operators are still suffering from investing so heavily in 3G, as Vodafone's recent financial results bear witness to. It would be a brave operator who would wish to invest in yet another spectrum, one with unproven technology and uncertain revenue streams. It is possible major fixed line operators, still sore at not having exploited the mobile revolution in the 1990s, may see the new spectrum as a means to take on the mobile operators and offer a new form of wireless connectivity. However, the majority of these have already invested in existing technologies such as WiMAX and WiFi, both of which are a great deal closer to offering fast wireless broadband. It would be surprising if any of the major operators in either mobile or fixed would rush to deploy services in this spectrum for blanket rollout. Rather it is probably larger operators targeting specific areas or smaller, niche players that will look to exploit the spectrum for small-scale, highly specialised services.

Falk Bleyl is senior product director, VoIP, at THUS.