Battle of the bands

On April 12th this year Germany became the first country in Europe to start the award of the 800MHz spectrum—the so called ‘digital dividend’. While amounts bid did not reach anything like the dizzying heights of 3G licensing that we witnessed in 2000, it should prove to be a valuable band. All eyes will be on Germany, as over the next few years other national regulators around the world are set to follow and release a significant amount of some of the most valuable spectrum that exists. How much will be released in each country and what it will be used for are key questions on everyone’s mind.

Ovum has recently developed, and will continuously update, an internal database of spectrum awards around the world to help answer these questions. To kick-off our new research in this area we have looked at what spectrum is likely to be used for LTE services and where in Europe these awards are taking place.

We can’t see it, touch it, or hear it, but radio spectrum is central to our lives. It is the essential raw material for all mobile services and, like many raw materials, it is a finite and therefore extremely valuable resource for which demand is increasing all the time.

As consumers demand broadband on the move and governments outline ambitions for ubiquitous broadband, more capacity will be required to support the huge demands placed on the network. Eventually, network operators will need to upgrade their infrastructure to nextgeneration (or 4G) technologies, meaning that more of this valuable resource will be required.

There are currently two main competing technologies that will be deployed for next-generation mobile services: mobile WiMAX and LTE. Each has its own benefits and the choice in many cases will come down to local market circumstances.

From a spectrum policy point of view, however, there is a key difference between the two. Mobile WiMAX is currently TDD based (using unpaired spectrum) whereas LTE will primarily be FDD in the short to medium term and uses paired spectrum. The TDD version of LTE, TD-LTE, will appear in Europe later on. By deciding how spectrum is packaged, when it is released, and the conditions attached to its use, regulators and policymakers have the ability to influence which technology is deployed.

However, it is not for national regulators to pick winners; that is something that the market must decide. Instead they must concern themselves with getting the enabling factors right. These include taking the necessary steps to ensure that adequate spectrum is available and packaged in such a way so that either technology can use it. Crucially, they must also design and structure rollout obligations in a way that complements the spectrum being offered and encourages bidding.

The auctions will be more sober affairs than the bidding frenzies that characterised the 3G spectrum auctions in the early part of the last decade, but regulators will still be keen to ensure that spectrum is awarded only to those who value it most. Only then can governments and regulators work towards creating an environment where the availability of mobile broadband is increased, and where it can play an important contributory role in achieving near-universal broadband.

While the common principles of technology and service neutrality will apply, a certain amount of harmonisation is both desirable and necessary. Spectrum policy is one of the few remaining areas where there is still a heavy-handed regulatory approach. Recognising the importance of market-based principles is a prerequisite for effective regulatory policy and will eventually dictate when auctions take place, how much spectrum is made available, how it is packaged, and the rules around how it should be used.

Slowly but surely, national regulators are implementing market mechanisms in the belief that the best way to secure the optimal use of spectrum is to limit regulation to the essentials, for example preventing harmful interference and ensuring fair competition. However, spectrum does not respect national borders and so, if one country uses a particular band for one purpose and a neighbour uses it for another, problems will quickly develop. For this reason, harmonisation between countries is encouraged and also serves to increase the size of the potential market and scope for economies of scale for equipment manufacturers.

The rationale for a coordinated approach is largely derived from the fact that next-generation technologies require large blocks of spectrum (either operated by a single party or multiple parties working collectively with contiguous spectrum) for their potential to be fully realised. Best performance for LTE FDD is achieved using 2 x 20MHz blocks, and truly national high-capacity networks will require spectrum at both low and high frequencies. Addressing these requirements in an integrated way, if achieved quickly, should give operators greater certainty over their future spectrum holdings, while continuing to support a competitive market outcome.

The digital dividend (800MHz) and 2.6GHz bands are set to be favourites for LTE. Taken together, they represent a very significant increase in the amount of spectrum that will be available for mobile broadband. It is here where operators will be looking to deploy next-generation technologies, in particular LTE, as no profile has yet been defined for mobile WiMAX at 800MHz. It is where all the efforts around standards and the efforts of the vendors in terms of developing equipment is being focused, not least because the existing 900MHz and 1800MHz bands are heavily used by existing GSM services and, before you are able to reuse those frequencies, you have to clear them of existing users. In the longer term, these bands can be refarmed, but at the moment there has been limited interest.

The 2.6GHz spectrum band, which has been identified globally by the ITU as the “3G extension band”, will be vital in satisfying the demand for greater capacity for mobile broadband and supporting networks such as LTE, which have already started to be deployed commercially around the world. However, progress so far has been slow, with the 2.6GHz band having been awarded in only a handful of countries (Finland, Hong Kong, Norway, and Sweden). That is set to change in 2010 with at least another seven countries (Austria, Denmark, France, Germany, Netherlands, Portugal, and Spain) scheduled to auction this band.

While higher frequencies are particularly suited to significantly increasing data capacity in high-demand zones such as dense urban areas, in rural areas the costs associated with deploying a network using this band are mostly prohibitive. The natural propagation characteristics of spectrum mean that, at lower frequencies, airways travel further and can penetrate buildings well.

It is for this reason that the digital dividend—the spectrum that is freed-up on switchover from analogue to digital TV—is also of interest. As such, in the medium term, we expect the digital dividend frequencies to be used for wide-area and in-building coverage, achieved with a small number of base stations, complemented by the 2.6GHz band to achieve a good amount of capacity for large numbers of end users in dense (urban) environments.

On the whole, the 2.6GHz and 800MHz bands will be the most popular, but in some instances (such as in Finland, France, and the UK) operators are considering re-farming existing bands—notably the 900/1800MHz bands. However, both these bands are heavily used by existing technologies and clearing them will not be straightforward (the UK provides a good example of the challenges that will be faced). As such, LTE will coexist with legacy mobile network technologies for at least the next ten years, particularly since GSM will still provide the backbone of voice communication.

Matthew Howett is a Senior Telecoms Analyst at Ovum

Leave a comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.