Spectrum re-farming: getting ready for the LTE capacity crunch

Friday, 23 September 2011 09:48  |  James Barton

With the advent of LTE, the prices of spectrum licences are escalating – but the spectrum itself is finite, and increasingly hard to obtain. Meanwhile, older technologies that have largely fallen out of use are occupying other spectrum bands – but there is a common solution to these problems. DT editor James Barton spoke to Fabricio Martinez, Services Director at Aircom, about how spectrum re-farming can reduce operator expenditure while bringing new technologies to emerging markets.

DT: Explain what you mean by spectrum re-farming, and how it can save operators money.

FM: Re-farming is the reuse of existing spectrum. Whatever spectrum operators currently own can be used to allocate capacity for a new technology, whether LTE or additional UMTS capacity – the latter is more likely for emerging markets.

As operators have already paid a lot for spectrum across different countries, it makes sense to squeeze the existing capacity and operate more than one technology. So for example if they were running a GSM network, they may be able to reduce the amount of capacity that they are using by 30 – 70%, and then allocate that newly freed-up capacity to new technologies (in this case LTE).

While it’s important that the re-farming exercise doesn’t impact network performance, migrating traffic over the existing GSM or UMTS networks will definitely represent a saving compared to acquiring a new LTE licence for the standard licensing period of 20 years.

DT: How exactly can 2G/3G spectrum be adapted to offer 4G/LTE services?

FM: We know one operator which has 10MHz of spectrum in 900 and 20MHz in 1800, as well as around 10MHz in 2.1. They are operating GSM in 900 and 1800, and UMTS in 2.1, and they want to employ additional capacity for data, and it doesn’t matter at this stage whether that is UMTS or LTE – what is important is allocating extra capacity for the new tech. To do this, we can re-farm 50% of the capacity that they have in 1800, and use those 10MHz of spectrum to allocate extra capacity for the UMTS network or for a new LTE network which will allow the operator to handle the market’s growth in data. In doing this, all customers and data can be migrated to 3G/LTE but there will still be enough capacity for voice on the 2G network.

DT: How long has the idea of re-farming been around for?

FM: Back in 2000, when the US market migrated from TDMA to GSM base stations, they had to re-farm part of their spectrum to allocate GSM capacity. Obviously most customers at the time were using TDMA-based phones, so they couldn’t get rid of this technology and had to run both at the same time without buying additional spectrum. Re-farming was the solution.

The reality is that the more spectrum you have, the more frequencies you have to plan your network, and provide capacity and quality of service. Through re-farming, you can have fewer frequencies and less capacity while still allocating good quality of service and capacity into your network by avoiding interference. If you originally have 50 frequencies – 50 channels for traffic – and this is reduced to 25, it’s now possible to use these channels to provide the same amount of capacity that the original 50 did. That’s the idea behind the re-farming process: coming up with a frequency plan and traffic management that allow an operator to handle their current 2G traffic while deploying new technologies.

It’s not a new process but it is becoming very relevant; this is due both to the advent of LTE and the increasing lack of spectrum. Previously, operators simply bought more spectrum at auction if they needed it but it’s getting to the point where there’s not that much left, so reusing it is a more appealing option.

DT: So operators are already re-farming spectrum across developed and emerging markets?

FM: Absolutely – operators in both developed and developing markets are allocating the extra capacity for UMTS. Thailand, Malaysia and Cambodia have all begun re-farming spectrum, while in Africa countries such as Nigeria, Ghana and South Africa are looking into the process. It is also becoming increasingly common in Latin America.

One potential issue is raised by regulation, particularly in emerging markets – licences need to be valid for use with technologies other than those that they were issued for; they need to be agnostic. Instead of a 2G/GSM licence, it needs to be sold as spectrum for wireless services, allowing operators to deploy either a 2G, 3G or LTE network in that spectrum band. Once the regulator has approved this, the operators must then decide whether they want look at re-farming their existing spectrum to deal with growth in data and lack of capacity.

Technology-agnostic licences would also allow operators to begin offering LTE services ahead of their competition in countries where the relevant licences have not yet been issued. Comparing the cost of an LTE licence to the cost of re-farming shows that there is a significant saving with the latter; 5MHz of re-used spectrum will last 2 – 3 years, while 10MHz will last 5 years.

DT: Do you think that regulators will be reluctant to make licences technology-agnostic, as this could lose them revenue?

FM: I think that this is about to change - not only because developing markets are running out of spectrum, but also because re-farming gives governments the capability of reselling spectrum that was previously in use. For example in Spain, through the re-farming process operators are returning 50% of their spectrum to the government, which can then be re-auctioned. If governments allow operators to be technology-agnostic, they force them to make better use of their spectrum so it’s actually a means of generating revenue. Even countries which have only recently licensed 3G spectrum - such as India - are likely to align themselves with this trend in order to better us their capacity and allocate more resources.

DT: Several emerging markets have been something of a test bed for LTE deployments – do you think re-farming spectrum for LTE will be more prevalent in these areas than in developed markets?

FM: It shouldn’t make much of a difference whether a market is more or less developed. With some of the limitations of developing markets - such as the cost of deploying networks and relatively low ARPU – re-farming represents another way of becoming more cost-effective. Rather than buying new spectrum, operators can get the most out of their existing investment.

In some developing markets, spectrum is not so well regulated and some of the frequencies that are being allocated for LTE (such as 2.6) are in use by the government or the military, which limits the options for LTE deployments. The 1800 band could represent an alternative to this.

If you look at the cost of re-farming, it’s nothing compared to the cost of a licence. It also extends the life of a network by 3 – 5 years via the added capacity. Compared to a 20-year licence, an operator would save around 20%, which in emerging markets would probably be an average of around US$300m. It’s a way for operators to squeeze resources and become more efficient.

---

Add this page to your favorite Social Bookmarking websites