Blog Post

Microsoft follows Google into South Africa to prove white space broadband potential

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

Microsoft(s msft) has launched a new trial of so-called white space broadband in the northern South African province of Limpopo. This adds to Microsoft’s existing “4Afrika” white space trials in Kenya and Tanzania, and Google(s goog)’s similar program in Cape Town, at the other end of South Africa.

White space broadband is an experimental technology that might end up bringing connectivity into rural and poor urban areas around the world – as long as local governments play ball by allowing the re-use of the relevant radio spectrum. This is one of the main drivers behind Microsoft and Google’s pilots in Africa: to lobby the authorities by proving the benefits of white space broadband.

Buffer zones

The technology works by using the buffer zones or “white spaces” between TV channels in order to carry data. These buffers exist to stop the TV channels from interfering with one another – don’t forget that TV is broadcast at pretty high power. Because white space broadband can be transmitted at relatively low power, the idea is that it can exploit these buffer zones without itself interfering with the surrounding broadcasts.

Microsoft and Google are, however, doing slightly different things in their trials. Google’s involvement extends mainly to its new spectrum database, which is a way of telling white space broadband client devices which frequencies are available for exploitation and where – this is why that trial is taking place in highly-developed and mountainous Cape Town, where the complexity of the local TV broadcast setup provides a great worst-case scenario for technical testing purposes.

Microsoft, on the other hand, is funding trials that demonstrate the commercial viability of white space broadband – the South African government wants 80 percent of its citizens to have broadband access by 2020, and Microsoft wants to prove that white space is the way to go.

Country by country

The Limpopo trial, which aims to connect local schools, is similar to Microsoft’s Kenyan pilot, in that it targets very rural areas that may not even be on the electricity grid (the Tanzanian pilot was more urban, dealing with high-density, low-income areas). The Limpopo pilot involves solar-powered base stations and – Microsoft being Microsoft – each school also gets a range of Windows tablets for pupils, laptops and training for teachers, projectors and teaching materials.

Now, about that commercial viability. According to a ZDNet report, Microsoft is predicting white space connectivity of up to 4Mbps at a cost of between 20-50 rand ($2.04-$5.10) a month. As regular broadband at speeds up around 1Mbps cost ten times as much, that’s nothing to be sneezed at (although there are signs that South Africa’s ludicrously overpriced and underspecced broadband may soon improve, thanks to the breaking-up of Telkom’s monopoly).

It doesn’t appear that what Microsoft is doing in Limpopo Province is that much different from what it already demonstrated in Kenya this year. What’s important, though, is for the company to convince the South African regulators to allow the legal use of TV white spaces for broadband purposes, and that means setting up a local pilot. We can no doubt expect more of these in other African countries in the coming year or two.

In the end, for both Google and Microsoft this all comes down to wanting to spread connectivity, and therefore those companies’ addressable markets. That said, this connectivity will also have major benefits for the economies of the countries concerned, so everyone should do well out of it.

4 Responses to “Microsoft follows Google into South Africa to prove white space broadband potential”

  1. Thanks Gertjan for raising an Important issue. Multinationals like to shine with the ministries, disregarding the local R&D institutions, from where they get the solutions and information. Most important in this regard should be the build up of local R&D capacity, to develop the technology of future broadband networks.

    John Dube,WiTS

    • Anonymous

      Agreed John and Gertjan. There is a growing need for Africa to build technology suited to its own environmnet and then even export this to other countries – Ushahidi comes to mind or even pay as you go. In the White Space domain South Africa and other African countries have a number of very unqiue aspects. Ubundant white space (with many adjacent channels available for channel aggregation), An urgent need for spectrum in UHF bands due to many low density rural populations that need connectivity, in the early phases there is likely to be a mix of analogue and DTV which needs some different protection regions in side bands. I predict that most of Europe will use White Spaces for machine to machine and to some extent urban USA may be similar due to limited white space and saturation of alternative broadband connectivity – rural USA looks promising – but the political barriers in the USA are huge with many senators complaining that people are throwing away valuable spectrum.

      I sincerely believe that South African research institutes and developers can eventually lead in this space due to the suitability of the technology to Africa and companies like Google / Microsoft / spectrum bridge / Neul etc. can come alongside or and put their trust in Africans and fund their research and field studies. This is not a Microsoft or Google story – this is a South African story with institutions like ICASA, research institutions like the CSIR, South African universtities like NWU, UCT and the South African government pulling together because they beleive this has huge potential and then running some field trials. They are happy for Microsoft or Google to also enter the playing field and try out ideas with them – because at the end of the day companies like MS/Google benefit from more connected people in Africa.

  2. Gertjan van Stam

    What about South Africans? I do not find any reference to what South Africans are doing? The Council for Scientific and Industrial Research (CSIR) in South Africa is one of the leading scientific and technology research, development and implementation organisations in Africa. It undertakes directed research and development for socio-economic growth. What is this authority saying?

    • David Meyer

      You are quite right – I should have noted that the CSIR is involved in Microsoft’s pilot (and Google’s for that matter). Other local participants in the Limpopo trial include the University of Limpopo and network firm Multisource.