According to 2012 estimates, internet penetration in Africa has reached 15.6%. Though the actual number of people on whom the internet has an impact is undoubtedly much higher, this statistic does demonstrate a significant infrastructural disparity between Africa and other continents. Currently, this connectivity gap is being filled by other media - if the astonishing growth rate in the African mobile market is anything to go by. And then, there is radio and television. In comparison to the all-singing, all-dancing Internet, these older media seem rather simplistic: nevertheless, radio especially remains one of the cheapest, most versatile and most widespread forms of mass communication there is - and a powerful tool for African educators, despite its lack of interactivity. But what if there were a way of combining the pervasiveness of radio with the enhanced connective power of the internet?
A recent article by Mawuna Remarque Koutonin, editor of Silicon Africa, went so far as to suggest that we should "Forget the Internet - education through radio is bigger, more affordable!" He makes a persuasive case for radio as a partial solution to the infrastructure problems besetting the Continent; existing broadcasting networks certainly provide an essential medium for the transmission of information, at least until the internet cables have caught up. For when we talk about broadcasting we are talking about infrastructure that is often already in place, and - in contrast to cables - covers wide areas all at once. To capitalise on these advantages, American companies began in 2007 to experiment with using radio waves to improve access to the Internet - specifically, by tapping the potential of unused areas of that part of the spectrum usually used for television.
The radio spectrum, over which television and radio are broadcast, is carefully divided up in most countries and allocated between different users to avoid interference between channels. This means that in between the parts of the spectrum that are used there are often "white spaces" to be found - frequencies whose only use is as a buffer between different stations. As television services go digital, buffer areas in the ultra-high frequency range (300 - 3000 MHz) are no longer required - and large swathes of the spectrum are freed up for other uses, including the provision of wireless broadband. Using broadcasting technology, it is possible to provide internet connectivity through the ether.
The infrastructural advantage white space technology offers is that it propagates waves in the same way as television. Signals from a single base station travel up to ten kilometres in all directions - covering an area of over 300 square kilometres. The waves can penetrate walls and obstacles; when transmitted and re-boosted over long distances they are capable of covering huge swathes of land for a fraction of the price of laid cables. What's more, the towers that diffuse the signal need only solar power to function, while the end users need no direct grid connection at all to receive it.
Microsoft began trials to test the viability of white space technologies in Africa this February. The first recipients were schools, healthcare centres and libraries situated near the small town of Nanyuki and settlement of Kalema in Kenya's rift valley. Next, university students in Dar es Salaam, Tanzania, were connected; and most recently, at the end of July, Microsoft chose Limpopo province, one of the poorest parts of South Africa, for its third initiative, using the University of Limpopo to spread broadband access to remote schools in the region, also providing laptops and solar panels for charging where there is no access to electricity.
These trials have demonstrated the wide-ranging versatility of white space broadband. Their success shows that rural development need not be a slow matter of connecting Africa to electricity and internet cables metre by arduous metre. The future can, and must, arrive more quickly than that, if Africa's infrastructure is to keep up with economic growth rates - and white spaces may well be a big part of the solution.