For many of us who already enjoy the benefits of the internet in South Africa, the most exciting tech development in 2016 will be when the mechanical diggers arrive to make trenches for the cables that will bring Fibre to the Home (FTTH) broadband to our doors. In our recent readers’ survey, “cheaper, faster internet access” topped the list of things you are most looking forward to over the next 12 months with a massive 45% of the vote.
For the majority of people in the country, however, affordable and reliable high speed internet access of any kind remains as foreign as snowball fights on Christmas day. The latest statistics from the General Household Survey show that more than half of all families still have no access to the net at home: something that’s frankly shameful in sub-Saharan Africa’s richest country.
“There are 7.4 billion people on Earth,” says Professor H Nwana, “If you were to rank us all in terms of income from Bill Gates to the poorest Nepalese, the first billion has an average annual income of $49 500 (R771K) and is prepared to spend up to $205 a month on communications services including subscription TV. The second billion has $12 000 annually, and can afford $53 a month.
“By the fourth billion, it’s $3 000 a year and $12 a month,” Nwana continues, “The affordability challenge kicks in.”
Nwana has spent most of his career thinking about how to get people online. The Cameroon-born technologist currently lives in the UK, where he oversaw spectrum policy for the British telecoms regulator, Ofcom, from 2009 to 2013. During this time, he led the team responsible for both the migration to digital television (which was completed in 2012) and auctioning off licences for 4G/LTE services.
The gamekeeper has turned poacher, and Nwana currently serves as the Executive Director of the Dynamic Spectrum Alliance (DSA), an organisation which works to promote dynamic spectrum allocation around the world and advise national communications regulators on how to allow its use in their country.
Dynamic spectrum allocation, which Nwana emphasises is a regulatory term and not a technology, is a mechanism through which network operators are allowed to send and receive wireless broadband signals over radio frequencies which have been reserved and licensed for other purposes but are currently unused. It’s best known in the form of TV White Spaces (TVWS), those frequencies licensed for analogue TV but which don’t carry local stations. They’re the ones that appear as static when you’re tuning your TV, and Nwana reckons there’s plenty of them in Africa.
Indeed, he argues, precious bandwidth is going to waste in digital broadcasting as well as in traditional terrestrial channels.
“82% of African countries today have less than 19 [satellite] channels on air,” Nwana explains, “In Cameroon, I can put all 19 channels on one MUX. There will always be room for spectrum to be used on a secondary basis.”
The year in White Spaces
2016 is promising to be a pivotal year for TVWS in South Africa. The technology is mandated by the National Broadband Policy, SA Connect, and the telecoms regulator ICASA has begun the process of consultation which should eventually allow widespread TVWS in South Africa. Microsoft and Google have both run well publicised and successful trials here, whileICASA and local network firms have also been running feasibility tests, investigating how the likes of SABC are using their spectrum allocations and what could be fitted around their channels without causing signal interference.
TVWS and dynamic spectrum allocation are ways to identify unused radio spectrum and fill it with internet signals. The technologies that are most likely to be used are WiFi over TV bands, 802.11af, which is related to WiFi but designed to not to interfere with neighbouring signals, and regional broadcasting standard WRAN 802.22.
According to Microsoft’s Director of Technology Policy Paul Garnett, who just happens to be Chair of the DSA, dynamic spectrum is best thought of as a “wireless local loop”, acting like ADSL over the local telephone network in that it is used to connect fixed routers in buildings with internet points a distance away. Garnett reckons that a single 8Mhz chunk of spectrum allocated for a TV channel but not used can provide 16Mbps speeds over a 10KM range.
The most common argument in favour of TVWS and dynamic spectrum allocation – which can apply to any to any licensed spectrum, not just those used for television – is that it provides a way of getting low cost internet access to people in rural areas or places where current providers are failing to serve the market.
Infrastructure investment is minimal, because there’s no long distance cabling required and a TVWS base station is is much cheaper than an LTE tower to set up. According to Microsoft, the networking equipment for its Limpopo trials, which connected six schools in the area, cost just $52 000 (around R800 000) for the whole project including routers at the schools, with almost the same again in licensing and approval charges. LTE base stations cost between $10 000 and $100 000 each depending on size and range, and are more complex to install.
Furthermore, DSA operators shouldn’t be subject to licence fees to operate since the spectrum they use has already been paid for by other services which are simply not using it.
Nwana says that focussing solely on the deployment costs misses the strongest argument in favour of dynamic spectrum: it’s not the technical solution that should be debated, it’s how to bring market forces to bear on the problem of affordable connectivity.
“As an ex-regulator the question is that if you have that sort of problem what do you do?” Nwana says, “And the first answer any good regulator will say is throw competition at the problem.”
TVWS and dynamic spectrum is just one way of promoting competition, he says, because it allows new companies to come in without the legacy of a phone network and shareholders to support. He compares it to the effect of allowing mobile operators to compete with fixed line telecoms firms two decades ago.
“Going back, there were less than 4 500 fixed lines in Nigeria [Africa’s most populous country – Ed] 25 years ago,” Nwana says, “You threw competition at the problem, allows the mobile networks allowed to compete, and now all Africans have a phone. The same principle applied to the broadband network will have the same effect. Competition must be allowed through new techs like TWVS.“
Happy to co-exist
Nwana says that opposition to TVWS and dynamic spectrum from the likes of the mobile industry organisation GSMA is uncalled for, and that many different forms of access technology will be needed to keep market forces working and reduce the cost of access.
“It’s not a model of one winning over the other,” he says, “My job is to defend citizens and provide access and affordability sooner. That means promoting competition from within the existing ecosystem and outside.”
Nwana is optimistic about the potential for TVWS in South Africa, and given the lack of progress made on digital migration – the process by which analogue TV transmissions are turned off and the spectrum resold for LTE/4G licences – and the desperate shortage of available spectrum for high-speed wireless broadband, it should come as no surprise that Nwana and his team have been consulting with ICASA around the issue. Indeed, he believes that once its use has been established, using free frequencies for unlicensed broadband will be common well after the digital migration takes place.
He does, however, sympathise with ICASA and says that we shouldn’t expect anything to happen too soon.
“Effectively we’re introducing disruption into the ecosystem,” Nwana concludes, “And most regulators are cautious about that and want to proceed carefully. There’s no need for us to do it too much haste.”[Main image – CC 2.0 BY Anthony Quintano, other images supplied by DSA.]