Reaching the other three billion
By Peter Guest | Published: 15 January, 2009
Telecoms entrepreneur Greg Wyler is well aware of the problems with putting telecoms cabling under Africa, having been among the first to endeavour to bridge the vast hinterlands that reinforce the “digital divide” and give African consumers access to modern telecoms infrastructure.
Now chief executive of O3B networks – so called because it aims to supply high speed communications and to the “other three billion” currently underserved by the existing networks, Mr Wyler explains that, despite a number of recent moves to build submarine cables supplying Africa with international capacity, the problem does not end when the cable hits the beach.
“Africa suffers from something – the Mercator Projection. We see Africa as smaller than it is physically, so running fibre across Africa is a rather daunting task,” he warns. Given that the United States is roughly a third the size of Africa, he argues: “If we said we were going to put fibre in Florida, and you were in Seattle, you’d say, ‘that doesn’t affect me.’ Even if you were in Georgia you’d think you were pretty far away. And that’s the problem with Africa. Just one spot is not going to give the coverage on the edge of the network all throughout the continent.”
The problems with distribution are manifold. Some routes would be “thin” – i.e. the regions they cross would have very little demand and take very little capacity and across such enormous distances the maintenance and repair bill would be so colossal as to render it uneconomic. Furthermore, there is the concern that the development of a continental wholesale fibreoptic cable is in some respects the same as that of an oil or gas pipeline, and by crossing national boundaries introduces the same political risk – that there is an imbalance of power towards those further up the chain who are by their geography able to restrict supply to those downstream.
O3b’s approach is to use satellites, which will beam capacity from the hardwired networks in Europe. This in itself is hardly revolutionary – satellite communications have been used across Africa for years – but the existing constellations are not geared for high speed internet.
The problem, Mr Wyler says, is that the height of the geostationary orbits currently serving the continent is such that the time lag introduced when bouncing a signal from them makes the interactive media of the new web sluggish or unusable. “They’re so far away that they have over half a second of latency. That makes the internet applications – interactive applications – not function at all. So what we’ve done is brought the satellites a lot closer to Earth, so they have about a tenth of a second latency. Fundamentally that means it’s snappier, websites will pop up like they do here.”
Mid-Earth orbit satellites provide a viable alternative, but, as with fibre, the Mid-earth orbit satellite network was built to serve the developed economies of the Northern Hemisphere. A map of their trajectories shows them orbiting in dense ball of twine patterns of 40 or more satellites. A constellation of that size is expensive. The beauty of O3b, Mr Wyler explains, is that they do not need 40. They need five.
The reason is simply that the company will be supplying capacity to historically underserved equatorial markets. To ensure continuous coverage over the northern reaches of the earth, the satellites’ orbits must be inclined, which means that as the Earth rotates underneath them they are in line of sight of any given point on the ground for much less time, hence more vehicles are needed.
By putting a satellite constellation around the equator, O3b obviates the necessity for many intersecting orbits. This makes the system a lot cheaper than its predecessors. Versus cable it is quicker to deliver to the user – requiring a relatively cheap base station on the ground – and scalable, as to increase capacity the company needs only to add another satellite to the ring. The relative proximity of the satellites to the Earth versus geostationary satellites also reduces their power requirements. Microwaves, like light, are subject to the inverse square law in signal transmission, so increasing the distance from the target greatly increases the required intensity of the signal.
Mr Wyler hopes that O3b will be used as a primary method of distributing capacity for operators, who can then resell that capacity, or for corporates looking to build private networks.
It may all sound like another Nigerian Space Programme, which made headlines with the launch of its first satellite in 2007 but was subsequently forced to shut the orbiter down eighteen months later. The fast talking Mr Wyler seems the epitome of a salesman, but the O3b project has an impressive roster of backers, including HSBC and the internet giant Google.
The first launch will be in 2009. “We’ve got a lot of interest and some commitments,” Mr Wyler says. “We don’t have a lot of capacity in the first launch, and we’re expecting Africa to be sold out pretty quickly.”
