Sunday, September 14, 2008

Space offers high-speed net access to Africa

Greg Wyler, a US technology entrepreneur seeking a worthwhile project after selling his semiconductor cooling devices business, found himself in Rwanda two years ago, trying to bring internet access to one of Africa’s poorest regions.

After rebuilding a communications tower at the top of the 15,000-ft Mount Karisimbi and laying a 450km fibre network to schools and universities, Mr Wyler realised that the problem was bigger than just connecting people’s homes to the internet, it was more an issue of a lack of a telecommunications network to link the homes to.

“It’s easy to build a WiMax [wireless data transmission] tower, but if you’re 5,000 miles from a fibre network, you can’t provide fast internet access,” he says.

The solution was to move on from this network in Rwanda to an altogether more ambitious effort to rethink the telecommunications backbone in markets where a comprehensive fibre network will not make commercial sense for many years, if ever.

What started as a series of Powerpoint slides only began to gain momentum a year ago, when Mr Wyler caught the attention of John Malone’s Liberty Global cable group, HSBC’s private equity arm, and Google, the US search engine group.

Since then, the three partners have injected about $20m each, and raised a smaller sum from Allen & Co, the media investment boutique, to bring the project to a stage where it is ready to announce an order for the satellites that will form the network.

The partners have been attracted by a confluence of technology and demand which has made viable Mr Wyler’s vision of using satellites instead of fibre to provide the “backhaul” connection between telecoms core networks and the cellphone towers springing up in even the remotest areas of Africa, Asia and Latin America.

O3b satellite

Existing fibre networks are only viable where there is sufficient demand in the end location, making them useless for most rural areas. “In places like Africa, the cost of bandwidth is 20 times what you pay in the US,” notes Larry Alder, product manager in Google’s alternative access group.

Most commercial communication satellites are in geostationary orbit, with fixed beams, which limits signal strength and bandwith. Established satellite operators such as SES Global and Eutelsat, which rely on geostationary constellations, have focused mainly on developed countries.

The 16 satellites with which Mr Wyler’s O3B Networks will launch are closer to earth, thereby providing a stronger signal than existing commercial satellites, and use rotating beams that can be targeted to focus bandwidth on the areas where it is needed. Unlike some other satellite communications pioneers such as Iridium and Inmarsat, which have no mass customer base, O3B Networks will not attempt to deal directly with consumers.

Instead, it will sell capacity on its satellites to telecoms companies that are building out networks of mobile communications towers to meet rising demand in emerging markets.

The growth in the number of mobile phone users in emerging markets is high. In Africa, for instance, annual mobile customer growth is close to 60 per cent, underscoring Mr Wyler’s confidence that there is “tremendous” pent-up demand.

If successful, O3B Networks could sharply reduce the cost of providing high-speed data services in many countries. Leased lines can represent up to a quarter of mobile operators’ costs at present, and their margins are already being squeezed by rising traffic volumes.

Google says it will also accelerate the pace of internet adoption in such markets. “This unleashes a lot of entrepreneurs to build local networks,” says Mr Alder.

To succeed, the partners must raise another $700m. It could face competition from established operators, such as Globalstar, which offers satellite phone services direct to consumers in many parts of the world and will upgrade its offering using similar satellites to O3B. But if the plan works, says Mike Fries, chief executive of Liberty Global, “it seemed to us like one of those rare chances that could go from zero to 60 very fast”.

Blogged with the Flock Browser

No comments: