Google has embarked on an ambitious $1bn plan to launch 180 satellites to provide internet access to remote parts of the world, according to the Wall Street Journal.
It seems the race is on to be the provider of choice for the enormous (and un-invoiced) emerging market, though who will advertise to it is yet to be revealed.
Kester Mann of CCS Insight told The Reg: It raises the stakes in the fight for internet supremacy in emerging markets with Facebook, which is pushing its internet.org initiative with similar ambitions. This push appears to be being driven by the leading US internet companies: we have heard few similar ambitions so far by Chinese companies such as Alibaba and Tencent.”
The project apes Motorola’s Iridium project which similarly was designed to provide global mobile phone access and filed for bankruptcy after Moto had spent $5bn on the launching the network. Iridium was sold to Iridium Satellite for $25m, in a deal backed by the US military, and is now prospering as a niche service.
Like Iridium, the Google network will use Low Earth Orbit satellites – as distinct from the Inmarsat network which uses geostationary satellites. While this requires more units, it provides advantages in bandwidth and particularly in latency as the higher a satellite's orbit, the longer it takes for the signal to be received. Reports say that the eventual number of satellites could double.
The team to launch the project is led by Greg Wyler, who joined Google from O3b Networks which has a fleet of eight Medium Earth Orbit satellites and provides coverage to 70 per cent of the world.
It is expected that Google is looking at Kymeta beam-steered antennas which are cheaper than phased array antennas and could track the vast number of birds Google is planning to launch. Kymeta already supplies O3b. Its CTO, Brian Holtz, recently joined Google as part of a major recruitment drive of satellite engineers.
What about Loons?
The venture joins prior Google projects Titian Aerospace drones and Google Loon high altitude balloon initiative as ways to reach consumers that otherwise could not get access to cat videos.
The Titian project uses drones which are expected to stay aloft for months at a time while Loon now relies on deals over radio spectrum with Google’s good friends at the mobile phone networks. Perhaps this is the least ambitious of the three and certainly more pragmatic than a claim by Larry Page at TED that Loon could cover the whole world.
Mann told El Reg: “The move feels bold. The costs are significant and may well spiral, while significant returns from low-spending customers in emerging markets are not guaranteed. Furthermore, there have been a number of high-profile failures by companies looking to deliver internet access by satellite in the past. Meanwhile, it should be remembered that the network side is just one part of the story – the device landscape is unclear. Lastly, the initiative could face major regulatory hurdles.”
Satellites – especially 180 or 360 of them – could easily cover the whole world. The issue is how – when it is providing internet access to the very poorest people in the world – Google will find anyone who wants to advertise to them. ®