Uber tech blog GigaOm links to an article in news.com which talks about an effort to make internet access a reality in underdeveloped regions. This project named Green WiFi is being run by a couple of employees from Sun (one former and one current). They have developed a prototype with a grant from the $100 laptop project and are now getting ready for a field deployment in Uttar Pradesh.
At a very high level, enabling internet access comprises of 2 distinct challenges. Having a high bandwidth broadband backbone in place and getting people access to this backbone - the so called last mile problem. The wires that cable companies put, the phone lines that dsl providers use and the wi-fi access that hotspots provide all address the last mile problem. Simple WiFi won't be able to solve the last mile problem for Indian villages are they are usually no where near a broadband backbone. So the Green WiFi project is going to use a series of routers - each connected wirelessly through 802.11b/g (wifi) - to connect the end user to the backbone and hence the internet. Their innovation is that they are coming up with a cheap and robust solar power charged battery set in order to provide power to these routers and more importantly they are coming up with a way to do elegant degradation when the router starts running out of power.
""What we bring to the table is an intelligent charge-controller. We put the router on a diet," Pomerleau said. The controller sits between the battery and the router and regulates power to the router depending on the charge level of the battery and the amount of incoming sunlight."
Absolutely critical for this idea to succeed in Indian villages.
The pilot project in Uttar Pradesh would be able to flesh out the technical abilities and limitations of this model. But the possibility of widespread deployment would depend on how far the internet backbone is from a typical Indian village. The cost of the router has been capped at $200 (Rs. 9000). The news.com report says that the wifi router nodes can be 1kilometer apart. I am not convinced about that being true in real deployments. But anyway quibbles aside, greater the distance from the backbone more the number of routers that have to be used and more the cost. WiMax which has a longer range would be a better fit for this kind of an effort. But we won't be seeing commercial WiMax before late 2007. WiFi is already here. So if this model can bring high speed internet access to atleast those villages which are close to the backbone, that would be a great achievement.