Wireless mesh networks in Africa

I’ve never been to Africa before, but I have many friends who have. I know that their internet access there (in Malawi, at least) could often be a lot better. The problem most likely arises from the lack of a solid data infrastructure. There are DSL solutions, but from what I’ve read they can be fairly unreliable, depending on where you are. Furthermore, Malawi often sees rolling black-outs. I don’t know exactly why they do this, but I imagine it probably has something to do with budgets.

Propagating network connections over long distances using wireless isn’t a new idea. In fact, it’s something many companies and municipalities across the world are considering in rural areas. All one needs is a pair of unidirectional, high-gain antennas pointed at each other and you can create a link between nodes miles apart. I’ve even seen experiments in the San Francisco bay area using nothing more than the asian frying scoop (round, mesh ladle of sorts) and a USB 802.11 adapter to connect to another one several miles away. But this isn’t the purpose of my post – just a little side note and something which may be a feature of my ideas later on.

What I AM thinking about:
My church, Flood, works closely with the African Bible College in Malawi. They’ve been around for quite a while, and even have a very good reputation for the quality of education they provide. But it’s only now that they’re getting broadband installed. All of a sudden they have an opportunity to connect their computer lab, their administrative offices, and their professors to a (faster) connection. The major hurdle, however, is that there’s probably no data cabling installed anywhere between buildings (or even within, for that matter). So the idea came to me when I read about MIT’s RoofNet project. Perfect! You buy a bunch of Netgear routers, flash them with a special distribution of linux and other customized software packages, and all of a sudden you have a pretty nice 802.11b/g mesh network. I could imagine littering these all over the campus, thus negating the need for dedicated ethernet links, additional switches, etc. I’m going to continue looking into this idea and talk to Flood’s resident networking pro, Seth Eaton, to see what he has been thinking of for this project. Perhaps the folks over at Meraki Networks (spin-off from RoofNet) wouldn’t mind donating some units in the perfect test location?? Seriously guys, please consider it and we can work something out =)

Maybe I’m going to Africa sooner than I thought!




1 Comment

Mesh network – yes good idea, however the problem Malawi is experiencing (and the rest of Africa) with high costs of internet access, has got nothing to do with the cost of equipment.

I deal with a local ISP/WISP in Malawi that actually provides free installations of their equipment (WiMax, yes WiMax, they deployed their WiMax network over 15 months ago)howeve the high cost comes in because of the cost of bandwidth.

On average you but 1MB syncronous CIR for $4,800.00 per month. The technology and the solutions to connect to the internet are available, it is the high cost of VSAT bandwidth which makes it unaffordable.

Leave a Reply