Bob Cringely has an idea:
I’m preparing a number of experiments using a Linksys WRT54GS 802.11g router mounted on the roof of my house in Charleston, SC. The router is running Sveasoft firmware that gives me direct control of transmitter power if needed. I’ll be testing it with a variety of antennas, mainly omnidirectional but perhaps I’ll try a sector antenna, too. The other end of the wireless connection will be a second WRT54GS mounted in my funky homebuilt airplane. Operating the pair of routers as an Ethernet-to-Ethernet bridge will give me much higher performance than I could ever expect from just using a WiFi card in my notebook computer. And I’ll be testing a variety of antennas on the belly of my funky airplane, too.
What I expect will happen is I’ll be able to fly near Charleston and connect to my home network at 1-2 megabits-per-second. I’ll learn which antennas work best together. I may even place or receive a few VoIP phone calls using phone software running on my notebook computer. It will be interesting to see how far I can get from Charleston before losing service. It will be VERY interesting to see whether I can connect to WiFi base stations other than my own. I’m guessing such war flying will be possible, though I can’t guess how reliable.
Now — strictly because I am twisted this way — let’s take this experiment a step further. Sveasoft supports mesh networking, though with a practical limit of three hops. Aerial WiFi links of 10+ KM ought to be possible and maybe a LOT longer. The hardware cost of a WRT54GS and antenna are on the order of $100. There are, at the moment I am writing this, more than 1,000 small aircraft flying on IFR flight plans in the U.S. So for not very much money you could have a 1,000-node aerial mesh that could serve not only airborne but also terrestrial users. Triple the money, and you could put in each plane a Locustworld mesh with two radios for each node and truly robust mesh networking.
There are 25,000 mobile phone cells in the U.S., but AirCell pretty much duplicates that network using 115 specialized cell sites aimed skyward. While WiFi ranges are less than cell ranges, a 1,000-node mesh is huge — effectively bigger than any hotspot network now in use. There have long been proposals to park Internet hubs in the sky over major cities in balloons or circling aircraft, but the problem is always the cost of running that satellite in the sky. This solution eliminates that cost of operation, replacing it with a symbiotic relationship where aircraft owners benefit from volunteering the use of their planes by getting free airborne Internet service.
ZDNet has more on wireless mesh networks.