I’m quite late in posting about “a great blog post”:1 by “Sascha Meinrath”:2 (one of the leaders of the “Champaign-Urbana Community Wireless Network”:3) about the costs of deploying a mesh-based wireless network.
Sascha’s calculations are based on a rural/suburban environment, which is where most of his experience has been. His cost numbers are quite impressive. For a network of reasonable size, using an open source solution like CUWiNware (Sascha’s open source mesh wireless OS) is *cheaper* than commercial alternatives, even after you include deployment, maintenance, and ongoing costs.
In a city environment, I suspect the numbers would be a little different, but the end result would be the same. First, you’d need a greater concentration of wireless nodes, since tall buildings would cause the wireless signals to degrade faster. Second, you’d need to have a two-pronged deployment strategy since getting wireless connectivity *to* a building doesn’t get wireless internet service *through* a building, to all of the apartments. This is where a mesh network where each mesh node also acted like a hotspot (something CUWiNware doesn’t yet do) would be helpful. You could extend the mesh throughout a building using the same nodes that are used to connect outside of the building.
Regardless, in an urban environment an open source mesh network would be even more important since it would allow people to join and extend the network without any central coordination.
[1]http://www.saschameinrath.com/2005_07_07_08_37__update_2_open_source_open_architecture_muni-wireless_costs_aka_–_how_non-proprietary_wireless_is_more_cost_ef
[2]http://www.saschameinrath.com
[3]http://www.cuwireless.net
Filed under: Community Wireless, Mesh
