ETC News
October 04, 2004
Recently ETC performed a quick site survey at the local University campus to deterine our mesh networking node's range. In this test only one node was placed at a central location (Second floor window of Cass science building), and no special antennas were used. We walked campus using a laptop and another identical mesh node powered by battery. In a true mesh deployment there would be significantly more nodes, but what we have demonstrated is that significant coverage is possible with only one node. A many noded mesh network installation would increase the robustness through network path redundancy.
To help determine the building signal penetration given their brick construction, we entered the lobbies of a number of the campus buildings. As expected the "deeper" into the building the more the signal faded. However the signal remained strong within the lobby suggesting a node placed either inside or just outside the building would sufficiently re-boost the signal for in-building use.
In the image above (click to zoom) the red circle represents the approximate limits of useful transmission coverage, while the area inside the green circle represents premium quality coverage. With planned antenna placement (i.e. not just stuck in a window) we would expect these limits to stretch. Care has to be taken to avoid paths which are heavily obscured by trees, as they tend to have the most impact on signal level.
Our philosophy on mesh networking deployment is not only that the per-node cost should be kept low, but that there should be libreral placement of nodes at convenient locations. This keeps both installation and upkeep simple. And while there are two towers and a few tall buildings on campus that we wouldn't mind taking advantage of, the light posts(light standards) would serve well as locations for nodes.
What is wireless mesh networking?
TrackBack URL for this entry:
http://engtech.ca/mt/engtech-tb.cgi/28