UnwirePDX-Watch Report Complete

Personal Telco Compatriots:
    Russell and I finished the report and put it online a couple of days ago:

  http://unwirepdx-watch.org/?page_id=12

It's 35 pages of municipal-mesh-testing-goodness. The cliff notes version
goes something like:

 * We found a useable coverage of 58% within 500 feet of the
Proof-of-concept APs
 * Connections were only acheived with a high probability (~90%) within
250 feet of APs
 * At 1000 feet, the coverage falls off to 48%
 * The backhaul network seems to have about 2% packet loss on average
 * Per-client throughput and latency is reasonable, when it isn't being
throttled
 * Client-to-client communication is totally filtered
 * Their upstream ISP is Cogent and all traffic seems to go through one
gateway

The city has already signed off on the network, but our results suggest
that they should have insisted MetroFi increase their mesh density to
obtain the desired coverage. But, then again, according to recent
interviews with Logan, they can't afford to deploy nodes more densely than
they do now. Given that, 90% seems like it might have been too high a bar
to set with the technology and business model they chose. It'll be
interesting to see how other cities deploy their networks, what coverage
criterion they choose, and how they verify it. In Portland, however,
citizens will receive the network that MetroFi chooses to build, and, if
it is deployed the same way the proof-of-concept network has been (as it
is assuredly safe to assume), we can expect around 60% usable outdoor
coverage.

Of course all the data, configuration, software, and results are
free-as-in-freedom. Enjoy,

-- 
Caleb Phillips

-- 
The Personal Telco Project - http://www.personaltelco.net/
Donate to PTP: http://www.personaltelco.net/donate
Un/Subscribe:  http://lists.personaltelco.net/mailman/listinfo/general/
Archives:  http://news.gmane.org/gmane.network.wireless.portland.general/
Etiquette: http://www.personaltelco.net/index.cgi/MailingListEtiquette