[Mesh] Fwd: [eecs-announce] Dissertation Talk: Community Cellular Networks

Hi! Great stuff. Mitar -------- Original Message -------- From: Kurtis Heimerl <kheimerl(a)cs.berkeley.edu> Subject: [eecs-announce] Dissertation Talk: Community Cellular Networks To: eecs-announce(a)lists.eecs.berkeley.edu Title: Community Cellular Networks Speaker: Kurtis Heimerl Advisors: Eric Brewer and Tapan Parikh Date: Tuesday, November 12, 2013 Time: 2:00 PM Place: 242 Sutardja Dai Hall (Not Soda, Jorge took the nice room :( ) Abstract: Cellular networks are one of the most important innovations of the 20th century, with over six billion subscribers across all continents on Earth. Despite this, hundreds of millions of people in rural areas remain without coverage. We see two reasons for this: First, cellular installations are very expensive to install in rural areas, with power costs dominating the price of the total install. Secondly, cellular equipment is traditionally installed “top-down”, with major nation-scale providers bringing coverage to rural areas. In this work, we propose the model of Community Cellular Networks, small scale, low cost, locally operated cellular networks. To enable community cellular networks, we utilize OpenBTS, an open-source implementation of the GSM um layer. Building on this, we implemented two key technological innovations. The first is Virtual Coverage. Virtual Coverage introduces a “sleep” mode into the GSM protocol, allowing the base station (BTS) to “sleep” when the network is idle, reducing power draw. An autonomous radio, the WUR, allows users to “wake” the BTS when they need to communicate. Secondly, we implemented the Village Base Station, a series of extensions to OpenBTS and FreeSWITCH allowing for easily customizable cellular networks. These innovations were evaluated in the context of an in situ deployment of both Virtual Coverage and the Village Base Station in a small community in rural Papua, Indonesia. Through this 10 month ongoing deployment, we found that Virtual Coverage reduced the night power draw of the network by 56.6%. We also found that the Village Base Station allowed us to build a network well suited to the community. As of now, it has handled over 200 thousand SMS and is financially sustainable for the local operator, even if they were forced to finance the network entirely. I plan on purchasing Rambutan for this talk as well. Take that under advisement. -- http://mitar.tnode.com/ https://twitter.com/mitar_m