Community Cellular Networks


December 3, 2013


Kurtis Heimerl


UC Berkeley


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 200,000 SMS and is financially sustainable for the local operator, even if they were forced to finance the network entirely.


Kurtis Heimerl

I’m Kurtis Heimerl. I work on technology for developing regions with Professor Eric Brewer in the TIER group, as well as with Professor Tapan Parikh in the Represent group. I also moonlight as a developer and open-source community manager at Range Networks, supporting the open release of OpenBTS.

My thesis topic is the The Village Base Station (VBTS), a low-cost, low-power GSM basestation (based off OpenBTS). Our goal is to provide community cellular: local, community-owned cellular networks in areas without existing cellular coverage. Write-ups of our progress are here.

I’ve worked on a wide variety of projects. Some examples include Umati, the crowdsourcing vending machine; PANTS/NetAPI, a modern sockets implementation; mPhone, a message-oriented phone system; and the Python implementation of the DTN bundle protocol.

My Master’s thesis was a project called Metamouse (not to be confused with this), which allows for more equitable sharing of existing educational content. You can see a video demonstration here. I also gave a CITRIS talk on this research which you can see here.