Technology and Infrastructure for Emerging Regions

By Tamara Straus 

When Kurtis Heimerl applied for the PhD program in computer science at UC Berkeley, he didn’t intend to focus on technology for developing countries. But several experiences propelled him in that direction.  

Some friends of ours have posted a video demonstrating their use of cellular repeaters to bring coverage some remote parts of Panama. Available here: http://vimeo.com/78562876
 

 Our paper on the Papua network, Local, Sustainable, Small-scale Cellular Networks, has been accepted for publication at ICTD'13. See you all in cape town!
 
P.S. We also visited the Rhizomatica network in Mexico over the summer, wonderful stuff. Hopefully I'll be doing a writeup on that soon. 

 We've just switched to using Disqus for comments for our site. Hopefully this will foment some discussion (and end our spam problem)!

 Though we're a bit late to update this, we won the community award for our work in lowering the power draw of cellular sites at NSDI '13. A link to the work is available here: www.cs.berkeley.edu/~kheimerl/pubs/vbts_nsdi_13.pdf

 This blog has detailed some parts of the Village Base Station project. Our goal is a cellular network designed to to be operated by small-scale organizations and NGOs. We're specifically interested in supporting community; making cellular networks that no longer just bring the outside world into your home but also bring your neighbors closer together. Finally, after months of preparation, hacking, tree climbing (not by us, unfortunately), flying, getting stuck on bad roads, and negotiatation with carriers...  The Village Base Station is open for business.

We’re in Desa, a small town in Papua, Indonesia, setting up TIER's Village Base Station, a low-power, low-cost GSM network designed for rural areas. While we're also interested in things like building locally-relevant and useful services for the users of our networks, one of the main values of bringing cellular phone service to a place like Desa is connecting people to the outside world. Being able to make a phone call or send an SMS to Wamena, the nearest large town, can save a day's worth of travel and a large amount of money.

Kurtis Heimerl and I arrived in Papua, Indonesia last week to deploy TIER's Village Base Station. The Village Base Station, or VBTS, is a system for building GSM phone networks that's particularly well suited for rural areas. VBTS costs less than 1/10th the price of traditional cellular equipment (in no small part due to its use of the excellent OpenBTS project), but where it really shines is its low power consumption: since it only requires a peak power of 80W, we're able to run our VBTS here solely on micro-hydro and solar power. In the coming months, we'll be working closely with WamenaCom, a local ISP, and the Misionaris Sekolahin primary school to build a local cellular phone network here in the Central Highlands of Papua. We're basically the phone company, like AT&T or Verizon, but we're not just providing phone service: we'll also be deploying local services to support the excellent work that Misionaris Sekolahin is already doing in this community. We're starting in the town of Desa, about 300km southwest of Jayapura. Wamena, the nearest town with cellular phone service, is a two hour drive away -- when the road is passable!

Field test mode shows it works!

Normally you'd expect a first-week trip report to be full of frustrating stories of things not working, but aside from drilling through one of the motherboards we brought we've been making smooth and steady progress. We set up our first base station the day we arrived. WamenaCom uses a pole strapped to a tree next to the school to provide Internet access to the community, so we decided to mount our base station there too. After a bit of impressive tree climbing by the WamenaCom and Misionaris Sekolahin folks, the first VBTS was powered, networked, and installed, and we were able to make test calls. Not bad for day one! Coverage has exceeded our expectations, with our single base station able to provide strong signal to central Desa, even reaching the neighboring town of Kota about a mile away across a valley.

Preparing the base station for installation.

Mounting the base station on the tree/tower.

Word that cell phone signal is going to be available in Desa has spread like wildfire throughout the community and neighboring villages (well, like wildfire would spread if not for all the rain we've been having). We've still got a lot of work left to do to make the network reliable before we can open up the network and start selling SIM cards to people. Since that first day, we've set up outgoing calling, built a billing system for the network (our friends at WamenaCom will be maintaining this network in the long term, so we want it to be financially sustainable) and done a *lot* of performance and reliability tuning, which is vital since our only connection to the outside world is a VSAT connection. We also have been doing training with WamenaCom on building ISP-grade long-distance WiFi networks; we'll be using the long-distance WiFi network they're building to connect our next VBTS installations in neighboring towns back to our central hub in Des.

Looking towards Desa from Kota (Desa is in the center right of the image).

Setting up the long-distance WiFi relay in Kota with WamenaCom.

Everyone we've interacted with is very excited that they'll be able to use their cell phones for calls and SMS around Desa, whether to communicate with family in other parts of Indonesia or just to send messages across the valley (a trip that can take hours of hiking up and down steep and slippery hillsides). We're looking forward to working with WamenaCom, Misionaris Sekolahin, and the rest of the community here to deploy VBTS over the next few months.

 I was recently at the Internet at Liberty conference in DC, where inter net activists (i.e., activists who use the inter net) congregated and discussed mechanisms for dealing with surveillance and monitoring by telecommunications companies. Particularly, I was demoing my own "Evil Basestation" project (built on OpenBTS), which teaches activists what the network knows about them and how it gathers than information.

Kashif Ali and I recently put together a video demonstrating some of the technology we've developed to enable extremely low-power GSM cellular infrastructure. That video is available here: www.youtube.com/watch?v=i8PElVGDe5M It's a very technical discussion, so you may be lost if you're not already involved in GSM/Wireless. We're thinking of producing a simpler video in the future.