"I want people to think big," said Brewer, a professor of computer science at the U.C. Berkeley who has spent the last five years seeking ways to use WiFi to improve life in rural communities in the developing world.

Brewer shared his views and experiences in a keynote to a group of fellow academics and engineers at an IEEE workshop on wireless mesh networks here Monday (June 11). He describes his work in the video below.

His efforts have resulted in significant humanitarian and technical milestones to date. Brewer and others part of the Berkeley Tier project helped establish remote eye clinics in India that have helped as many as 3,000 people regain vision. They also assisted a team that set the world's record for the longest WiFi link—a 220 mile connection in Venezuela that delivered 6 Mbits/s.

"The secret is you have to be able to see far," he said of the link that stretched between a relay station on an Andes peak at 4,200 meters to one at sea level.

The effort used a long distance version of WiFi dubbed WiLD developed at Berkeley and installed in rural communities in India and Africa to improve medical care and education. The technology uses existing low cost WiFi hardware, but to extend range it adds a time division multiple access layer, simplified acknowledgement structures, new forward error correction code and dual antenna capabilities.

Cellular will link communities in the urban areas of developing nations, but the technology requires dense populations to make it economical to install relatively expensive base stations. WiFi is a cheaper pay-as-you-go technology more suitable for rural areas which represent a strategic piece in the problem of global poverty, he said.

"The beauty of WiFi is you don't need the carriers, and rural areas can't wait for carriers," Brewer said

As many as four billion people live on as little as two dollars a day, a largely rural population that could swell to eight billion in 25 years and is rapidly migrating to big city slums in search of a better life, according to Brewer.

"It's the largest and fastest growing part of the population," he said. "We can give people in rural areas hope and mitigate the urban migration."

Plenty of challenges are still ahead, the biggest being the poor quality of available electricity in developing nations. Power spikes and drops in places such as rural Africa account for as many as 90 percent of the faults on the Berkeley networks, frying equipment, causing it to hang or corrupting flash memory chips.

AirJaldi, a Berkeley partner in India, has developed a $2 low power detection circuit that disconnects routers so they don't hang when power levels drop. It is also designing a 7W solar and a hybrid solar/electric power module.

"You want to put these mesh networks in places where you may not have power," Brewer said.

The lack of open spectrum is another major headache. Some African countries forbid transmitting data across open spaces. Others let local carriers own all the nation's spectrum including WiFi bands. Although the 2.4 GHz band is available in about half the countries in Africa, the 5.8 GHz band is free in less than ten percent of them, he said.

Finding funding is one challenge that has eased slightly in recent years. Brewer's work is currently funded by the National Science Foundation. The U.S. State Department is providing funds for a remote health care project in Pakistan and other government bodies are showing signs of interest in projects.

"Private foundations represent the big unknowns," said Brewer. "Historically they have avoided technology funding, but there is a role [engineers can fill] in showing them how to do work that is technically savvy," he said.

Today, the notion of a 20-50 Mbit/s Gibraltar-to-Ghana backbone based on a ten-hop WiFi mesh is just a concept Brewer described for a keynote speech. Long term such networks could become important pieces of infrastructure for poor rural communities.

"People still do not fully appreciate all the value WiFi has to offer," Brewer said.