Community Networks Growing the Internet

Using TV Whitespace To Connect Communities in Malawi

About twenty years ago, people here at the International Centre for Theoretical Physics (ICTP) in Trieste, Italy started thinking about what happens to the people who come here after they go home. Specifically, what happens to the academics and researchers from the developing world who come to work with us. 

At the time, they’d come, work with us, do some amazing things, and then go back to places where they didn’t have the Internet. With no Internet, it was very hard for them to do research. They cannot download papers, they cannot contact colleagues. So, we started teaching people about internet technologies, and, at that time, we already had the vision that wireless is much cheaper and much quicker to deploy than fibre, or copper or anything else. We started teaching people about wireless and WiFi when it wasn’t even called WiFi.

We did the first link in Nigeria, at Obafemi Awolowo University in Ife-Ife. That was one of the first WiFi links, or proto-WiFi links, on the whole continent. From there, more and more people got in touch with us and started asking for training.

An interesting thing happens when you create a long-distance wireless Internet connection to somewhere, which is that you also make it possible to bring the Internet to other, nearby places, as well.  A few years ago, we worked on connecting hospitals in Malawi. At one of the hospitals, there was a girls’ school nearby, and they said ‘Oh, we’d love to be on the Internet, why don’t you connect us?’ Then there was a boys’ school further up the road that said ‘Oh, we want to get the Internet as well.’ And then once the Internet is in the schools, you can do some clever stuff like giving communities access to the bandwidth at night. There is no one in the school at night, so that bandwidth is going to be wasted.  Why can’t we give that to communities or public parks, or you know, telecentres or whatever, in the evening? 

One of the things we’re working with a lot right now is TV white space. In most of the world, they only have two or three TV channels, so all the rest of that broadcast spectrum is going unused. So, in Malawi, we started measuring TV white space, producing maps, and the local university went to the regulator with the maps that we produced together, and they said ‘Look, we’re only using two channels out of a hundred, can we please use some of that spectrum to connect schools and hospitals?’  So, the government of course could only say yes. We had one pilot project, then another other pilot project, and that’s the story of how the TV white space started.

There are advantages and disadvantages to using TV white space for the Internet. The disadvantages are that the equipment is much more expensive, because it is not a standard yet, antennas are really, really huge — they’re basically TV antennas — and the throughput is lower, so it’s not as fast. The huge advantage for us, though, is that the penetration is much better. It can go through walls much more easily. It can go through vegetation.  In many of these countries, when it rains, you don’t get WiFi on the university campuses, because trees are wet, and the signal doesn’t pass through wet leaves. It can also cover huge distances, and by using high-powered radios, we could limit the number of access points. 

Malawi gave our local partners a license to use TV white space for one year, and then that worked, so now they extended it for a second year. Now, Malawi is going to become the first country in Africa to regulate TV white space. If an ISP wants to use TV white space to provide internet to their customers, they can do it. Now, that’s having a snowball effect, so we helped launch a similar project in Mozambique last year, near Maputo. This year we’re going to go into three other places to connect about 10 schools.  What we’re trying to do there is to kind of extend the network, not only to TV white space sites, but then to have some MESH networks to connect communities.  

The ICTP is a bit different from some organizations, because if you ask us what difference our work makes on the ground, we don’t always know. We partner with local Universities and train people here, and then they go home and create networks. They see the changes, and they evolve their projects as they see fit. And that’s a great thing, they already know what they need. 

Note about the Internet Society and its support for Community Networks
The Internet Society is dedicated to helping the unconnected connect. We do this through funds provided by our Global Engagement team and through our grants programme called Beyond the Net. We are keen supporter of ICTP’s. Their team will be working with us to help train more community networks in Africa on radio frequency and spectrum basics. 

This article is part of a series on Community Networks.

Growing the Internet Internet of Things (IoT) Open Internet Standards

The Internet as a lifeline in Asia-Pacific

Disaster management is becoming more and more prominent in the region’s policy agenda—and with good reason. UNESCAP in a recent report noted that more than half of the world’s 226 natural disasters in 2014 occurred in Asia-Pacific. Active tectonic plate movements in the Pacific and Indian Oceans, areas that also generate a lot of tropical typhoons, combined with increasingly denuded forests have made the region the most disaster-prone in the world. Perhaps more alarmingly, the study found that Asia-Pacific countries are largely ill-equipped to deal with these crises.

This may soon change, with the help of technology. Around the region, numerous pilots seek to help developing countries monitor shifting temperatures and weather conditions—water levels, soil moisture, wind speed, rainfall rate—using solar powered sensors and cheap, tiny, single board computers like Raspberry Pi. These early warning systems can transmit data wirelessly through mobile networks and satellite links to a central server, helping to detect impending hazards like landslides and flash floods.

In Vanuatu, ICTs were widely credited as instrumental to minimising the death toll from Cyclone Pam. As the category 5 cyclone approached the country, several government agencies, working with mobile operators, drew up a plan to coordinate evacuation and recovery. A task force was formed, additional staff contracted, and free SMS alerts sent to citizens throughout the archipelago. Communications infrastructure was badly affected, but satellite phones dispatched in advance allowed first responders to organise relief efforts. Cyclone Pam, despite being one of the most powerful to hit the South Pacific, suffered 11 casualties—much lower than the 1987 category 4 cyclone that struck Vanuatu and left 48 dead.

ICTs were just as useful in the wake of the violent earthquake in Nepal in April. The tremors, which flattened entire villages and displaced close to half a million people, mobilised the global community to act, in no small part through online channels. Internet giants Google and Facebook each deployed tools that let individuals post or search for updates on friends and relatives in the country. Telecom carriers and VoIP applications offered free mobile and landline calls to Nepal, and regional e-commerce firms like Paytm collected donations for earthquake victims.

Thousands of volunteers across the world, including Nepalese expatriates, plotted feeds from social media networks, news and satellite images onto online maps. On the ground, groups like the World Food Programme turned to lightweight, inflatable mobile data antennas, which look like big beach balls, to restore connectivity in remote locations, allowing aid workers to collect, upload and verify information. Geo-tagging enabled platforms like ShakeMap to continuously show the disaster’s impact in real-time, giving a clearer picture of which roads were blocked, or who was stranded where—much like Ushaidi had done during the 2011 earthquake in Christchurch, New Zealand. Online connectivity was essential particularly when phone lines became congested, and survivors relied on the Internet to reach out to their loved ones.

Such emergencies highlight the value of low-cost and low-energy networks that can be deployed easily, especially in the immediate aftermath of a disaster, when communication is most critical. Nepal, for instance largely preserved its international connectivity throughout the quake and its 300 aftershocks. Much of the damage was at the last mile which, along with severed power lines, impaired the average person’s ability to go online.  

One solution lies in systems that can connect with each other even when the Internet is down. Commotion Wireless, for instance has developed an open source software that allow devices to communicate peer to peer using unlicensed spectrum, forming a mesh network that is highly redundant, and capable of sharing an Internet connection that is beamed or made available close by.

A similar idea in the Philippines takes resilience a step further. Showcased at the 12th APT ICT Development Forum last month, the project, jointly conducted by university-based Ateneo Innovation Centre, the DOST ICT Office and Japanese partners, employs mobile transponders that relay urgent messages from victims and rescuers to vehicles roving disaster areas. Equipped with one terabyte near-cloud servers, the latter will act as computing nodes, aggregating information that can be passed on to other vehicles until it reaches the global Internet.

A crucial ingredient to this concept is delay-tolerant networks, which incrementally store and move data along until it gets to its destination—a good option for places that lack reliable and instantaneous bandwidth. But perhaps even more crucial are policies that enable innovators to test and scale technologies that may be useful in disaster relief and preparation. As an example, many of these initiatives are pushing to have freer use of TV white space or more broadly, frequencies in the 700 to 800 Mhz bands, which have wider coverage and can penetrate through metres of debris.

The onset of climate change can only mean more frequent, unpredictable and severe cyclones, droughts, avalanches and tsunamis that injure hundreds and cause devastation that may take years, if not decades to repair. ICTs are becoming more advanced, available and in many ways cheaper but at the end of the day, these tools are precisely, tools. It is the human component—coordination, collaboration and trust between various stakeholders—that make technologies work before, during and after every calamity.