Research

As we have already indicated, our research is about making broadband in remote areas affordable and fast. There are a number of problems we need to solve, not all of them are technical, but being computer scientists we are mostly interested in the technical issues. If you are interested in any of these topics, please get in touch with us.

• Embedded Applications One of the important things to know about our masts and the relays on people’s roof tops is that they each contain quite a powerful computer that we can program.

  • Usually we can do this programming remotely, but occasionally we mess up and have to make a trip to the mast in question. It’s no fun climbing up a mountain in miserable weather, so we want to develop software that helps us configure the masts, installs new software and tries to prevent us from messing up.

  • We can use these individual computers to make intelligent decisions about what to do with the data they are forwarding. For example, if you are using voice over IP – also called internet telephony – even short interruptions are unacceptable, but if you are downloading a large picture, you would probably be more concerned about the overall speed of download, even if there are short interruptions. Smart software could understand that the two kinds of data should be treated differently.

  • Smart software at the masts could figure out automatically how to reconfigure the network in case of a failure. It could also figure out how to reduce the power needed for the computers and wireless cards. For self-powered masts this is very important.

• Mast location Finding locations for masts is a very difficult task. The mast we have between Sabhal Mòr Ostaig and Arnisdale is at the lowest point (around 310m) that has line-of-sight to both places. It is quite hard to find such a position. Also, look at the map and see if you can establish a better ring that will serve the communities involved. And it’s not just about finding connections with line-of-sight; there are other factors:

  • Ease of access. It’s a slog climbing up 310m and something that is dangerous in bad weather.
  • Proximity of a power supply. Self-powered masts are more expensive, both to build and to maintain.
  • Access rights. We have to put these masts on someone’s land.

  In addition, we need more than just line-of-sight. In order to get a strong signal the line-of-sight has to be well clear of obstructions.

  There are many computational problems here involving algorithms, optimization, computational geometry, user interfaces and encoding of geo-spatial data. interfaces

• Self-powered masts A mast needs about the same power (around 20W) as a high-efficiency light-bulb; and it needs it all the time. It would require a rather large solar panel to produce enough energy in a Scottish winter, so we need to use a combination of solar and wind, and we need to have adequate battery storage. But what is the right combination? To answer this question we need to understand weather patterns, equipment cost and risk analysis.

• Transmission over water Water can act as a mirror to wireless transmission between two masts. It can reflect a signal that interferes with the direct signal what this means is that broadband can be affected by the tide. One way to get round this is to move masts to higher positions, but this is not much use to people on low-lying islands. We are investigating less costly methods of dealing with this problem.

In addition to the electronic and computational problems, there are social and economic issues. For example, given that there is some subsidy for broadband in remote communities (as there is for some urban communities) what is the most effective way of providing the subsidy? Also, we imagine that the patterns of usage of internet will be different in remote communities; there may be an increased demand for internet telephones and internet radios. This is something that we need to measure.