Managed networks reduce interference by carefully selecting the Wi-Fi Access Points (AP's) location. They are available for enterprise business users, but are expensive, while private AP's are deployed at random without channel and power coordination. The use of central servers for coordination outside enterprise environments has so far not been a success. Avoiding interference without central coordination is not an easy problem to solve, as each configuration change affects neighbors, eventually reaching thousands (or even millions) of networks that have to be reconfigured.
Maseng AS is a start up company which is developing a software suite for Wi-Fi access points (AP) which addresses these problems. The system is called Empathetic Radio and is based upon a completely distributed peer to peer architecture. Control channels are encrypted backhaul (wired) between APs which exchange neighbor reports and form clusters in which all Wi-Fi networks disturb at least one other network and then perform channel and power allocation. All operations are automatic without operator or customer intervention.
The software suite is tailored to one particular AP brand chosen by the ISP. Their customers who live in multi-dwelling units (flats or apartments) suffer not only from co-channel interference but also from adjacent channel interference from their neighbors. This is because the AP radio filters are cheap. To overcome this problem this filter shall be measured and taken into consideration in the algorithms allocating channels.
The Master thesis is focused upon measuring the suppression of this filter between the adjacent channels to the operational channel. This will be done by using two networks, each equipped with one AP and a client.
The progress of the thesis will be followed by weekly meetings from UiO and Maseng AS as jointly agreed upon when requested by the candidate.