Self-Optimisation in Future Mobile Networks

~£60k: 01/08/2008 to December 01/08/2011

Standard are being defined for the Long Term Evolution (LTE) of cellular access technologies.   These future networks will have decentralised
control, with as many decisions as possible being delegated to the "Evolved Node-B" (the network element equivalent to what we currently call
the base station or phone mast). E-node-B's will be capable of self-configuration and self-optimisation based on the current radio environment, via
collecting and monitoring data and exchanging information between themselves. This self-optimisation aspect of the LTE architecture breaks
away from the traditional business model of the telecommunication operator service provider, and much research is needed to identify appropriate
self-optimisation strategies. This project will investigate specific areas and opportunities in this LTE/self-optimisation context that are
of importance to Motorola. We will co-opt or build a (initially simple) network simulator, and use this to test a number of different strategies
for how E-node-B's will be able to self-optimise based on collected data, testing each method via different network loads and traffic
distributions. An interesting feature that the project would probably focus on is the evolution (via evolutionary algorithm) of robust
self-optimisation strategies.

Project Team:

Supervisors : David Corne (Heriot Watt), Chris Murphy (Motorola)

Research Student:  Andrew Thompson