A genetic algorithm approach to solve channel assignment problem in cellular radio networks

We propose an efficient algorithm for channel assignment problem in cellular radio networks. The task is to assign channels to different cells in a spectrum-efficient way i.e. the minimum frequency band is used and at the same time interference among calls from different cells is avoided. The simplest version of this problem, where only co-channel interference is considered, is NP-complete. We propose a solution based on genetic algorithm approach. The contribution of this paper is twofold. (1) A novel way of generating the initial population which creates near optimum solutions when the population is large is presented. In all the cases we tried the best of these solutions are better or as good as those obtained by existing methods. (2) A genetic mutation operation is defined which improves the quality of the previously obtained initial solutions to an optimum solution. Simulation results are reported.