Hybrid channel allocation in cellular network based on genetic algorithm and particle swarm optimisation methods

Abstract

The challenge in wireless cellular network is to suit the changeable traffic demand by assigning appropriate channels from limited frequency spectrum and to maintain a desirable stratum of interference level. The system capacity can be improved by a reduction in interference effect by applying effectual channel assignment technique. This work proposes, genetic algorithm (GA) and particle swarm optimisation (PSO) techniques, with hybrid channel allocation for interference reduction. In GA, integer genetic representation for crossover and mutation operation is applied and graph theory based fitness function is designed. In PSO, hard and soft constraints are used for designing fitness function. The proposed GA and PSO compute co-channel and co-site interferences represented by interfering edges, the computation time and generations/iterations required. The signal-to-interference ratio (SIR) is determined, considering attenuation of the signal predicted by applying Hata propagation model. The performances of the proposed methods are applied on benchmark instances and are compared with the reported literature. The proposed PSO shows improvement in performance than GA in terms of reduction in interference, required computation time, generations/iterations required also improvement in the SIR.