An effective multi-objective hybrid immune algorithm for the frequency assignment problem

Abstract

This paper presents a novel multi-objective hybrid metaheuristic that combines an artificial immune system (AIS) with genetic operators to address the multi-objective Frequency Assignment Problem (FAP) in cellular mobile networks, seeking to minimize three objectives simultaneously: Total interference, Maximal interference, and the number of used frequencies. The proposed approach inherits its evolutionary cycle from a multi-objective immune network supplemented by a FAP-specific local search as well as clonal selection for a better intensification. Moreover, a genetic mutation is applied at each generation to enhance the exploration of the research space. Computational experiments performed over COST259 instances and based on the hypervolume metric, show the efficiency of the newly proposed hybrid algorithm, and corroborated by the comparisons we did with the most frequently referred algorithms in the related literature. Furthermore, the effect of the main parameters and the interaction between them is analyzed using statistical tools.