Hybrid Genetic Algorithm and Tabu Search For Solving the Frequency Assignment Problem in Cellular Network

Abstract

As more individuals joined the mobile area, the demand for the mobile service is increased. However , the limited number of frequency allowed raises the problem of frequency allocation. Simply it defined as the allocation of different frequency to each cells and its neighbor. This problem is considered an NP –problem. This paper proposed a hybrid algorithm between genetic algorithm(GA) and tabu search techniques (TS) to solve this problem. The proposed algorithm embeds the features of the local search, elite list and intensification from TS into mutation operation of GA. The aim is to find an algorithm which behave more efficient than the primary algorithms in terms of speed, complexity and search space problem like cycling in TS and premature convergence in GA. Test results show that the proposed algorithm can get rid of the premature convergence problem in GA and can find the target solution faster than TS.