Novel Network Selection Mechanism Using AHP and Enhanced GA

Abstract

To support high bandwidth with high mobility is the predominant objective of next-generation networks. As some networks such as UMTS (universal mobile telecommunication system) can offer a wide geographical coverage with relatively low data rate, while other networks such as WLAN (Wireless Local Area Network) can support much higher bandwidth at a local level, network selection is becoming one of the most significant challenges for the next-generation networks. In this paper we propose an efficient network selection mechanism which combines two mathematical techniques. Analysis hierarchy progress is utilized to achieve the weights of QoS factors based on the user preference, and genetic algorithm can solve the multi-objective optimization problem. As pure genetic algorithm has weak local search capabilities and low efficiency around the Pareto Optimal Solutions , it may even cannot reach the Pareto optimal solution. To solve this problem, we propose the enhanced Pareto genetic algorithm combined with Tchebycheff method, which can improve efficiency of optimization and ensure a better convergence to the true optimal network.