Vertical Handover Algorithm by Fuzzy TOPSIS for Improving System Performances

Abstract

The convergence of wireless heterogeneous networks has led the goal of “Always Best Connected” as the primary challenge for next-generation network selection. In order to improve the overall performances of the system, various criteria corresponding to user preferences and network conditions as well as QoS parameters are supposed to be considered comprehensively. Hence, network selection is a typical multi-attribute decision making issue. In this context, the analytic hierarchy process (AHP) and technique for order preference by similarity to an ideal solution (TOPSIS) are exhaustedly used as representative multi-attribute decision methods (MADM). However, various traffic types have different demands for network. AHP can only calculate the weight of one traffic type at a time. Therefore, different traffic types need to be analyzed separately. Moreover, these two methods are susceptible to subjective influences and are not suitable for handling contradictory and vague information, so they are prone to frequent handover, easily resulting in ping-pong effects. Thus, this paper provides a network selection algorithm which combines AHP, TOPSIS and fuzzy logic. Four traffic types and key indicators are evaluated respectively. Simulation results show that the implemented algorithm can significantly compensate for the above defects and make efficient decisions for all traffic types.