M-RATS: Mobile-based Radio Access Technology Selector for Heterogeneous Wireless Environment

Wireless heterogeneous environments are a common reality today and will be so even more in the future. This paper is focused on providing a wireless access selection method for the future mobile terminals that will use naturally inspired algorithms for best network selection. The novel algorithm uses mobile terminal measurements from different radio access technologies within a given time window. The network selection is based on specified criteria, whose number can be increased as needed. Each criterion is modeled with a membership function associated with Fuzzy Logic controllers, which are optimized by using Particle Swarm Optimization. Each criterion is weighted, with the aim to assign the influence on the decision for a radio access network. Hence, our proposed algorithm implements past knowledge of the service performances of available wireless networks used by the mobile, to make a decision for network selection on given time intervals. The main contribution in the algorithm for wireless access network selection is in the usage of nature inspired algorithms, such as Particle Swarm Optimization - PSO, for optimization of Fuzzy Logic controllers, as well as usage of a Genetic Algorithm for optimization of decision making based on multi-criteria inputs. The simulation analysis has shown that the proposed algorithm gives a better probability of satisfied users compared to other existing wireless network selection algorithms.