A network selection strategy in resources mobility environment

Various wireless networks have emerged and constituted a complex radio access network environment. In cognitive environment, network resources are always varying with time and space. When a mobile cognitive user with a multi-mode terminal generates a call request, how to select the best network? Considering that the traffic load varying with time, we introduce back propagation (BP) neural network to predict its changing trend. At the same time, because parameters obtained through measurement or prediction are imprecise and uncertain, we introduce fuzzy logic to deal with such uncertainty. Then synthesizing technology preference, operator preference and user demand, we propose a multi-attribute decision making (MADM) network strategy based on BP neural network and fuzzy logic. Finally, we evaluate the performance of our strategy through simulations. The results show that our strategy can effectively increase users´ utility-price and reduce their dropping probability.