Predictive RSS with fuzzy logic based vertical handoff algorithm in heterogeneous wireless networks

摘要

To achieve seamless mobility in integrated wireless and mobile networks, an intelligent vertical handoff mechanism is required to guarantee Always Best Connected. In this paper, a vertical handoff decision algorithm for nonreal time and real time services and network selection for WLAN/WiMAX/UMTS are presented. The conditions of handoff decision procedure consider predictive received signal strength of the target networks, received signal strength (RSS) of current serving network and residence time in the target network if their conditions are consistently true during dwell time to reduce connection dropping and unnecessary handoffs. Back-propagation neural network is used for RSS prediction. The handoff policy while mobile is located in WLAN/WiMAX and UMTS are different to prolong staying in WLAN/WiMAX due to its high bandwidth and low cost, and to minimize handoff delay for real time services in WLAN/WiMAX. The merit function is used as a handoff metric for evaluating network performance based on user preference in order to decide candidate networks for user. Fuzzy logic based network selection strategy using quantitative decision algorithm is the handoff criteria to determine which access network should be chosen. Simulation results show that the proposed handoff scheme can choose the optimum network through considering between the user preference and network conditions and outperforms the other approaches in the number of handoffs, handoff call dropping probability and Grade of Service (GoS).