Fuzzy Q-learning based vertical handoff control for vehicular heterogeneous wireless network

As a novel fundamental platform for providing real time access to wireless network, vehicular communication is drawing more and more attentions in recent years. IEEE 802.11p is a main radio access technology which supports communication for high mobility terminals. Due to the limited coverage, it is usually deployed coupling with cellular network to achieve seamless mobility. In cellular/802.11p heterogeneous network, vehicular communication has the characteristics of short span of time associating with Road Side Unit (RSU). Moreover, the media access control (MAC) scheme of IEEE 802.11p decides that packet collision probability increasing followed by the increasing of user quantity, which leads to the decreasing of network throughput. In response to these compelling problems, we propose a fuzzy Q-learning based vertical handoff (FQVH) control strategy for supporting the mobility management. FQVH has online learning ability and can give optimal handoff decisions adaptively with no need for prior knowledge on handoff behavior. Simulation results verify that it can adjust handoff strategies to different traffic conditions, and keep users always connected to the best network.