Learning route from beaconing and interest dissemination in vehicular sensor networks

Vehicular sensor networks can be used to collect information for the purpose of driving assistance, environment monitoring and entertainment. However, due to the vehicle movement and lossy feature of wireless channel, providing a reliable multi-hop communication between a sensor node and a sink node is particularly challenging. In this paper, we propose a protocol which learns the optimal route in the exchange of hello messages and interest messages. The proposed protocol uses a fuzzy logic to evaluate a direct wireless link and uses a Q-learning algorithm to learn the best route from a multi-hop perspective. We use computer simulations to show the effectiveness of the proposed protocol.