An MDP-Based Optimal Network Selection Algorithm for Heterogeneous IoV

In the V2V and V2X-based heterogeneous vehicular networking system, the vehicles need to select the optimal network, between each network. For network selection techniques generally can't meet the random and memoryless states of the networks and vehicles, and can't predict the states transition, network optimal decision algorithm based on VMDP is proposed. The algorithm established the VMDP state space and state transition probability matrix, and used VIA to get the optimal network selection strategy. Through the performance comparison with other algorithms, the simulation results show that the algorithm can not only effectively achieve the vehicle optimal network selection, but also achieve an improvement of 10%~30% on the communication link quality and vehicular switching times. The proposed algorithm can effectively improve the performance of vehicle communication.