Dynamic Spectrum Management with Movement Prediction in Vehicular Ad Hoc Networks

Vehicular networks have been widely considered as a promising architecture to enable vehicles-to-vehicle (V2V) and vehicle-toinfrastructure (V2I) connections, and the wireless access to the Internet for the vehicles. Current sparse deployment of the roadside units (RSUs) requires multi-hop data transmission among the vehicles and the RSUs, and thus vehicular ad hoc network (VANET) has been introduced. As one of the key resources in VANETs, radio spectrum has to be carefully managed to facilitate high spectrum utilization efficiency and network throughput. In this paper, we propose a dynamic spectrum resource management scheme for VANETs, with the assistance from the movement information of the vehicles that are usually available to the network. The link status prediction is made based on the vehicle location and movement information, for optimizing the spectrum band allocation to achieve maximized system reward. The optimization problem is solved, allowing the vehicles to utilize multiple spectrum bands and make decisions in a distributed and parallel manner. Extensive simulation results are also presented to demonstrate the significant performance improvement of the proposed scheme compared to the existing scheme that ignores the spectrum management optimization with vehicle movement information.