Coalitional Graph Games for Popular Content Distribution in Cognitive Radio VANETs

Popular content distribution is one of the key services provided by vehicular ad hoc networks (VANETs), in which a popular file is broadcast by roadside units (RSUs) to the onboard units (OBUs) driving through a particular area. Due to fast speeds and deep fading, some file packets might be lost during the vehicle-to-roadside (V2R) broadcasting stage. In this paper, we propose a peer-to-peer (P2P) approach to allow the OBUs to exchange data and complement the missing packets. Specifically, we introduce a coalitional graph game to model the cooperation among OBUs and propose a coalition formation algorithm to implement the P2P approach. Moreover, cognitive radio (CR) is utilized for vehicle-to-vehicle (V2V) transmissions so that the P2P approach does not require additional bandwidth. Simulation results show that the proposed approach performs better under various conditions, relative to the noncooperative approach, in which the OBUs share no information and simply respond to any data request from other OBUs.