中文摘要：

针对认知车载Ad Hoc网络（CVANET）信道的动态特性,以部分可观测马尔科夫决策过程（POMDP）为模型对认知车辆用户的频谱感知和频谱接入过程进行研究,提出基于POMDP模型的分布式机会频谱接入算法,并通过贪心算法降低POMDP算法计算量,最后通过仿真研究影响认知车辆用户吞吐量的主要因素,并验证算法的可行性。仿真分析结果表明,认知车辆用户通过本算法接入吞吐量得到有效提高,降低了交通中广播风暴的可能,并且降低了计算量。

英文摘要：

For the dynamic features of cognitive vehicular Ad Hoc network （CVANET） channel state, a study on spectrum sensing and access of cognitive vehicle users was presented, which was based on framework of partially observable Markov decision process （POMDP）. Then, a decentralized opportunity spectrum access based on POMDP model was proposed. Because there is a huge computational complexity, the greedy algorithm was sought. Finally, the influence factors of cognitive vehicle user throughput through the simulation were studied, and the feasibility of the algorithm was verified. The simulation results show that cognitive vehicle user＇s access throughput is effectively improved by this algorithm which avoid the happening of broadcast storm and also simplify the calculation.