蚁群路由算法在车载自组网中的研究和应用.pdf

Abstract Today, our countr has strongl advocated the development of Internet. As technolog advances, more and more people have paid attention to non-traditional networks. SinceVANET(VehicularAdHocNetwork)hasbecomemoreandmoreclose to the human’s life, VANET graduall comes into our view. Firstly, we stud on VANET’s routing algorithms and ABR(Ant-based Routing Algorithm) in this paper. Thenweanalyzedthe shortcomingsoftheantcolon algorithm,especiall thenetwork congestion problems caused b local optimum and problems, e.g. optimal path’s ant pheromone can’t accumulate quickl enough. To solve these problems, this paper introducesthe standardQ-learning algorithminreinforcement learning and assessment mechanisms of channel to improve ABR’s qualit . Finally, we designe a new added-service discover protocol based ABR for VANET, and then we verif the protocol’sfeasibilit andeffectivenessinsimulationplatform. ThispaperresearchedontheVANET’sroutingalgorithmanditsclassificationand the principle of classical ant colon algorithm. Based on these theories,we stud ant colon algorithm’s classic application in vehicular ad-hoc network.At the same time, this paper stud the principle of reinforcement learning theor and the principle of standard Q-learning,and Q-learning’s application in vehicle ad-hoc network. To deal withtheexistingproblemofABRwhileitisappliedinVANET,thispaperdesignakind of Q-learning-based ABR algorithm, whose effectiveness is verified b experiment; then we achieve the applications of the improved ABR algorithm to VANET‘s environments.Inthispaper,tomake antrouting algorithmmore suitableforthe actual requirements of VANET, we also designed an improved ABR algorithm based on assessmentmechanismsofchannel,andthentheimprovementschemeisappliedtothe VANET.Basedonthestud ofVANET’sservicediscover strategy,wedesignakindof value-added service discover strateg which based on ABR algorithm. Finally, we conduct