Research On Spectrum Resource Management In Vehicular Networks

Vehicular network is used to reduce traffic accidents,improve the efficiency of the transportation system and the experience of transportation participants,which has attracted wide interest in academia and industry.Cognitive vehicular network(CVN)is proposed to overcome the shortage of spectrum resources caused by dynamic network topology and increasing applications requirements,in which vehicles are allowed to opportunisticly access the unused licensed frequency bands such as TV white space through spectrum sensing and dynamic spectrum allocation technologies.This brings not only new available frequency bands,but also new technology issues to be solved such as spectrum sensing and spectrum resource allocation and management in vehicular network environments.Spectrum resource management technology is required to allocate spectrum resource and improve spectrum efficiency.It is of great scientific significance and has wide potential applications.Therefore,this dissertation investigates efficient new mechanisms,new methods and implementation technologies for spectrum resource management in vehicular networks.This dissertation analyzes the key technical issues on spectrum resource management in vehicular networks and proposed corresponding methods to improve spectrum efficiency,to increase the channel reliability and lower down the dependence on common control channel.The research details and contributions are as shown below.1.Periodic spectrum resource allocation schemes face the problem of repeated computation and time delay for spectrum users to wait for allocaiton.Spectrum allocation is modeled as a multi-step decision-making problem in this dissertation and a low complexity sub-optimal algorithm(LCSA)is proposed.Dynamic Programming method is used to solove the optimal allocation strategy.By this way,spectrum broker is able to store optimal strategy in advance and use them when spectrum requests come.Delay and repeated computation are both reduced.The system status feature subspace is constructed to further reduce complexity.Simulation results show that LCSA algorithm approaches 95% of the optimal strategy performance when the computational complexity is significantly reduced.2.Fast convergence reserved ALOHA(FCR-ALOHA)protocol is propsoed to reduce time slot conflicts and to accelerate the time slot reservation coordination process in safety message broadcasting of vehicular networks.FCR-ALOHA utilized merging collision detection and time slot election to make one of the collided vehicles to keep reserving the collided time slot in the subsequent communication process.By this way,the time required to re-reserve a time slot is highly reduced,and the data transmission failure of vehicles without reserved time slot is reduced.Simulation results show that the FCR-ALOHA protocol obtains advantages over the typical reliable slot reservation ALOHA(Reliable Reserved ALOHA,RR-ALOHA)protocol under different vehicle density and network topology change intervals.In a dense vehicle environment,Pakect Delivery Ratio(PDR)of FCR-ALOHA is 20% higher than that of RR-ALOHA protocol.In a sparse vehicle environment,FCR-ALOHA reduced the number of access collisions by 50% compared to RR-ALOHA.3.A novel Cross Layer Anti-collision Algorithm(CLAA)for transportation system information collection applications is proposed to improve the estimation accuracy of the number of vehicles as well as to reduce the algorithm complexity.CLAA reduced the complexity of Bayesian inference-based vehicle number estimation algorithm at the MAC layer,and expands the slot type at the physical layer to further improve the vehicle number estimation accuracy.Simulation results show that the CLAA algorithm can maintain high system efficiency under different initial frame lengths,which is superior to the current mainstream DFSA algorithm.When the number of vehicle tags varies from 1 to 1024,the average system efficiency of the CLAA algorithm can reach 35.64%,which is 7.3% higher than the existing ILCM(Improved Linearized Combinatorial Model)algorithm.The coefficient of variation of the system efficiency of the CLAA algorithm is 88.4% lower than that of ILCM algorithm.4.Anti-collision Selective Broadcast(acSB)protocol is proposed to reduce the delay caused by channel conflicts and hidden node issuses for basic broadcast service without common control channel.The acSB protocol utilizes special design of broadcast channel sequence and listening channel sequence,asynchronous broadcast algorithm and probabilistic forwarding algorithm to reduce delay in single-hop broadcast scenario,channel conflict caused by hidden nodes and broadcast forwarding.The acSB protocol provides a good alternative for broadcast service when the common control channel fails,and effectively solves the problem that the system cannot broadcast information when the common control channel fails.The simulation results show that the acSB protocol greatly reduces the number of broadcast conflicts,and reduces transmission delay while ensuring the stability of information transmission and information coverage.The dissertation propsed a effective spectrum resource allocation algorithm and reliable effective channel access technologies.These results provide available solutions for spectrum resource management and effective realization of vehicular networks,helping to the development of vehicular network technologies and industry.