Research On Call Admission Control In Infrastructure-based VANET

With the increase of service demand in wireless network, the types of services has become diverse. Multimedia services are sensitive to delay, and have important influence on user experience. In infrastructure-based VANET, because of vehicles mobility and limited resource, the connection time of wireless link between a vehicle and the infrastructure is too short, which introduces high packet loss rate. As a result, the network performance and quality of service is degraded. Therefore, how to ensure the quality of service and allocate the limited resource reasonably is an important research topic in VANET. Admission control mechanism, as an important part of resource management, is a key technology to ensure quality of service.Based on the analysis of some traditional classic admission control algorithms, this thesis proposes a mobility-based call admission control (MB-CAC) algorithm by integrating characteristics of VANET. To reduce the number of control frames and the load of AP, MB-CAC adopts the beacon frame sent by AP to carry resource usage information. After vehicles obtain the usage of resource from the beacon frame, a vehicle initially determines whether its service request can be accepted by AP. If so, the vehicle sends a service request control frame, or else refuses to send a service request. In order to ensure high-priority service request is accepted by AP, low-priority services set smaller values of the channel idle time, AIFS, and contention window. When receiving a service request, the AP differentiates the type of service according to its initial time. If it is handoff-call, the AP performs handoff-call admission principle. Otherwise, the AP carries out new-call admission principle. In handoff-call admission principle, to ensure the priority of handoff-call, the AP reserves bandwidth resource for it according to adjacent AP admission ratio and handoff-call arrival probability. In new-call admission principle, because the number of accepted new-calls has influence on handoff-call, the AP sets accepted threshold K for new-calls.By means of Markov chains analysis and simulation, the thesis evaluates the performance of MB-CAC in terms of handoff-call dropping probability (HCDP), new-call blocking probability (NCBP), channel utilization, and throughput. Analytical and simulation results demonstrate that MB-CAC algorithm has advantages over HCDP and NCBP by maintaining high channel utilization and throughput.