Relay Selection Algorithms In Vehicular Networks

Vehicular ad hoc networks(VANETs)are capable of supporting the communications between vehicles,and between vehicles and infrastructure.As an important component of intelligent transportation system(ITS),VANETs are expected to achieve intelligent traffic control,traffic safety alarm and enhance the transmission performance of vehicular users,and have broad application prospects.In VANETs,in the case that the source vehicles(SVs)cannot communicate directly with the access point(AP),they may need to select the relay vehicles(RVs)for information forwarding.However,the mobility of vehicular users,the complicated characteristics of the communication links,the dynamic changes of the network topology and the diversification of user services pose difficulties and challenges to the relay selection algorithm design for VANETs.In this thesis,we study the relay selection algorithms in VANETs.Considering the VANET scenario which consists of a number of SVs and its corresponding AP,and assuming that the SVs are transmitting throughput-sensitive service and delay sensitive service,in this thesis,a user service characteristic-oriented relay selection algorithm for is proposed.Firstly,the candidate RV set is determined according to the SVs service characteristics,then,by applying the network calculus theory,the arrive curve of SVs and the service curve of the candidate RVs are modeled and the transmission performance of each SVs through the candidate RVs is evaluated.Jointly considering the transmission performance of all SVs and RVs,we define the function of joint network transmission performance and formulate the RV selection problem as the optimization problem which aims at maximizing the network transmission performance.The optimization problem is solved based on the Kuhn-Munkres algorithm.Stressing the VANETs application scenario that one RV is allowed to forward data packets for multiple SVs,we propose a relay selection algorithm based on joint utility function optimization.Jointly considering the transmission performance of all SVs and RVs,we define the utility function which is a function of effective throughput and link characteristics.Under the constraints of SV service requirement and RV resource sharing,the joint utility maximization problem is formulated,and then solved by using the Lagrangian dual method to obtain the optimal RV strategy.