A Study On Energy Efficiency Of RSU For Vehicular AD HOC Networks

Vehicular Ad Hoc Network (VANET) is a special type of of mobile ad hoc networks for vehicular communcation. In recent years, it has received a great amount of attention from both industry and academia because of the road safety and environmental pollution problems. A VANET consists of moving vehicles installed with on board units (OBUs) and roadside units (RSUs) deployed along a road, and can be used to provide safety and multimedia services for vehicles. This thesis studies the RSU energy saving problem in VANET, and proposes several RSU scheduling mechanisms or algorithms for different road scenarios in order to increase the energy efficiency of RSUs.First, a best position based scheduling mechanism (BPSM) is proposed for delay tolerant services in a general road scenario. The proposed scheduling mechanism chooses the best time for the communication between a RSU and a vehicle based on the distance between the RSU and the vehicle in order to save the energy of the RSU. Simulation results show that BPSM can effectively improve the energy efficiency of a RSU. Meanwhile, considering that a RSU with limited energy cannot work long when its traffic load is heavy, a Step Probability Scheduler (SPS) mechanism is proposed to balance the residual energy of RSUs and thus extend the working lifetime of the RSU system. Simulation results show that SPS can effectively balance the energy consumptions of RSUs.Then, a simple but efficient ON/OFF scheduling based RSU energy saving mechanism is proposed for a remote road scenario. In this algorithm, when no request is received, an RSU will alternatively work in an ON state and in an OFF state; Once a request is received, it will extend the ON state for a certain period of time. To analyze the performance of the proposed algorithm, a mathematical model is established to calculate the request acceptance probability and energy efficiency of an RSU. Simulation results show that the mathematical model established is basically accurate and thus can be used to analyze the performance of the proposed scheduling mechanism. Meanwhile, the proposed scheduling algorithm can effetively improve the energy efficiency of an RSU.Finally, a Least Load off First (LLOF) scheduling algorithm is proposed for a city road scenario. The algorithm improves the energy efficiency of the RSU system at the cost of the average service delay. It evaluates the impact of an RSU on the system performance before deciding whether to close the RSU. As the traffic loads of RSUs are different, the algorithm performs the evaluation starting from an RSU with the lowest traffic load until all RSUs are evaluated, and finally obtain the RSU scheduling result. Simulation results show that LLOF can effectively improve the energy efficiency of the RSU system.