Research On Energy Management Strategy In EH-CRSN

Energy Harvesting Cognitive Wireless Sensor Network(EH-CRSN)is a novel network which introduces cognitive radio(CR)technology and energy harvesting(EH)technology into traditional wireless sensor network(WSN).CR technology alleviates the problem of spectrum shortage in WSN,but sensor nodes must consume a large amount of energy to support CR functions,such as channel sensing and spectrum switching.EH technology can supplement the energy of sensor nodes and extend the life cycle of EHCRSN.The introduction of EH technology has made the development of EH-CRSN.However,it brings challenges to energy management.In order to extend the life cycle of network as much as possible and improve the energy efficiency of the nodes,it is necessary to study energy management strategies of nodes deeply.Firstly,according to the characteristics of the limited battery capacity and battery capacity degradation of the sensor node in EH-CRSN,this thesis studies how to determine the sampling rate of each energy harvesting sensor node under the condition of battery capacity degradation to maximize the network utility of the sensor node in its life cycle.To solve this problem,an adaptive sampling rate control algorithm(ASRC)is proposed by establishing a mathematical model.The algorithm adaptively adjusts the sampling rate according to the current battery level under the condition that the node meets the battery capacity constraints and link capacity constraints,and effectively manages the energy use of the node to optimize network performance and extend the network life cycle.The algorithm considers the impact of battery imperfections on performance,which is more realistic.The simulation results show that the ASRC algorithm can maximize the network utility and improve the overall performance of the network under the premise of maintaining energy sustainability of the node.Secondly,this thesis considers that the existing channel selection algorithms in EHCRSN lack the process of dynamic learning,so a channel selection algorithm based on Qlearning is proposed.Under the condition of non-ideal spectrum sensing,the algorithm introduces energy efficiency into the reward function by establishing a Q-learning framework.The node can interact with the channel environment and learn from it,and select the channel with the largest long-term cumulative average reward as the priority to sense and try to access.In addition,from the perspective of large energy consumption of sensor cognitive function,a scheme is proposed to collect energy by using radio frequency signals when the channel is busy to supplement the energy of the node and extend the life of the node.Simulation results show that compared with the existing channel selection algorithm,this algorithm can effectively improve the energy efficiency of spectrum sensing and adapt to the dynamic radio environment.