Research On Key Technologies Of Energy Efficiency Optimization For Wireless Rechargeable Sensor Networks

Internet of things(IoT)technology is improving the quality of people’s life and has played an important role in smart home,intelligent transportation,environmental monitoring,public safety and other fields.As an important part of IoT,wireless sensor networks(WSN)can collect a large amount of environmental information with low cost and pro-vide a reliable data source for IoT.However,the limited battery power of sensor nodes(SN)limits the development of WSN.With the help of wireless charging technology,wireless rechargeable sensor networks(WRSN)can charge SN by using mobile charger(MC),which solves the problem of limited WSN lifetime.How to improve the energy efficiency of WRSN has attracted more and more attention.In this thesis,the key technologies and factors for WRSN energy efficiency opti-mization are studied.The research is carried out in two aspects: routing protocol and MC charging path planning.The energy efficiency oriented heterogeneous paradigm routing(EEHP)protocol and policy gradient path planning(PGPP)algorithm are designed and implemented.The effectiveness of the algorithm is proved in simulation experiments.The main work of this thesis is listed as follows.Aiming at the problem that the current research on routing protocols in wireless sen-sor networks is mainly focused on non-rechargeable WSN,and there is no routing protocol that can give fully leverage its technical advantages according to the features of WRSN.This paper designs EEHP protocol which based on the heterogeneous network structure.In this protocol,SN contains normal sensors(NS)and cluster heads(CH).The CH with high battery capacity transmit the data of SN in its cluster to the base station in a multi hop manner.The protocol uses a lowest transmission cost algorithm to plan the transmis-sion path for CH.By transmitting data with this path,CH can reduce energy consumption and improve the energy efficiency.The protocol can also transfer the energy consump-tion workload to the CH closer to the base station,which is convenient for MC to charge these SN,and can further improve the energy efficiency of the network.Comparative experiments proves the effectiveness of EEHP protocol than other protocols.The charging path planning of MC is a NP-hard problem.The task is to plan a path for MC that can not only charge the SN in time,but also reduce the driving distance,so as to reduce the energy loss of MC.In this thesis,the method of deep reinforcement learning is used to solve this problem,and the PGPP algorithm is proposed.In this algo-rithm,the SNs in WRSN are seen as the environment,and the base station and MC are seen as the sensors and actuators of the agent.A reward function which can measure the energy efficiency of MC is designed,and the policy gradient algorithm is used to learn the strategy.Comparative experiments show that compared with other MC charging path planning algorithms,this algorithm has certain advantages in the speed of path planning and the energy efficiency of WRSN.This thesis designs a simulation environment for WRSN.A variety of routing pro-tocols and MC charging path planning algorithm are simulated in this environment.The process of the simulation environment uses the event scheduling method to drive the op-eration of the system,which can calculate the state transition information of the environ-ment in a short time.The system can directly jump to the latest state,which shortens the time required for the simulation experiment.The simulation environment can measure the performance of different WRSN routing protocols and charging planning path algorithms,and improve the efficiency of the research on WRSN.