Optimization Design Of Charging Model In Wireless Rechargeable Sensor Networks

Traditional sensors are powered by a compact button cell,and due to the limited capacity of the button battery,the network lifetime is limited by the sensor's battery capacity.In recent years,with the innovation and breakthrough in wireless energy conversion technology,the electromagnetic coupling resonance technology appeared which can transfer energy through wireless to a certain distance within the scope of the receivable devices.Aiming at the problem of wireless charging in wireless sensor networks,many scholars proposed mobile charging models and methods,including mobile charging schemes based on point-to-point charging and point-to-multipoint charging models.For point-to-multipoint charging model,mobile charger(MC)can transfer energy to multiple target devices within its charging range concurrently at a selected landmark location.For point-to-point charging scenario,MC can transfer energy to only one target devices within its charging range.Scholars use different charging models to achieve their respective charging optimization goals.In this paper,we design and implementation the charging schemes based on point-to-point and point-to-multipoint charging model.(1)Design the charging scheme based on point-to-point charging model.This paper explores the problem of maximizing network lifetime under the point-to-point charging model by one MC.Since there is only one MC in sensor networks,the network size cannot be too large.We firstly analyze the maximum and minimum size of the network in this paper.Secondly,we propose an effective charging scheme under a given size,which includes charging sequence and charging time.As for the problem of charging sequence,by using the residual energy of the node and the distance between the nodes as the weight,we convert the charging sequence problem into a class TSP problem.As for the problem of charging time,we aim to maximize the charging time,and the residual energy of the node should to meet the given threshold as the constraint,convert the charging time problem into an optimization problem to calculate charging time of each sensor node.(2)Design the charging scheme based on point-to-multipoint charging model.Under the point-to-multipoint charging model and the limited capacity of MC,we first analyze the size of the network,that also means,the maximum number and the minimum number of sensor nodes can be charged by one MC.Secondly,we use Set Covering Problem to determine the stop positions while MC traveling in the network,and get the shortest path of MC by Hamiltonian circuit.Then,under the goal of maximizing the network lifetime,the residual energy of the node at any time is higher than the given threshold be a constraint,we convert the problem of charging time into an optimization problem.Finally,we determine the opening sectors of MC at each stop points.(3)We code these two charging schemes of point-to-point charging and point-to-multipoint charging model on Eclipse platform.A lot of comparative experiments are carried out,and we investigate the effects of different experimental parameters on the experimental results.The result demonstrate the effectiveness and feasibility of our charging schemes.