Research On Omnidirectional Chargability In Wireless Rechargeable Sensor Networks

With the wide application of Wireless Sensor Network(WSN),more and more literatures studied the WSN,many of which focused on power supply.As we all konw,battery is still the bottleneck of mobile devices and frequent battery replacement consumes a lot of manpower,physics and financial resource.In this context,Wireless Rechargeable Sensor Network(WRSN)emerged.In WRSN,charger node provides en-ergy for sensor node by wireless power transfer.The change of energy source greatly increases the reliability,flexibility and expansibility of the network.This paper focuses on omnidirectional charging in WRSN.The notion of omni-directional charging was first proposed by us by which an area is omnidirectionally charged if a device with directional antennas at any position in the area with any ori-entation can be charged by directional chargers with power being no smaller than a given threshold.In this paper,we mainly address two important problems:one is to detect whether the target area achieves omnidirectional charging given a deterministic deployment of chargers,the other is to determine the probability that the target area achieves omnidirectional charging given a random deployment of chargers.We make the following four key contributions:1.To the best of our knowledge,we are the first to propose and study the omnidirec-tional charging problem,and establish the empirical directional charging model involving chargers and devices.Omnidirectional chargability is extensively ap-plied to millimeter wave cellular networks and wireless rechargeable sensor net-works.2.We propose an efficient method to detect whether a target area achieves om-nidirectional charging under a given charger deployment.At first,we develop piecewise constant approximation and area discretization techniques to partition the target area into subareas and approximate powers from chargers as constants.Thus we only focus on the Omnidirectional chargability of subareas.Then we propose the Minimum Coverage Set extraction technique which reduces the con-tinuous search space to a discrete one and thereby allows a fast detection algo-rithm.Finally,if all the subareas achieve omnidirectional charging,the target area achieves omnidirectional charging,or the target area does not achieve om-nidirectional charging.3.We derive an upper bound of the omnidirectional charging probability for a ran-dom deployment.We first replace the target area by grid points on triangular lat-tices and prove that if chargers' power with reasonable relaxation achieves the omnidirectional charging of grid points,then the target area achieves omnidirec-tional charging.Then we obtain an upper bound of the omnidirectional charging probability for a random point in the area,and finally derive the omnidirectional charging probability for all grid points and the target area.4.we conducted simulations and field experiments to verify our theoretical find-ings.The results show that our algorithm outperforms comparison algorithms including an adapted algorithm based on the full-view coverage detection algo-rithm in wireless sensor networks by at least 1.2×,and the consistency degree of our theoretical results and field experimental results is larger than 93.6%.