Research On Energy Optimization In Wireless Rechargeable Sensor Networks

Internet of Things (IoT), act as key parts of next generation of information technology, which are the important fusion era of the informatization and automation. Wireless sensor networks, acts as important fundamental parts of IoT, these technologies utilize local area network or internet to connect thing with thing, thing with human, human with human, to obtain informational, automat-ic and intelligent networks, which is called "the interconnection of all things". Wireless sensor networks utilize multiple technologies to transfer information to terminal database for further data fusion and decision design.Wireless rechargeable sensor networks (WRSN) utilize multiple technologies to harvest ener-gy, could help to get rid of such stored energy limitation, as these power are harvested sustainably from ambient surroundings. Energy management and optimization are essential problems in sensor networks, and they play as key role in obtaining network optimization objects. This hot research area attracts a lot of attention from engineers and scientists, and has been witnessing many impor-tant research promotions. However, there are several fundamental problems need to be revisited, especially in the harvesting and utilization process. This dissertation focus on optimal energy distri-bution in wireless rechargeable sensor networks, combine with efficient algorithms and scheduling methods, to design optimal energy distribution schemes. These schemes aim to improve whole network utilization and provide technical guidance to system implements. The main work and contributions are summarized as follows:1. A brief review of the background, overview, and related works on wireless rechargeable sensor networks is provided.2. Research in non-contact charging scenario, on optimal movement strategy of single mobile charging resource. The sensor nodes are deployed randomly in scenario, and one single ener-gy resource is moving with point way. When the energy resource stops at a location, it stays for some duration to charge near sensor nodes. This dissertation design an algorithm with low computation complexity, to find optimal stop locations and the corresponding durations, to minimize the total charing delay. We first propose an novel optimal solution using the linear programming method. To further reduce the computational complexity, we reduce the search area to an smallest enclosure disk, and then discretize this disk into a finite number of subareas with discretizing accuracy ε. These subareas are put into the linear programming solver to obtain the optimal stop locations and the corresponding durations. Then, for system implement consideration, these stop locations are further merged into a smaller number. Our heuristic solution has a provable approximation ratio of (1+θ)/(1 - ε) by comparing with the theoretical optimal method. Numerous simulation results demonstrate the efficiency of our algorithm, and further provide guidelines for parameters setting.3. Research in non-contact charging scenario, on optimal static placement of multiple charging resources by considering limited onboard energy storage. In a general scenario with ran-domly deployed sensor nodes, we optimally deploy static energy resources to charge sensor nodes for executing tasks. We aim to deploy minimum number of energy resources, while supporting all sensor nodes to work sustainably. We first propose a binary search algorithm that yields an optimal required charging power for each sensor node. We then discretize the candidate charger placement area into finite subareas with a threshold ε. Base on the area discretizing and optimal required charging powers, we design a PTAS algorithm to find a near optimal number of chargers placed in the network. The effectiveness of our proposed algorithm has been demonstrated by extensive simulation results and system implement ex-periments.4. Research in contact charging scenario, on optimal movement strategy of single mobile charg-ing resource. We propose a novel Energy Synchronized Charging (ESync) protocol, which is a mobile charging protocol to provide energy for wireless rechargeable sensor networks. Observing the limitation of the Traveling Salesman Problem (TSP)-based solutions when nodes energy consumptions are diverse, we construct a set of nested TSP tours based on their energy consumptions, and only nodes with low remaining energy are involved in each charging round. We first design a power-a method to cluster the sensor nodes based on their energy consumption ratio, then construct a set of nested TSP tours and TSP tour selection al-gorithm. Furthermore, we propose the concept of energy synchronization to synchronize the charging requests sequence of nodes with their sequence on the TSP tours, to reduce average charging delay of sensor nodes. Experiment and simulation demonstrate ESync protocal can significantly simultaneously reduce charger travel distance and nodes charging delay.5. Research in wireless charging scenario, on optimal network-view task scheduling scheme to maximum charged energy utilization. In a general scenario with randomly deployed sensor nodes, a mobile charger moves along a fixed trajectory, to charge sensor nodes for executing given tasks. In this dissertation, we investigate the problem of assigning a given set of tasks in a wireless rechargeable sensor network while maximizing the charger’s velocity to minimize the charging delay. We first propose an online task assignment algorithm, namely Lower Bound assignment (LB), that yields a quantifiable lower bound on the charging velocity while guaranteeing a feasible assignment. This algorithm further enables the transformation of our considered task assignment problem into a variation of the classical multiple knapsack problem. We then present a fully polynomial-time approximation scheme with a (2+ε)-approximation ratio, namely ACT, that is built upon an existing greedy algorithm designed for the original knapsack problem. Numerous simulation results demonstrate the efficiency of our algorithm, and further provide guidelines for parameters setting.The conclusions are drawn with future work at the end of the dissertation.