Research On Coverage And Lifetime Optimization Of Wireless Sensor Networks Based On Evolutionary Algorithm

With the rapid development of radio frequency identification technology and wireless power transfer technology,wireless sensor networks(WSNs),which include traditional wireless sensor networks and wireless rechargeable sensor networks,have been widely used in military,social production and life.The WSN is composed of a large number of wireless sensor nodes,which can monitor and cooperate with the monitoring area in real time,and collect and process information from monitoring objects and monitoring areas.In WSNs,coverage optimization and the lifetime optimization have become the major issues in the current research.WSNs are generally used to monitor the surrounding environment.In order to prevent the occurrence of coverage holes,the monitoring area is required to be completely covered.At the same time,in order to ensure that the network can work for a long time,the lifetime of WSNs is required to be maximized.In terms of the newly emerging wireless rechargeable sensor networks,charging and energy replenishment scheduling have also become the focus of research.This thesis aims at maximizing the coverage optimization and lifetime optimization of WSNs,establishes a reasonable problem model,and proposes efficient methods based on evolutionary algorithms.Finally,reasonable wireless sensor networks can be obtained.The main work is summarized as follows:(1)The coverage optimization problem of WSNs targets at using a certain number of wireless sensors to cover a specified monitoring area.By optimizing the layout of these wireless sensors,we can maximize the coverage and minimize the redundancy.To solve this problem,a multi-agent genetic algorithm(MAGA)for coverage optimization of wireless sensor network is proposed.Firstly,the model of wireless sensor network coverage optimization is constructed.Then,by combining MAGA with virtual force algorithm(VFA),the appropriate evolution operator and objective function are designed,where the objective function is solved by Monte Carlo method with penalty function.Compared with Random algorithm and genetic algorithm(GA),it illustrates that the proposed algorithm is effective in solving the coverage optimization problem of WSNs.(2)The lifetime optimization of WSNs is to plan a certain number of wireless sensor nodes which are randomly deployed in the monitoring area so as to maximize the lifetime of WSNs.To solve this problem,a MAGA with redundant sensor rescheduling for lifetime maximization of WSNs is proposed.Firstly,this problem is transformed into a disjoint set cover problem.Then,by adopting the genetic algorithm,an efficient energy function and a redundant sensor rescheduling operator are designed.By comparing with the MAGA and the genetic algorithm with schedule transition operations(STHGA),the experimental results show that the proposed algorithm is not only suitable for point coverage problems and small-medium sized area coverage problems,but also is suitable for solving large-scale area coverage problems.(3)The lifetime optimization of wireless rechargeable sensor networks(WRSNs)targets at replenishing sensor nodes in the monitoring area to maximize network lifetime.To solve this problem,a memetic algorithm(MA)with energy replenishment strategy for lifetime maximization of WRSNs is proposed.Firstly,a local search operator and an energy replenishment strategy are designed.Then the problem is converted to the minimum set coverage(MSC)problem and solved by the MA.Compared with the traditional MA,the effectiveness of the proposed algorithm and the applicability of the obtained charging strategy are illustrated.