Research On Optimization Methods For Coverage In Wireless Sensor Network

The coverage of wireless sensor network(WSN)is considered as the primary problem in vast applications,and it determines the service scope of WSN.Besides,the coverage optimization is a support technique of WSN.Thus,based on the analysis and conclusions of domestic and overseas researches,the coverage optimization algorithms are studied in wireless sensor networks in this dissertation.The major research contents and productions are shown in the following areas:To solve the hole repair of area coverage problem in WSN,we first consider the probabilistic detection model with false alarm rate,which is a more realistic sensing model in the application.Then the criteria of coverage holes for the requirement of network service quality is defined.To repair the coverage holes,two algorithms are proposed.The first algorithm is the incremental repair algorithm,which mainly considers the lowest coverage probability in the region of interest.And the second algorithm is minimum repair algorithm,which mainly considers the minimum sensor nodes to repair the coverage holes.Both the two algorithms solve the area coverage problem successfully.To solve the hole repair of barrier coverage problem in WSN,a KM Algorithm for minimum movement of mobile sensors is proposed.The aim of the algorithm is to reduce the energy consumption of sensor nodes and mend the barrier gaps.Besides,combined the coverage probability and network energy,a criterion of the network maintenance indicator is established for barrier coverage.Based on the network maintenance indicator,this dissertation proposes two algorithms,which mainly consider the minimum sensor nodes and the minimum energy consumption,respectively.Both the two algorithms solve the barrier coverage problem successfully.To solve the target-barrier coverage optimization problem in WSN,this dissertation first proposes a determined deployment algorithm,which is named as target-barrier based on minimum nodes algorithm.The algorithm constructs the target-barrier circle of a single target.And if there are many targets in the ROI.target-barrier circles can be merged,or some circles are deleted by the algorithm.Next,a greedy algorithm is considered for rpairing target-barrier coverage with mobile vehicle,which solves the target-barrier coverage problem successfully.To solve the probabilistic coverage optimization problem in directional sensor network(DSN),probabilistic sensing models are considered for DSN to have practical considerations.The three probabilistic sensing models are exponential decay probabilistic sensing model,the fault-tolerant probabilistic sensing model and the hybrid probabilistic sensing model.Then,based on our probability sensing models,a probabilistic coverage algorithm is introduced for random deployment to compute the coverage probability of DSN.Since the directional sensor node can relocation the sensing direction,a second order prticle swarm optimization algorithm is proposed to repair the probabilistic coverage problem in DSN.To solve the three-dimensional(3D)barrier coverage optimization problem in camera sensor network(CSN),a 3D probabilistic sensing model for the camera sensor is first proposed.The sensing model defines a new realistic resolution criterion for the crucial feature of the intruder.Based on the 3D probabilistic sensing model,this dissertation studies the deployment strategy,coverage probability and network robustness of barrier coverage in CSN.Performance results demonstrate that our barrier coverage with more practical considerations can provide a desirable surveillance level.Moreover,compared with local face-view barrier coverage and full-view barrier coverage,the proposed barrier coverage is more reasonable and closer to reality.In this dissertation,the coverage optimization problem of WSN is systematically researched.A series of sensing models of sensors and optimization algorithms are proposed for different coverage problems.In comparison with other correlative methods through the simulation experiments,the proposed algorithms are verified to be feasible,available and advanced.