Research Of Holes Detection And Redeployment In Wireless Sensor Network

As one of the fundamental issues in wireless sensor network (WSN), coverage is one of the key points which affect the network performance and the Quality of Service (QoS). According to the deployment way of nodes, it can be classified as deterministic or random deployment. In addition, battery depletion or intentional destruction of the sensors may result in non-uniform sensor deployments. Due to this non-uniformity, coverage holes may exist in networks. Thus, it’s very important for detecting and repairing coverage holes of WSN. The methods of detection and re-deployment coverage holes are studied in this paper. The main research achievements are as follows:The sensing model of the nodes is inevitably affected by the environment factors, therefore the probabilistic detection model with false alarm rate is proposed. A method of using an unmanned aerial vehicle (UAV) is proposed for boundaries of holes detection and redeployment, for the advantages of high accuracy, saving energy and no environmental impact in deployment. In this thesis, the architecture and communication model of UAV and WSN is designed.The re-deployment strategy based on depth-first of probability is investigated. The joint detection probability iso-sensing map is established by the data center in the UAV. In particular, according to the set minimum threshold of the joint detection probability, boundaries of holes are detected. The new sensor node is deployed in the lowest position of joint detection probabilities within the communication radius of node. An experiment is presented to verify the performance of the proposed algorithm. Contrast with different methods such as random deployment and MAX_MIN_COV, the proposed method attains higher detection probability.The re-deployment strategy based on the Delaunay Triangulation is researched. The joint detection probability iso-sensing map is established with UAV. Furthermore, binary image is formed by the data center. Thus, the size and the position of hole are detected. The UAV flight path problem is transformed into Traveling salesman problem with genetic algorithms, and a deployment method based on Delaunay Triangulation is applied for the uncovered regions. Simulation results show that the coverage rate has been efficiently improved by the proposed method.