Studies On The Coverage And Connected Optimization Algorithm In WSN

Along with the recent breakthrough of computing and sensor technologies, Wireless Sensor Network (WSN) has become a very popular research topic in academy as well as a cutting edge technology in industry. It has attracted a lot of interests from different areas. The reason is that not only it has been used in many real world applications, but also it is believed to have great potential yet to be explored. The two fundamental research problems in WSN are the monitor area coverage and the network connectivity. These two problems will have directly impact on whether the WSN functions or not. Moreover, they will affect the network’s Quality of Service (QoS) and the survival time. Therefore, the further investigation of the above two problems will contribute both theoretical significance and practical value.This dissertation will explore innovative technologies to improve the above two problems, i.e. use the energy efficient blind spot coverage control to mitigate the monitor area coverage problem and apply mobile deployment optimization on the sensor nodes to improve the network connectivity. The dissertation contributions can be summarized as the following points:1. In the deterministic networks, work has studied the optimal deployment and the communication problems in the deterministic deployment. A node deployment method has been proposed using the regular hexagon mesh. The benefit is that not only the nodes can achieve the seamless coverage for the monitoring area, but also the required number of nodes can be reduced to the minimum. The result shows that the regular hexagonal grid method requires fewer nodes-and consumes less energy. This can be proved both using mathematical theoretical analysis and by comparing its result with the one in simulation. In the random deployed networks, the network monitoring area has been divided using the regular hexagonal grid. Hence, the full coverage problem for randomly distributed nodes can be analyzed, the probability relationship of completely seamless coverage can be counted by dividing the nodes into grids of area, and the relation of the number of nodes and detection area can be counted by the Poisson distribution model. The proposed method has improved the theoretical properties:the optimal coverage problem of the seamless network will have a higher theoretical and practical guidance value.2.To accomplish the re-deployment of nodes with mobility in the network, the network coverage can be improved by reducing the blind area. Through analyzing the characteristic and properties of the Vor onoi method, the dissertation presents a method to approximately estimate the blind area. Moreover the dissertation proposed two algorithms that can be used to re-deploy the nodes with mobility:CBAH is suitable for the cluster network and DLBAH can be applied in the distributed network architecture. Thanks to simulation, the result shows that both CBAH and DLBAH can effectively reduce the blind area in random networks. Compared with the VOR algorithm and the VOR-MO algorithm, both the network coverage ratio and the average node moving distance can be improved.3. Regarding the problem of how to effectively monitor the multiple targets in the sensor network, this dissertation has proposed a multiple objective collaborative monitoring algorithm (MOCMA). This method has considered the relationship between the nodes in the network and nodes that have been monitored, and it aims to contribute to the problem of monitoring multiple targets in an effective fashion. In the MOCMA method, the data mining method with the association rules will be applied first. This is followed by cooperatively controlling the work status based on the previous data mining result. The benefit will achieve a better quality of monitoring as well as improving the energy consumption by the network nodes. The MOCMA method has also considered the cooperation among the network nodes, so that the nodes can be self-scheduled in the most effective way. This effectively improves the network survival time and the network coverage ratio.The dissertation is supported by the National Natural Science Foundation of China (No.60672137,61171075), the Specialized Research Fund for the Doctoral Program of Higher Education of China(No.20060497015), the State Key Laboratory of Software Development Environment (No.SKLSDE-2009KF-2-02),and the New Century Educational Talents Plan (No.NCET-08-0806).