Research On Deployment Of Mobile Nodes In Wireless Sensor Networks

As an intelligent information system aggregating information collection, processing and transmission, wireless sensor networks is a self-organized network with dynamic topology, which is composed of thousands of static or mobile sensor nodes. In the application research of wireless sensor networks, proper deployment of sensor nodes is the basis and premise of network's normal operation. The deployment effect directly determines the detection quality for the target environment. As a special application of wireless sensor networks, mobile sensor networks with the advantages of adjustable node location, controllable node movement, flexible network deployment and so on opens up a new avenues to solve the nodes' deployment problem. Aiming at the optimal deployment problem of nodes in mobile sensor networks, based on the geometry split graph, this paper mainly studies the distributed deployment strategy about determining the nodes' optimized positions with part of the network information to achieve the purpose of effectively reducing nodes' deployment overhead, improving network's coverage performance and optimizing network resource's utilization rate. The main work of this paper is as follows:Firstly, based on the improved Voronoi diagram, an distributed deployment algorithm(DDA) in mobile heterogeneous networks taking into account nodes' residual energy is proposed. Selecting nodes' residual energy as the constraint parameter, the algorithm respectively determines the velocity vector of two types of nodes produced by the coverage subinterval segmentation in every moment. The simulation results show that this algorithm can improve the coverage rate and coordination speed of the network, as well as enhancing the balance of the residual energy of different nodes in the case of large-span node radius.Secondly, the paper proposes a disperse control autonomous deployment algorithm(ADA) in mobile hybrid sensor networks based on the Voronoi-Laguerre diagram. First, the algorithm puts forward the concept of Voronoi-Laguerre controlled polygon, which is constructed from the subintervals polygon adjusted by the intersection of neighbor nodes' perception circles. The target candidate point of node is ultimately ascertained according to the controlled polygon. Each node in the network updates its position by rounds to improve the network coverage. The simulation results show ADA algorithm has obvious advantages in the aspects of network coverage rate, deployment speed and nodes' distribution uniformity.Finally, aiming at the coverage problem of directional mobile sensor networks, a distributed deployment strategy(DDS) based on the virtual omni-directional perception circle is proposed. DDS uses the equal-area virtual omnidirectional circles to replace nodes' sensing sectors to determine the candidate target location of every node. By making full use of the rotation of directional nodes, DDS gets the node's perspective direction to face the outside of the detection area. All nodes adjust the perspective direction and move reasonably to improve the network coverage continuously. The simulation results show DDS has obvious advantages in improving the coverage rate of network and enhancing the nodes' distribution balance.