Research Of Reliable Deployment Scheme In Wireless Sensor Network

With the rapid development of microelectronic technology, wireless communication technology,computing technology and embedded processing technology, the wireless sensor network (WSN)comes into being studied widely by academia and industry. The node deployment problem is one ofthe basic problems of WSN, which is the foundation of network operation and applicationimplementation. Sensor nodes may be arbitrarily distributed in the configuration area, which reflectsthe status of monitoring and tracking of a region in a WSN. Node deployment is of great significanceto improve the viability of a WSN, and it is the key to plan and design the WSN. The reasonabledeployment would largely improve the performance of the entire network and expand the coverage ofthe monitoring area. Based on the theme, this thesis mainly focused on two topics as follows:First, the fault tolerance repair deployment issue in randomly deployed WSN is investigated.Connectivity is a critical issue in WSN, only when the network is connected the collected data cantransmit to the viewer successfully. At the same time, in order to ensure the reliability of the network,fault tolerance against node failures is also critical. Based on the above two factors, this thesis designsa repair deployment algorithm to guarantee a specified level of connectivity (such as k-connectivity)and fault tolerance characteristics. Initially, a network model based on graph is established. Thenbased on the network model, a partial k-connectivity repair algorithm and a full k-connectivityrepair algorithm are proposed to repair the connectivity of the WSN, respectively. After that, thecomputational complexities in terms of both running time and memory space are given. Finally,simulation results show that the proposed two repair algorithms can consistently offer a betterperformance gain over other existing algorithms.Second, the optimal deployment problem of line WSN is investigated. In order to improve thelifetime of the line WSN and avoid the “energy hole” problem brought by uniform deployment, thisthesis proposes an optimization deployment strategy based on a particle swarm model. Initially, a linenetwork model is established which is based on flow constraints, and a node optimal deploymentproblem is formulated according to the model. In the optimization problem, the theoretical expressionof the network lifetime is used as the objective function. Then, by using the intelligent optimizationtheory, a particle swarm optimization (PSO)-based network deployment scheme is harnessed to tacklethe proposed optimization problem. In the optimization process, the optimal solution is regarded asthe final configuration position of a node. Finally, simulation results show that the proposed PSO-based network deployment scheme can consistently offer a better performance gain over otherexisting network deployment schemes.