Research On WSN Routing And Clustering Algorithms Based On Particle Swarm Optimization

Wireless sensor network is composed of a large number of micro sensor nodes which are randomly or manually deployed in monitoring regions. Wireless sensor networks have been widely used in disaster monitoring, environment monitoring, defense investigation systems, health care and so on. Much attention have been attracted by wireless sensor networks due to widespread application. However, because of their typically deployment in unattended monitoring area, the energy of sensor nodes can not be replaced or supplemented. Therefore, it is important for wireless sensor networks routing and clustering algorithms how the limited energy to be effectively used and the network lifetime to be prolonged.This thesis is based on wireless sensor networks. In order to prolong the network lifetime and balance network load, the existing clustering and routing protocols have been analyzed and compared with. Combined with the particle swarm optimization,much reaserch has been made on the WSN routing and clustering algorithms. The main contents can be included as follows:(1).A centralized, energy-balanced and uneven clustering and routing protocol, which are based on particle swarm optimization, has been analyzed. In this protocol, inter-cluster multi-hop routing is combined with non-uniform clustering mechanism. Unlike other routing protocols, a special node is used as cluster head. Based on particle swarm optimization, an effective particle encoding scheme and multi-objective function have been used in inter-cluster multi-hop routing algorithm. As the energy consumption and latency of cluster head node have been taken into account in the inter-cluster routing algorithm, the energy consumption and latency of cluster head have been balanced. Uneven clustering is also based on particle swarm optimization. The particles are encoded as a complete cluster solution. The energy consumption of sensor nodes and cluster head nodes have been taken into account in the clustering algorithm. As the simulation results shown that the clustering algorithm make network lifetime improved, the network energy consumption balanced and latency reduced.(2).Compared with previously analyzed algorithm, the performances of newly analyzed protocol have been greatly improved. However, in order to make the lifetime of network improved, the lifetime of the cluster head node must be improved. Since the particle swarm optimization is easy to be fallen into local optimum and differential evolution algorithm has a good global search capability, a hybrid algorithm of particle swarm optimization and differential evolution has been used in the inter-cluster routing stage to improve the life of the network and reduce the latency of the network. For large-scale wireless sensor networks, convergence of the algorithm is important for an algorithm in the clustering stage. In order to accelerate the convergence of clustering algorithm, the particle swarm optimization should be improved. Before the position of particle is updated, a local improvement stage is introduced. Convergence of the algorithm has been accelerated by introducing local improvement stage.