Study On Clustering Routing Protocol In Large-scale Wireless Sensor Networks

Wireless sensor network is a self-organizing wireless communication network composedby a large number of sensor nodes. It can be used to perceive the objective world and transferinformation. As a completely new research field, its huge potential and broad applicationspace have attracted wide attention of scholars from a variety of countries. As one of the mostimportant part of wireless sensor network, the design goal of routing protocol is highreliability, low cost and easy to maintain. In this paper, we do a full research of the flatrouting protocol and hierarchical routing protocol. At the same time, we analysis theadvantage and the disadvantage both of the two kind of typical routing protocol. The mainresearch work in this paper can be summarized as the follows:(1)Inspired by data clustering algorithm, this paper proposes a clustering algorithmnamed APBCS that based on Affinity Propagation. In APBCS clustering algorithm, a newdefinition called node competing strength is presented, the three parameters, i.e., the ratio ofnodes residual energy, the distance of nodes and the neighbor nodes density, are considered tobecome the cluster heads. Furthermore, the algorithm uses a ring based multi-hop clusterheads communication. The Simulation results show that, compared with LEACH algorithmthe cluster head selection is better reasonable in APBCS algorithm.(2)By learning the chain strategy of PEGASIS, this paper proposes a clustering routingprotocol named CRPBIC by means of using Hybrid Particle Swarm Optimization. CRPBICtakes advantage of intelligent algorithm to design the chain between clusters in wirelesssensor network. The Simulation results show that, compared with LEACH algorithm andAPBCS algorithm, CRPBIC protocol prolong the lifetime of the network efficiently.(3)When used individually in the large scale WSN, CRPBIC protocol can’t performancevery well. With the increase of the network size and the number of sensor nodes, thealgorithm complexity and operation cost will increase at the same time. This limits its scopeof application especially in large-scale wireless sensor networks. Therefore, this paperimproves CRPBIC protocol. Firstly, the network is divided into different areas, each area usesCRPBIC algorithm to select the cluster heads at the same time, and then the cluster headscomposes a chain by using improved Hybrid Particle Swarm Optimization. At last, the chainhead in different area will form a routing tree according to certain rules. The simulation showsthat after improving the CRPBIC protocol, it is more feasible in a large-scale network. Itprolong the lifetime of the network efficiently.