Wireless Sensor Networks Based On Particle Swarm Optimization, Self-organizing Clustering Algorithm

With the development of modem networks and wireless communication technology, Micro-Electro-Mechanism System (MEMS), Micro-sensor technology, and artificial intelligence, Wireless Sensor Network (WSN) has been developed rapidly. Recent years, wireless sensor network has been applied to a lot of domains, such as country security, environment detecting military affairs reconnaissance, traffic management, and agriculture etc.. In WSN, sensors are characterized mainly by small size, dense deployment, and limited energy, those characters baffle the development of WSN. Additionally, a new cluster algorithm which could save node energy has been proposed. Cluster algorithm is a strategy which could divide network into a number of clusters, cluster algorithm uses the method of cluster head competition in the each cluster, which manages the network topology, and uses the data fusion to compress transmission data. Cluster algorithm becomes the primary strategy to save energy of network.In order to prolong the lifetime of the WSN, it is essential to design a self-organized clustering algorithm to accommodate the characteristics of sensor network. This paper proposes a kind of load balancing and energy efficiency clustering model for WSN according to the characteristics of WSN, which optimizes clustering and route based on the energy of sensor and distance between sensors. The paper takes advantage of PSO to optimize the problem of cluster. In order to optimize the model in the help of PSO, this paper designs a type of updating mechanism about the position vector and speed vector of particles, and penalty function strategy is introduced to solve two primary constraints. The simulation results show that our algorithm greatly improves the performance and efficiently saves the energy of WSN in the help of MATLAB 7.1.