Abstract
The most challenging task in wireless sensor network is energy efficiency, as energy is the major constraint in the wireless sensor network to improve the life time of the network. Hence developing algorithms to improve network life time is the major task. In wireless sensor network most of the energy is wasted while gathering the data, hence an efficient algorithm which conserves energy has to be designed. Thus our proposed work A Novel Data Gathering Algorithm for Wireless Sensor Networks using Artificial Intelligence (NDGAI) uses mobile element and deals with the conservation of energy while gathering the data. Appropriate clustering, cluster leader selection and proper path determination of mobile element helps to conserve energy and improve the over all network life time. In our proposed work initially the clusters are forged by using Amended Expectation Maximization(AEM) algorithm, which is the maximum likelihood estimate. It is used along with Gap statistic method to find the optimal number of clusters. AEM algorithm helps in obtaining the centres of the cluster with maximum number of nodes near the cluster centres. For each cluster, Cluster Leader (CL) is selected by using Fuzzy Logic. Fuzzy logic selects the node which is near to the cluster centre by using parameters such as Closeness of node to the Cluster Centroid, direction of node towards base station, number of Neighbouring Nodes. After the CL’s are determined, to reduce the path length virtual points(VP) are selected so that mobile element reaches this virtual point and collects the data. These VP’s are selected only when the CL has data in it. The mobile elements can reach these virtual points intelligently by using optimal path,that is obtained by using hybrid of Particle Swarm Optimization and Artificial Bee Colony algorithm. Thus the mobile element travels in the optimal path and gathers the data from the entire network intelligently and efficiently with less amount of energy. With this approach the performance and life time of the network is improved while gathering the data. The simulation results are compared with Scalable Grid-Based Data Gathering Algorithm for Environmental Monitoring Wireless Sensor Networks (SGBDN) and proved that the proposed method is better than SGBDN .