Self-organized clustering algorithm based on regional relevance in wireless sensor network

In view of limited resources and amounts of data in Wireless Sensor Network(WSN),clustering compressions are always employed to reduce transmission packets.Based on the regional relevance,a self-organized clustering algorithm was proposed for the wavelet compression in sensor network.The algorithm used the correlation of actual regional data for clustering.Actually,during the wavelet data compression procedure,the data correlation was detected in the cluster head,so that to insure better correlation and adjust the clusters structure.Meanwhile,data dependence among the nodes was analyzed in Sink,to form large-scale clusters with better relevance,so that to improve the efficiency of wavelet compression in a long period.Theoretical and experimental results show that the proposed algorithm can eliminate the data redundancy by using temporal and spatial correlation as much as possible,improve the efficiency of wavelet data compression,and reduce network energy consumption.