An improved k-means clustering algorithm over data accumulation in Delay Tolerant Mobile Sensor Network

Delay Tolerant Mobile Sensor Network (DT-MSN) possesses high delay tolerability. And the real-time requirement of data is reduced, which results in data packet accumulation. The limitations of nodal buffer and great data accumulation have created data management problem. Thus, it seems particularly important to complete the task of quick analysis of the collected data in DT-MSN. To solve the problem of data accumulation, an improved k-means clustering algorithm is proposed based on linear discriminant analysis (LDA), namely LKM algorithm. In the algorithm, we firstly apply the dimension reduction method of LDA to change the high-dimension dataset into two dimensional dataset, then we use k-means algorithm for clustering analysis. Simulation results show that LKM algorithm shortens the sample feature extraction time, and improves the accuracy of k-means clustering algorithm, thus enhancing the performance of k-means clustering algorithm to analyze and process vast data.