A Grid-Based Approach To Continuous Clustering Of Moving Objects

With the rapid advances in wireless devices and positioning technologies, tracking and clustering of moving objects has drawn increasing attention. Previous methods of clustering moving objects merge clusters by searching all the existing clusters, which have an obvious decline in efficiency as the number of clusters increases. This paper proposes a gridbased approach to continuous clustering of moving objects. We first employ dynamic grid to narrow the searching area when merging clusters, and then develop an efficient split algorithm to handle the split of clusters, which avoids multiple splits of one cluster during a period of time. At last, a comprehensive experimental evaluation has been conducted to validate our approach, and the results indicate the efficiency and effectiveness of our algorithm, especially for large data set.