A Routing Computation Strategy Based on Improved Clustering Algorithm

A routing computation strategy based on improved clustering algorithm is proposed in this paper. It uses a clustering algorithm based on K-means and Dijkstra (CAKD) in topology partitioning. The routing information of each node can be computed during the clustering process, and CAKD reduces the time cost in routing computation and the storage in storing routing information. Experiments showed that, compared with using Dijkstra to cluster, CAKD cost less time. Meanwhile, compared with the strategy of global shortest path, the routing computation strategy based on improved clustering algorithm costs less storage in storing routing information.