A New Approach for Gateway-Level Load Balancing of WMNs through k-Means Clustering

Wireless mesh network (WMN) has emerged as a key technology because of their advantages over other wireless networks. Due to the dynamic infrastructure, the traffic volume of the WMN goes in an increasing order, thus balancing the load of the network becomes very crucial. Hence the problem of load balancing is addressed in this paper and for which the cluster based architecture of WMN is considered. In this architecture, a network is subdivided into clusters and each cluster contains a cluster head. Now the problem is also subdivided into the load balancing within each cluster, which is the responsibility of the cluster head. An appropriate selection of a cluster head is very important as it performs a vital role in increasing the network performance. The paper proposes a clustering method based on k-means approach to divide the network into k clusters to manage the load in small scale and hence to reduce the overall load of WMNs. The proposed approach works at the gateway level. The simulation results show that the performance of the WMNs is improved with the proposed clustering method.