Topology Stability-based Clustering for Wireless Mesh Networks

In the past, many clustering algorithms for adhoc networks have been proposed. Their main objective is to solve the scalability issue of ad-hoc networks by grouping nodes into clusters. The challenge in MANETs for those clustering algorithms is to cope with the high node mobility which affects the stability of the cluster structures. Wireless mesh networks consist of a static backbone and a number of mobile nodes. In the backbone of a wireless mesh network the topology is relatively static. However, topology changes occur due to frequent link losses and temporary link instability. Due to the static nature of the backbone, mobility-based approaches are not suitable in this case. In this paper, we state the important aspects for stable clustering in wireless mesh networks with unidirectional links based on the investigation of a 45-node wireless mesh testbed. We analyze well-known clustering algorithms and their performance in a large-scale testbed. Finally, we propose a new clustering algorithm called Stable Link Clustering Algorithm (SLCA).