CBLA: Cluster-based load-aware channel assignment for multi-channel wireless mesh networks

The capacity of wireless mesh networks can be significantly improved by multi-channel communications; channel assignment is an important issue in multi-channel Mesh network. We propose a novel distributed channel assignment algorithm CBLA(Cluster-Based Load-Aware), which takes advantage of both static and dynamic channel assignment approaches. CBLA uses clustering to reduce the complexity of channel assignment problem, estimates the traffic load by low-overhead statistic information of local packets. Adaptive dynamic channel assignment results in a better link load equilibrium. We also introduce a new routing metric, which combines hop count, channel diversity and cluster information together. A detailed performance evaluation shows that CBLA significantly improves aggregated throughput and reduces average packet delay.EI03546-5513