Centralized Scheduling for Wireless Mesh Networks with Contention-Reduced Media Access

We propose a centralized scheduling approach for wireless mesh networks targeting machine-to-machine applications. Generally centralized scheduling offers good performance comparing with simple distributed schemes such as CSMA/CA which easily arises contention problems due to interference or "back off" wait. This contention problems are prominent when nodes are meshed, and densely deployed. At the same time, however, centralized scheduling requires complex function, or large overhead in both of central and branch nodes. The proposed scheme achieves high efficiency of transmissions and simplicity combining following features. First, our scheme applies sparse coding technology to random access to reduce contention, regarding nodes' transmission requests as a sparse signal. Second, we developed interference-aware node grouping in order to keep both of transmission efficiency and the sparsity of nodes' transmission requests. Finally, our scheme coordinates scheduling and routing by using node group information. Simulation results show that our scheme performs better especially in dense mesh networks.