A Control-Message Quenching Algorithm in OpenFlow-Based Wireless Mesh Networks with Dynamic Spectrum Access

In recent years, Open Flow is becoming a popular network architecture. It can provide fine-grained monitor to the flows in the network. The separation of control plane and data plane makes it flexible. However, to the best of our knowledge, most OpenFlow protocols are designed for wired networks. And the switch state and controller bottleneck are still major issues in OpenFlow network. In this paper, we first proposed a wireless OpenFlow based Cognitive Radio network. In order to solve the channel selection issue in CRN, we introduce the Hierarchical Dirichlet Process(HDP) model to sense and classify the wireless channel. To reduce the state of switch, we use wildcard rule scheme. In addition, an algorithm is proposed in the paper to further quench the requests to the controller. The simulation results show that our proposed CMQ algorithm with information from neighbors outperforms the conventional CMQ algorithm in throughput and packet delay performance.