A Distributed Cluster based Interference-Traffic aware CA for MRMC WMN

• Main Authors: Raju, B., Athota, K., Negi, A.
• Format: Conference Proceeding
• Language: English
• Subjects: Channel Assignment; Clustering; Clustering algorithms; Computer science; Interference; Network servers; Network topology; Spine; Spread spectrum communication; Telecommunication traffic; Throughput; Wireless mesh networks
• DOI: 10.1109/WCSN.2009.5434790

Wireless Mesh Networks (WMNs) have the potential for improving network capacity by employing multiple radios and multiple channels (MRMC). Channel Assignment (CA) is a key issue that plays vital role in defining WMN throughput by efficient utilization of available multiple radios and channels there by minimizing network interference. The two important issues that are needed to be addressed by CA algorithm are Connectivity and Interference. CA problem is proven to be NP-Hard [2] [4]. Further difficulty is added by constraints of varying network conditions such as interference and traffic load. In this paper we propose a Distributed Cluster-based Interference-Traffic aware connectivity preserving CA (DCITCA) algorithm. DCITCA involves two phases of operation. In phase 1, first step prepares r-hop clusters in the given network topology where r defines maximum hop distance between any mesh node in the cluster to the cluster head. Second step performs interface-to-link binding to ensures that every link in the network topology receive a communication channel and channel dependency problem is controlled. In phase 2, the channels are assigned to interfaces by selecting least interfering channels in the neighborhood. Interference is measured based on the traffic load carried in the neighborhood. Proposed CA algorithms can be easily implemented on commodity IEEE 802.11 hardware. Our simulations demonstrate that proposed algorithm DCITCA reduces overall network interference when compared to existing CoMTaC[7] CA algorithm.