Distributed channel assignment and scheduling algorithms for multiple channels multiple interfaces wireless mesh networks

• Main Author: Luong, Trung Tuan.
• Other Authors: Lee Bu Sung
• Format: Theses and Dissertations
• Language: English
• Published: 2011
• Subjects: DRNTU::Engineering::Computer science and engineering::Computer systems organization::Computer-communication networks; DRNTU::Engineering::Computer science and engineering::Theory of computation::Analysis of algorithms and problem complexity
• Online Access: http://hdl.handle.net/10356/44603
• Institution: Nanyang Technological University

Wireless mesh networks have been widely deployed in the past few years. This increasing popularity is based on two basic characteristics: ease of deployment and increase in network capacity. Research in wireless mesh networks has focused on single channel networks, in which a common channel is required for simple routing control in multi-hop networks. Single channel wireless mesh networks do not allow multiple communications to take place concurrently due to channel interference, which leads to inefficient utilization of the available spectrum. This thesis focuses on the case where wireless nodes, equipping with multiple wireless interfaces, endeavor to effectively utilize the multiple available channels to improve the network capacity to meet user requirements. Channel interference causes significant throughput degradation in single channel wireless mesh networks. One approach to overcome this problem is the use of multiple channels multiple interfaces (MCMI) networks, where each wireless node has a dedicated interface per channel. We call this the N-N MCMI networks. A set of channel scheduling algorithms was proposed: Random, Round Robin, and Round Robin+. Among the proposed algorithms, Round Robin+ algorithm is able to minimize the intraflow channel interference and mitigate the effect of interflow channel interference, thus it has the best performance. With effective channel scheduling algorithms, N-N MCMI networks have shown significant improvement in network capacity. Wireless technologies can provide a large number of available channels, thus it is expensive to have a number of interfaces equal to a number of available channels for each node. Thus, the number of interfaces at each node is expected to be fewer than the number of available channels in the network. We call this the N-M MCMI networks ($M<N$).