Toward an Access Infrastructure for Mobile Cloud

The provision of mobile cloud service calls for a wireless access infrastructure that offers high bandwidth to mobile users. Among all enabling technologies, wireless mesh networks (WMNs) have the advantage of low deployment cost and widely available user equipments. To provide more bandwidth, access points in WMNs are commonly equipped with multiple wireless interfaces (radios) that can operate on multiple non-overlapping channels in parallel. The objective of channel assignments in a multi-channel, multi-radio MWN is to reduce co-channel interference experienced by links so as to increase network capacity while maintaining network connectivity. Prior studies addressing this issue majorly considered effects of co-channel interference at the link layer. In this study, the authors consider co-channel interference at the physical layer. Furthermore, most existing methods are based on heuristic or game theory. This study applies simulated annealing technique to the channel allocation problem. The objective function for this approach is defined as the total signal-to-interference radio (SIR) experienced by each link. To maintain network connectivity, the proposed approach limits the set of assigned channels for each radio. Experimental results show that, compared with traditional heuristic-based and game-theoretic approaches, the proposed simulated annealing algorithm results in more operative links.