Abstract :
In distributed wireless mesh networks, highly reliable channels are typically preferred and thus suffer from heavy contentions; on the other hand, unreliable channels are often discarded, leading to bandwidth waste. Indeed, each node, in contributing to overall network performance, should balance between utilization on reliable channels and unreliable channels. Stability plays an important role, either on each node or the whole network. We propose a distributed multi-phase maximum weighted matching algorithm, making use of both reliable and un-reliable channels. We prove that the algorithm will achieve maximized overall network throughput with relatively high stability in a distributed manner. We also apply channel bundles to effectively improve stability in the network, and the problem proves to be NP-hard. The approximate ratio and complexity of the algorithm are also analyzed in a structural manner. The time complexity is O(Delta3 + (log* n)2), and overhead complexity is O(Delta times (Delta + n) + nalpha + log n), with Delta denoting maximum number of node degree in a n nodes network. Simulation results show that p-stable design effectively improves network stability, especially upon the existence of large number of unreliable channels.
Keywords :
computational complexity; radio networks; telecommunication network reliability; telecommunication network topology; wireless channels; NP-hard problem; channel utilization; distributed multiphase maximum weighted matching algorithm; distributed wireless mesh network; reliable channel usability; time complexity; Algorithm design and analysis; Bandwidth; IP networks; Interference; Mobile communication; Stability; Telecommunication network reliability; Throughput; Usability; Wireless mesh networks;