Robust resource allocation for multi-hop wireless mesh networks with end-to-end traffic specifications

In this paper, we consider the robust resource allocation problem targeting for providing end-to-end rate guarantee in multi-hop multi-radio multi-channel wireless mesh networks. We incorporate the concepts of interference margin and outage probability to characterize the network robustness under homogeneous and heterogeneous wireless channel conditions. The robust resource allocation problems are formulated as mixed-integer nonlinear programming (MINLP) problem by explicitly taking into account practical radio switching, co-channel contention and multi-path routing constraints. To reduce the complexity, we exploit the special property of the problems and decompose them into the feasibility-checking problem, and propose a binary search algorithm to find the optimal solution of the problems using an iterative procedure. Using traces collected from an indoor wireless testbed, we evaluate the proposed algorithms in terms of interference margin and outage probability. The simulation results show that our algorithms are superior to existing scheme under moderate channel variations and external interference.