An iterative framework for optimizing multicast throughput in wireless networks

This paper shows that, under certain conditions, a wireless network can bemodeled by a directed configuration graph with possible hyperarc links if the transmission schedule is given. Assume single multicast session. The maximum achievable multicast throughput equals the max-flow min-cut bound of the configuration graph. An optimization framework is proposed to maximize the multicast throughput via iterative updates of the transmission schedule. It is demonstrated that the optimal multicast throughput can be obtained without exploring either a large number of hyperarc links or a large number of cuts, although efficient suboptimal algorithm is needed to avoid searching link combinations and to reduce the complexity further to polynomial in the number of nodes. It is also shown that, when the configuration graph has hyperarc links, the minimum cut can no longer be obtained using the well-known flow augmenting path algorithm. An alternative algorithm is proposed.