Minimum weight multicast scheduling in multi-channel wireless mesh networks for real-time voice applications

In this work, we consider the problem of scheduling multicast real-time voice traffic in multi-channel, TDMA wireless mesh networks. The main challenges to solve this problem lie in (a) ensuring strict delay constraint while scheduling a real-time voice call, (b) providing admission control while maximizing the number of requests accepted and (c) scheduling a multicast voice call in an online manner without affecting ongoing calls in the network. Although, scheduling real-time traffic has received considerable attention in recent past, to our knowledge, these set of challenges have not been addressed in prior work. In addition to above set of challenges, we consider the weighted version of the problem where each node has a weight (e.g., function of residual energy) associated with it, and our goal is to find a minimum weight multicast tree. To achieve this goal, we propose an online, centralized algorithm - Real-Time Multicast Scheduling(RTMS), to schedule real-time multicast traffic in TDMA based wireless mesh networks. RTMS runs in polynomial time with a logarithmic factor of approximation on the weight of the tree. Further, our multicast scheduling can be used in any TDMA based mesh network (e.g., WiFi [5]) to schedule multicast voice or video calls. In comparison to state-of-art work done in MSA [15], RTMS increases call acceptance rate by more than 50%, and results in 33% reduction in the weight of the tree. Moreover, RTMS performs close to optimal with the call acceptance rate 80% compared to optimal. Further, the weight of the tree produced by RTMS is only 25% more compared to the optimal weight.