Bottleneck-first scheduling for real-time traffic in IEEE 802.11 infrastructure-based mesh networks

This paper studies the real-time traffic scheduling in IEEE 802.11 infrastructure-based wireless mesh networks. Providing strict latency guarantee for real-time traffic in a wireless mesh network is difficult, and one of the main challenges is the difficulty in coordinating temporal operations of the mesh access points (APs). In this paper we propose a bottleneck-first scheduling scheme (BFS) for voice traffic. In the proposed scheme, a central station is responsible for making scheduling decisions for all the real-time packet transmissions at the APs. Scheduling decisions at the APs with a higher traffic load are done before those with a lower traffic load. At each AP, voice packets with more end-to-end hops are scheduled first. Numerical results show that the proposed scheduling scheme achieves low transmission delay and high capacity in the mesh networks, compared to the simple first-come-first-serve scheduling scheme.