Traffic Characterization for Efficient TWT Scheduling in 802.11ax IoT Networks.

To reduce packet collisions and enhance the energy efficiency of stations, Target Wake Time (TWT), which is a feature of the 802.11ax standard (WiFi 6), allows the allocation of communication service periods to stations. While effective TWT allocation requires characterizing the traffic pattern of stations, in this paper, we empirically study and reveal that the existing methods (i.e., channel utilization estimation, packet sniffing, and buffer status report) do not provide adequate accuracy. To remedy this problem, we propose a traffic characterization method that can accurately capture inter-packet and inter-burst intervals on a per-flow basis in the presence of factors such as channel access and packet preparation delay. We empirically evaluate the proposed method and confirm its superior traffic characterization performance against the existing ones. We also present a sample TWT allocation scenario that leverages the proposed method to enhance throughput.