Autonomous Dynamic Frequency Selection for WLANs Operating in the TV White Space

This paper describes the challenges of frequency channel selection for WLANs operating in the TV White Space (TVWS) and proposes a dynamic frequency selection (DFS) technique as a solution. Based on the statistics gained from past usage experience of channels, of the TVWS channels´ vacancies and channel occupation by the adjacent WLANs, each WLAN evaluates and prioritize the channels independently using a priority function. Then by selecting the channel of the highest priority among vacant channels, each WLAN can quickly find a channel in the low interference regime and whose vacancy is adaptive to its traffic load distribution as well. Performance of the proposed DFS is evaluated through simulation using QualNet, based on TVWS channels with realistic feature and by setting the weight factors of different metrics in the priority function to different values. The evaluation results show that compared to conventional DFS, the proposed DFS avoids adjacent APs selecting the same channel and reduces channel switch rate, and is more adaptive to diverse environments. As a consequence, it maximally reduces total time consumption of channel switch by 40% than random selection.