Deep Learning-Aided Channel Allocation Scheme for WLAN

• Published in: IEEE wireless communications letters Vol. 12; no. 6; pp. 1007 - 1011
• Main Authors: Lee, Woongsup, Seo, Jun-Bae
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.06.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Artificial neural networks; Channel allocation; Channels; co-channel interference; Cochannel interference; Computing time; Data transmission; Deep learning; Deep neural network; Local area networks; Machine learning; Neural networks; Optimization; Performance enhancement; Performance evaluation; Training; Unsupervised learning; Wireless communication; Wireless LAN; Wireless networks
• ISSN: 2162-2337; 2162-2345
• DOI: 10.1109/LWC.2023.3257128

In the wireless local area networks (WLANs) based on the IEEE 802.11 technology, the limited set of channels is shared by a large number of access points (APs), which inevitably results in severe co-channel interference (CCI) among APs utilizing the same set of channels. In order to improve the performance of data transmissions in WLANs, the channel allocation must be carried out with care by considering such CCI among APs. In this letter, we propose a deep learning (DL) based channel allocation scheme to minimize the overall CCI experienced by the APs, thereby improving the network's performance. To this end, a deep neural network (DNN) structure and an unsupervised learning-based training methodology are designed. The performance evaluation demonstrates that the proposed DL-based scheme achieves near-optimal performance with low computational time complexity.