Deep Reinforcement Learning-Based Channel Allocation for Wireless LANs\n with Graph Convolutional Networks

2019/05/17 by Kota Nakashima, Nakashima, Kota, Shotaro Kamiya +9 · 2 citations
Computer Science · #Cooperative Communication and Network Coding #Energy Efficient Wireless Sensor Networks #FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Networking and Internet Architecture (cs.NI) #Signal Processing (eess.SP) #Wireless Networks and Protocols #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.1905.07144

openalex publication_date 2019/05/17 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28

Abstract

Last year, IEEE 802.11 Extremely High Throughput Study Group (EHT Study\nGroup) was established to initiate discussions on new IEEE 802.11 features.\nCoordinated control methods of the access points (APs) in the wireless local\narea networks (WLANs) are discussed in EHT Study Group. The present study\nproposes a deep reinforcement learning-based channel allocation scheme using\ngraph convolutional networks (GCNs). As a deep reinforcement learning method,\nwe use a well-known method double deep Q-network. In densely deployed WLANs,\nthe number of the available topologies of APs is extremely high, and thus we\nextract the features of the topological structures based on GCNs. We apply GCNs\nto a contention graph where APs within their carrier sensing ranges are\nconnected to extract the features of carrier sensing relationships.\nAdditionally, to improve the learning speed especially in an early stage of\nlearning, we employ a game theory-based method to collect the training data\nindependently of the neural network model. The simulation results indicate that\nthe proposed method can appropriately control the channels when compared to\nextant methods.\n

Cited by

• Graph Neural Networks for Resource Allocation in Interference-limited Multi-Channel Wireless Networks with QoS Constraints
• Hierarchical Multi-Agent Reinforcement Learning-based Coordinated Spatial Reuse for Next Generation WLANs

Related

• Graph Neural Networks for Wireless Networks: Graph Representation, Architecture and Evaluation
• Decentralized Channel Management in WLANs with Graph Neural Networks
• Carrier-Sense Multiple Access for Heterogeneous Wireless Networks Using Deep Reinforcement Learning
• GCN-Driven Reinforcement Learning for Probabilistic Real-Time Guarantees in Industrial URLLC
• Application of End-to-End Deep Learning in Wireless Communications Systems