Dynamic Spectrum Access for Femtocell Networks: A Graph Neural Network Based Learning Approach

This paper concerns the dynamic spectrum access problem for femtocell networks, where traffic loads are different among cells. We model the interference relationship of the femtocell networks with conflict graphs, and a graphical game is employed as the channel access coordination mechanism. A graph neural network based architecture is proposed, which directly maps traffic loads to the channel access scheme for each femtocell. With our method, each femtocell first estimates the qualities of all available channels based on the information from its neighbors, and then the channels of the highest quality are accessed. A multiagent reinforcement learning framework is designed to train the proposed architecture to make accurate estimations of channel quality.

[1]  Husheng Li,et al.  Multi-Agent Q-Learning for Competitive Spectrum Access in Cognitive Radio Systems , 2010, 2010 Fifth IEEE Workshop on Networking Technologies for Software Defined Radio Networks (SDR).

[2]  Ananthram Swami,et al.  A Survey of Dynamic Spectrum Access: Signal Processing and Networking Perspectives , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.

[3]  Bin Li,et al.  Learning-Based Spectrum Sharing and Spatial Reuse in mm-Wave Ultradense Networks , 2018, IEEE Transactions on Vehicular Technology.

[4]  Alagan Anpalagan,et al.  Opportunistic Spectrum Access with Spatial Reuse: Graphical Game and Uncoupled Learning Solutions , 2013, IEEE Transactions on Wireless Communications.

[5]  Alagan Anpalagan,et al.  Dynamic Spectrum Access in Time-Varying Environment: Distributed Learning Beyond Expectation Optimization , 2015, IEEE Transactions on Communications.

[6]  Kobi Cohen,et al.  Deep Multi-User Reinforcement Learning for Distributed Dynamic Spectrum Access , 2017, IEEE Transactions on Wireless Communications.

[7]  Xu Chen,et al.  Distributed Spectrum Access with Spatial Reuse , 2012, IEEE Journal on Selected Areas in Communications.

[8]  Shane Legg,et al.  Human-level control through deep reinforcement learning , 2015, Nature.

[9]  K. J. Ray Liu,et al.  Game theory for cognitive radio networks: An overview , 2010, Comput. Networks.

[10]  Jeffrey G. Andrews,et al.  Uplink capacity and interference avoidance for two-tier femtocell networks , 2007, IEEE Transactions on Wireless Communications.

[11]  Jeffrey G. Andrews,et al.  Femtocell networks: a survey , 2008, IEEE Communications Magazine.

[12]  Max Welling,et al.  Semi-Supervised Classification with Graph Convolutional Networks , 2016, ICLR.

[13]  Alagan Anpalagan,et al.  Opportunistic Spectrum Access in Cognitive Radio Networks: Global Optimization Using Local Interaction Games , 2012, IEEE Journal of Selected Topics in Signal Processing.

[14]  Rajarathnam Chandramouli,et al.  Distributed Learning in Secondary Spectrum Sharing Graphical Game , 2011, 2011 IEEE Global Telecommunications Conference - GLOBECOM 2011.

[15]  Prabhu Chandhar,et al.  Area Spectral Efficiency of Co-Channel Deployed OFDMA Femtocell Networks , 2014, IEEE Transactions on Wireless Communications.

[16]  Haibo He,et al.  Q-Learning for Non-Cooperative Channel Access Game of Cognitive Radio Networks , 2018, 2018 International Joint Conference on Neural Networks (IJCNN).

[17]  Alagan Anpalagan,et al.  Load-Aware Dynamic Spectrum Access for Small-Cell Networks: A Graphical Game Approach , 2015, IEEE Transactions on Vehicular Technology.

[18]  Robert W. Heath,et al.  Is the PHY layer dead? , 2011, IEEE Communications Magazine.

[19]  Haibo He,et al.  Deep Learning Based Energy Efficiency Optimization for Distributed Cooperative Spectrum Sensing , 2019, IEEE Wireless Communications.

[20]  Dirk Wübben,et al.  Cloud technologies for flexible 5G radio access networks , 2014, IEEE Communications Magazine.

[21]  Le Song,et al.  2 Common Formulation for Greedy Algorithms on Graphs , 2018 .

[22]  Jeffrey G. Andrews,et al.  Femtocells: Past, Present, and Future , 2012, IEEE Journal on Selected Areas in Communications.

[23]  Husheng Li,et al.  Multi-agent Q-learning of channel selection in multi-user cognitive radio systems: A two by two case , 2009, 2009 IEEE International Conference on Systems, Man and Cybernetics.