Recent studies have shown that machine learning (ML), particularly reinforcement learning (RL) in conjunction with radio resource management (RRM), has great promise as a methodology for self-reconfiguration and autonomous network performance optimization. In parallel with cognitive radio platforms, radio access networks, 5G network slicing development, and innovations beyond 5G, these technological advancements can be readily applied to space-based networks for a variety of use cases. However, the community has yet to demonstrate desired network optimization for the highly dynamic user equipment quality of service (QoS) requirements of space-based networks and provide seamless inclusion of delay-tolerant networks. No apparent solutions exist that successfully address the space-based operations' challenging and dynamic operating requirements by exploiting offline RL for optimized RRM. The work described in this paper focuses on applying offline RL algorithms to network-slicing RRM in space-based networks by leveraging 5G cellular network slicing methods. Numerical simulations establish the feasibility of offline RL algorithms for Radio Access Network (RAN) slicing in space missions. This paper presents findings pointing to the efficacy of offline RL algorithms and promising technologies in enabling intelligent communication/network optimization applicable to space-based scenarios.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Offline Reinforcement Learning and Cognitive Radio Resource Management for Space-based Radio Access Network Optimization


    Contributors:


    Publication date :

    2023-06-20


    Size :

    1302045 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Slice Management in Radio Access Network via Deep Reinforcement Learning

    Khodapanah, Behnam / Awada, Ahmad / Viering, Ingo et al. | IEEE | 2020


    Multilayered reinforcement learning approach for radio resource management

    Kevin, Collados / Gorricho Moreno, Juan Luis / Serrat Fernández, Juan et al. | BASE | 2013

    Free access

    Cognitive Radio Network Throughput Maximization with Deep Reinforcement Learning

    Ong, Kevin Shen Hoong / Zhang, Yang / Niyato, Dusit | IEEE | 2019


    Multi-layered reinforcement learning approach for radio resource management

    Kevin, Collados / Gorricho Moreno, Juan Luis / Serrat Fernández, Juan et al. | BASE | 2014

    Free access

    Noncooperative Resource optimization for NOMA Based Fog Radio Access Network

    Qiu, Yu / Zhang, Haijun / Long, Keping et al. | IEEE | 2020