5th International ICST Conference on Cognitive Radio Oriented Wireless Networks and Communications

Research Article

Analysis and implementation of reinforcement learning on a GNU Radio cognitive radio platform

Download1054 downloads
  • @INPROCEEDINGS{10.4108/ICST.CROWNCOM2010.9170,
        author={Yu Ren and Pawel Dmochowski and Peter Komisarczuk},
        title={Analysis and implementation of reinforcement learning on a GNU Radio cognitive radio platform},
        proceedings={5th International ICST Conference on Cognitive Radio Oriented Wireless Networks and Communications},
        publisher={IEEE},
        proceedings_a={CROWNCOM},
        year={2010},
        month={9},
        keywords={Chromium Cognitive radio Convergence Markov processes Media Access Protocol Receivers Sensors},
        doi={10.4108/ICST.CROWNCOM2010.9170}
    }
    
  • Yu Ren
    Pawel Dmochowski
    Peter Komisarczuk
    Year: 2010
    Analysis and implementation of reinforcement learning on a GNU Radio cognitive radio platform
    CROWNCOM
    IEEE
    DOI: 10.4108/ICST.CROWNCOM2010.9170
Yu Ren1,*, Pawel Dmochowski1,*, Peter Komisarczuk1,*
  • 1: School of Engineering and Computer Science, Victoria University of Wellington, PO Box 600, Wellington, New Zealand
*Contact email: yu.ren@ecs.vuw.ac.nz, pawel.dmochowski@ecs.vuw.ac.nz, peter.komisarczuk@ecs.vuw.ac.nz

Abstract

We present a physical cognitive radio system implementation under the GNU Radio platform with the aim of evaluating a reinforcement learning spectrum management scheme. In our experiments we examine the packet transmission success rate of the cognitive user for a variety of channel utilisation parameters. We derive analytical expressions using Markov chain analysis for the learning convergence time and secondary user packet transmission success rate in the general case of large-scale networks. Our results show that the reinforcement learning scheme significantly improves system performance.