Machine Learning and Intelligent Communications. Second International Conference, MLICOM 2017, Weihai, China, August 5-6, 2017, Proceedings, Part II

Research Article

Distributed Joint Channel-Slot Selection for Multi-UAV Networks: A Game-Theoretic Learning Approach

Download
137 downloads
  • @INPROCEEDINGS{10.1007/978-3-319-73447-7_59,
        author={Jiaxin Chen and Yuhua Xu and Yuli Zhang and Qihui Wu},
        title={Distributed Joint Channel-Slot Selection for Multi-UAV Networks: A Game-Theoretic Learning Approach},
        proceedings={Machine Learning and Intelligent Communications. Second International Conference, MLICOM 2017, Weihai, China, August 5-6, 2017, Proceedings, Part II},
        proceedings_a={MLICOM},
        year={2018},
        month={2},
        keywords={Multi-UAV network Joint channel-slot selection Weighted interference minimization game Potential game Stochastic learning automata},
        doi={10.1007/978-3-319-73447-7_59}
    }
    
  • Jiaxin Chen
    Yuhua Xu
    Yuli Zhang
    Qihui Wu
    Year: 2018
    Distributed Joint Channel-Slot Selection for Multi-UAV Networks: A Game-Theoretic Learning Approach
    MLICOM
    Springer
    DOI: 10.1007/978-3-319-73447-7_59
Jiaxin Chen1,*, Yuhua Xu2,*, Yuli Zhang2,*, Qihui Wu1,*
  • 1: Nanjing University of Aeronautics and Astronautics
  • 2: PLA University of Science and Technology
*Contact email: chenjiaxin0507@sina.com, yuhuaenator@gmail.com, yulipkueecs08@126.com, wuqihui2014@sina.com

Abstract

Unmanned aerial vehicle (UAV) has found promising applications in both military and civilian domains worldwide. In this article, we investigate the problem of distributed opportunistic spectrum access under the consideration of channel-slot selection simultaneously in multi-UAV networks from a game-theoretic perspective, and take into account the distinctive features of the multi-UAV network. We formulate the distributed joint channel-slot selection problem as a weighted interference minimization game. We prove that the formulated game is an exact potential game, and then use the distributed stochastic learning automata based joint channel and time slot selection algorithm to achieve the pure-strategy Nash equilibrium. The algorithm does not need information exchange among UAVs in the network which is more suitable for dynamic and practical enviroment. The simulation results demonstrate the effectiveness of the algorithm.