Cognitive anti-jamming channel selection based on Bandit learning in aeronautical swarm network

被引：0
|
作者：
Qiu Q. [1 ]
Li H. [2 ]
Zhang H. [2 ]
Luo J. [2 ]
机构：
[1] China National Aeronautical Radio Electronics Institute, Shanghai
[2] Faculty of Information Technology, Beijing University of Technology, Beijing
来源：
Li, Haitao (lihaitao@bjut.edu.cn) | 1600年 / Huazhong University of Science and Technology卷 / 49期
关键词：
Aeronautical swarm network; Channel selection; Cognitive anti-jamming; Kl-UCB[!sup]++[!/sup] algorithm; Multi-armed bandit model;
D O I：
10.13245/j.hust.210504
中图分类号：
学科分类号：
摘要：
To solve the problem that multiple airborne radios access to the same channel would result in collision and degrade the system performance in aeronautical swarm network (ASNET), the cognitive anti-jamming channel selection scheme based on multi-armed bandit (MAB) learning was investigated. The ASNET anti-jamming channel selection MAB game model was constructed firstly, and the accurate algorithm was given to estimate the number of radios in dynamic swarm networks. Then, a collision avoidance (CA) kl-UCB++ anti-jamming channel selection strategy was proposed based on the prior estimated information. Furthermore, the theoretical upper bound of the number of channel collisions was derived. Simulation results show that the proposed CA kl-UCB++ anti-jamming channel selection strategy can reduce the collisions on available channels with lower cumulative regret, which can effectively improve the ASNET anti-jamming ability in frequency domain. © 2021, Editorial Board of Journal of Huazhong University of Science and Technology. All right reserved.
引用
下载
收藏
页码：20 / 25
页数：5
相关论文
共 12 条
  • [1] 8
  • [2] 5
  • [3] AREF M A, JAYAWEERA S K., A novel cognitive antijamming stochastic game, Proc of Cognitive Communications for Aerospace Applications Workshop, pp. 1-4, (2017)
  • [4] WANG B, LIU K R., An anti-jamming stochastic game for cognitive radio networks, IEEE Journal on Selected Areas in Communications, 29, 4, pp. 877-889, (2011)
  • [5] SLIMENI F, SCHEERS B, CHTOUROU Z., Jamming mitigation in cognitive radio networks using a modified Q-learning algorithm, Proc of International Conference of Informatics, Multimedia, Cyber, and Information System, pp. 1-7, (2015)
  • [6] DASTANGOO S, FOSSA C, GWON Y., Competing cognitive resilient networks, IEEE Transactions on Cognitive Communications Network, 2, 1, pp. 95-109, (2016)
  • [7] GWON Y, DASTANGOO S, FOSSA C, Et al., Competing mobile network game: embracing antijamming and jamming strategies with reinforcement learning, Proc of International Conference on Communications and Network Security, pp. 28-36, (2013)
  • [8] GWON Y, DASTANGOO S, KUNG H., Optimizing media access strategy for competing cognitive radio networks, Proc of Global Communications Conference, pp. 1215-1220, (2013)
  • [9] ROSENSK I, SHAMIR O, SZLAK L., Multi-player bandits-a musical chairs approach, Proc of International Conference on Machine Learning, pp. 155-163, (2016)
  • [10] BOURSIER E, PERCHET V., SIC-MMAB: synchronization involves communication in multiplayer multi-armed bandits, Proc of International Conference on Neural Processing and Information Systems, pp. 12048-12057, (2019)
12