Continual Deep Reinforcement Learning to Prevent Catastrophic Forgetting in Jamming Mitigation

被引:1
作者
Davaslioglu, Kemal [1 ]
Kompella, Sastry [1 ]
Erpek, Tugba [1 ]
Sagduyu, Yalin E. [1 ]
机构
[1] Nexcepta, Gaithersburg, MD 20878 USA
来源
MILCOM 2024-2024 IEEE MILITARY COMMUNICATIONS CONFERENCE, MILCOM | 2024年
关键词
Anti-jamming; reinforcement learning; deep learning; catastrophic forgetting; continual learning;
D O I
10.1109/MILCOM61039.2024.10773861
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Deep Reinforcement Learning (DRL) has been highly effective in learning from and adapting to RF environments and thus detecting and mitigating jamming effects to facilitate reliable wireless communications. However, traditional DRL methods are susceptible to catastrophic forgetting (namely forgetting old tasks when learning new ones), especially in dynamic wireless environments where jammer patterns change over time. This paper considers an anti-jamming system and addresses the challenge of catastrophic forgetting in DRL applied to jammer detection and mitigation. First, we demonstrate the impact of catastrophic forgetting in DRL when applied to jammer detection and mitigation tasks, where the network forgets previously learned jammer patterns while adapting to new ones. This catastrophic interference undermines the effectiveness of the system, particularly in scenarios where the environment is non-stationary. We present a method that enables the network to retain knowledge of old jammer patterns while learning to handle new ones. Our approach substantially reduces catastrophic forgetting, allowing the anti-jamming system to learn new tasks without compromising its ability to perform previously learned tasks effectively. Furthermore, we introduce a systematic methodology for sequentially learning tasks in the anti-jamming framework. By leveraging continual DRL techniques based on PackNet, we achieve superior anti-jamming performance compared to standard DRL methods. Our proposed approach not only addresses catastrophic forgetting but also enhances the adaptability and robustness of the system in dynamic jamming environments. We demonstrate the efficacy of our method in preserving knowledge of past jammer patterns, learning new tasks efficiently, and achieving superior anti-jamming performance compared to traditional DRL approaches.
引用
收藏
页码:740 / 745
页数:6
相关论文
共 27 条
[1]   QoS and Jamming-Aware Wireless Networking Using Deep Reinforcement Learning [J].
Abuzainab, Nof ;
Erpek, Tugba ;
Davaslioglu, Kemal ;
Sagduyu, Yalin E. ;
Shi, Yi ;
Mackey, Sharon J. ;
Patel, Mitesh ;
Panettieri, Frank ;
Qureshi, Muhammad A. ;
Isler, Volkan ;
Yener, Aylin .
MILCOM 2019 - 2019 IEEE MILITARY COMMUNICATIONS CONFERENCE (MILCOM), 2019,
[2]  
Baldo N, 2011, ACM S MODEL ANAL SIM, P293
[3]  
Brockman G, 2016, Arxiv, DOI [arXiv:1606.01540, DOI 10.48550/ARXIV.1606.01540]
[4]   Timely and Covert Communications under Deep Learning-Based Eavesdropping and Jamming Effects [J].
Costa, Maice ;
Sagduyu, Yalin E. .
JOURNAL OF COMMUNICATIONS AND NETWORKS, 2023, 25 (05) :621-630
[5]   DeepWiFi: Cognitive WiFi with Deep Learning [J].
Davaslioglu, Kemal ;
Soltani, Sohraab ;
Erpek, Tugba ;
Sagduyu, Yalin E. .
IEEE TRANSACTIONS ON MOBILE COMPUTING, 2021, 20 (02) :429-444
[6]  
Dewey D., 2014, AAAI SPRING S
[7]   Deep Learning for Launching and Mitigating Wireless Jamming Attacks [J].
Erpek, Tugba ;
Sagduyu, Yalin E. ;
Shi, Yi .
IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING, 2019, 5 (01) :2-14
[8]   ns-3 meets OpenAI Gym: The Playground for Machine Learning in Networking Research [J].
Gawlowicz, Piotr ;
Zubow, Anatolij .
MSWIM'19: PROCEEDINGS OF THE 22ND INTERNATIONAL ACM CONFERENCE ON MODELING, ANALYSIS AND SIMULATION OF WIRELESS AND MOBILE SYSTEMS, 2019, :113-120
[9]  
Haarnoja T, 2018, PR MACH LEARN RES, V80
[10]  
Han GA, 2017, INT CONF ACOUST SPEE, P2087, DOI 10.1109/ICASSP.2017.7952524