A Dual-Tier Policy-Oriented Anti-Jamming Scheme Based on Deep Reinforcement Learning

Xingyun Chen; Haichuan Ding; Ying Ma; Xuanheng Li; Jianping An; Yuguang Fang

doi:10.1109/TWC.2026.3653807

A Dual-Tier Policy-Oriented Anti-Jamming Scheme Based on Deep Reinforcement Learning

Xingyun Chen, Haichuan Ding, Ying Ma, Xuanheng Li, Jianping An, Yuguang Fang

Department of Computer Science

Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review

3 Downloads (CityUHK Scholars)

Abstract

With the proliferation of software-defined radio technology, malicious jamming attacks against wireless communications have become more aggressive and flexible, which could easily create a complex and highly dynamic jamming environment by varying both the jamming parameters and the jamming policies. Such a complex jamming environment makes it challenging for most of deep reinforcement learning (DRL) based anti-jamming schemes in rapidly identifying effective strategies. In this paper, we have developed a dual-tier policy-oriented anti-jamming (DPA) scheme based on DRL to facilitate swift adaptation to the complex jamming environment. Unlike existing works, an upper-tier jamming pattern recognition (JPR) network is introduced to extract underlying jamming policy-related information which serves as a guidance for the lower-tier deep recurrent Q-network on anti-jamming decision-making. The output of the JPR network can enable the sharing of experiences among various jamming patterns originated from the same jamming policy and facilitate more efficient and targeted anti-jamming strategic learning. Extensive experimental results demonstrate that the superiority of our DPA scheme over other DRL-based benchmark schemes in terms of both anti-jamming performance and convergence speed. © 2026 IEEE

Original language	English
Pages (from-to)	10652-10668
Journal	IEEE Transactions on Wireless Communications
Volume	25
Online published	23 Jan 2026
DOIs	https://doi.org/10.1109/TWC.2026.3653807
Publication status	Published - 2026

Funding

The work of Haichuan Ding was supported by the National Natural Science Foundation of China under Grant 92367201 and Grant 62201045. The work of Ying Ma was supported by the National Natural Science Foundation of China under Grant 62301052. The work of Xuanheng Li was supported in part by the National Natural Science Foundation of China under Grant 62271100; in part by the Science and Technology Program of Liaoning Province under Grant 2023JH2/101700366; in part by the Fundamental Research Funds for the Central Universities under Grant DUT24ZD127; in part by the Open Research Fund of the National Mobile Communications Research Laboratory, Southeast University, under Grant 2025D02; and in part by the Xiaomi Young Talents Program. The work of Yuguang Fang was supported in part by Hong Kong Jockey Club Charities Trust under the JC STEM Lab of Smart City under Grant 2023-0108 and in part by Hong Kong SAR Government under the Global STEM Professorship.

Research Keywords

anti-jamming communication
cognitive radio network
deep recurrent Q-network
Deep reinforcement learning
dynamic spectrum access

Publisher's Copyright Statement

COPYRIGHT TERMS OF DEPOSITED POSTPRINT FILE: © 2026 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. Chen, X., Ding, H., Ma, Y., Li, X., An, J., & Fang, Y. (2026). A Dual-Tier Policy-Oriented Anti-Jamming Scheme Based on Deep Reinforcement Learning. IEEE Transactions on Wireless Communications, 25, 10652- 10668. https://doi.org/10.1109/TWC.2026.3653807

Access Output

10.1109/TWC.2026.3653807

428662249.pdfPost-print, 1 MB

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@article{e4e1137b094d4cff86e3adfa127ad0ed,

title = "A Dual-Tier Policy-Oriented Anti-Jamming Scheme Based on Deep Reinforcement Learning",

abstract = "With the proliferation of software-defined radio technology, malicious jamming attacks against wireless communications have become more aggressive and flexible, which could easily create a complex and highly dynamic jamming environment by varying both the jamming parameters and the jamming policies. Such a complex jamming environment makes it challenging for most of deep reinforcement learning (DRL) based anti-jamming schemes in rapidly identifying effective strategies. In this paper, we have developed a dual-tier policy-oriented anti-jamming (DPA) scheme based on DRL to facilitate swift adaptation to the complex jamming environment. Unlike existing works, an upper-tier jamming pattern recognition (JPR) network is introduced to extract underlying jamming policy-related information which serves as a guidance for the lower-tier deep recurrent Q-network on anti-jamming decision-making. The output of the JPR network can enable the sharing of experiences among various jamming patterns originated from the same jamming policy and facilitate more efficient and targeted anti-jamming strategic learning. Extensive experimental results demonstrate that the superiority of our DPA scheme over other DRL-based benchmark schemes in terms of both anti-jamming performance and convergence speed. {\textcopyright} 2026 IEEE",

keywords = "anti-jamming communication, cognitive radio network, deep recurrent Q-network, Deep reinforcement learning, dynamic spectrum access",

author = "Xingyun Chen and Haichuan Ding and Ying Ma and Xuanheng Li and Jianping An and Yuguang Fang",

year = "2026",

doi = "10.1109/TWC.2026.3653807",

language = "English",

volume = "25",

pages = "10652--10668",

journal = "IEEE Transactions on Wireless Communications",

issn = "1536-1276",

publisher = "IEEE",

}

TY - JOUR

T1 - A Dual-Tier Policy-Oriented Anti-Jamming Scheme Based on Deep Reinforcement Learning

AU - Chen, Xingyun

AU - Ding, Haichuan

AU - Ma, Ying

AU - Li, Xuanheng

AU - An, Jianping

AU - Fang, Yuguang

PY - 2026

Y1 - 2026

N2 - With the proliferation of software-defined radio technology, malicious jamming attacks against wireless communications have become more aggressive and flexible, which could easily create a complex and highly dynamic jamming environment by varying both the jamming parameters and the jamming policies. Such a complex jamming environment makes it challenging for most of deep reinforcement learning (DRL) based anti-jamming schemes in rapidly identifying effective strategies. In this paper, we have developed a dual-tier policy-oriented anti-jamming (DPA) scheme based on DRL to facilitate swift adaptation to the complex jamming environment. Unlike existing works, an upper-tier jamming pattern recognition (JPR) network is introduced to extract underlying jamming policy-related information which serves as a guidance for the lower-tier deep recurrent Q-network on anti-jamming decision-making. The output of the JPR network can enable the sharing of experiences among various jamming patterns originated from the same jamming policy and facilitate more efficient and targeted anti-jamming strategic learning. Extensive experimental results demonstrate that the superiority of our DPA scheme over other DRL-based benchmark schemes in terms of both anti-jamming performance and convergence speed. © 2026 IEEE

AB - With the proliferation of software-defined radio technology, malicious jamming attacks against wireless communications have become more aggressive and flexible, which could easily create a complex and highly dynamic jamming environment by varying both the jamming parameters and the jamming policies. Such a complex jamming environment makes it challenging for most of deep reinforcement learning (DRL) based anti-jamming schemes in rapidly identifying effective strategies. In this paper, we have developed a dual-tier policy-oriented anti-jamming (DPA) scheme based on DRL to facilitate swift adaptation to the complex jamming environment. Unlike existing works, an upper-tier jamming pattern recognition (JPR) network is introduced to extract underlying jamming policy-related information which serves as a guidance for the lower-tier deep recurrent Q-network on anti-jamming decision-making. The output of the JPR network can enable the sharing of experiences among various jamming patterns originated from the same jamming policy and facilitate more efficient and targeted anti-jamming strategic learning. Extensive experimental results demonstrate that the superiority of our DPA scheme over other DRL-based benchmark schemes in terms of both anti-jamming performance and convergence speed. © 2026 IEEE

KW - anti-jamming communication

KW - cognitive radio network

KW - deep recurrent Q-network

KW - Deep reinforcement learning

KW - dynamic spectrum access

UR - http://www.scopus.com/inward/record.url?scp=105028851168&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-105028851168&origin=recordpage

U2 - 10.1109/TWC.2026.3653807

DO - 10.1109/TWC.2026.3653807

M3 - RGC 21 - Publication in refereed journal

SN - 1536-1276

VL - 25

SP - 10652

EP - 10668

JO - IEEE Transactions on Wireless Communications

JF - IEEE Transactions on Wireless Communications

ER -

A Dual-Tier Policy-Oriented Anti-Jamming Scheme Based on Deep Reinforcement Learning

Abstract

Funding

Research Keywords

Publisher's Copyright Statement

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this