基于虚假数据检测的信息物理系统安全学习控制方法

苗金钊; 刘金良; 孙乐; 查利娟; 田恩刚

doi:10.11999/JEIT250537

基于虚假数据检测的信息物理系统安全学习控制方法

doi: 10.11999/JEIT250537 cstr: 32379.14.JEIT250537

1.
南京信息工程大学计算机学院南京 210044
2.
南京林业大学理学院南京 210037
3.
上海理工大学光电信息与计算机工程学院上海 200093

基金项目: 国家自然科学基金(62373252, 62273174)

详细信息

作者简介:
苗金钊：男，博士生，研究方向为网络安全、无人系统和智能控制

刘金良：男，教授，研究方向为强化学习、网络化系统优化、网络安全和隐私保护

孙乐：女，教授，研究方向为数据挖掘、深度学习、云计算和服务计算

查利娟：女，教授，研究方向为最优化控制、数据驱动、智能控制和网络化系统

田恩刚：男，教授，研究方向为网络控制系统、网络攻击、非线性随机控制和数据驱动

通讯作者:
刘金良　liujinliang@vip.163.com

中图分类号: TP274
计量
- 文章访问数: 16
- HTML全文浏览量: 9
- PDF下载量: 0
- 被引次数: 0
出版历程
- 修回日期: 2025-09-16
- 网络出版日期: 2025-09-23

A Learning-Based Security Control Method for Cyber-Physical Systems Based on False Data Detection

1.
School of Computer Science, Nanjing University of Information Science and Technology, Nanjing 210044, China
2.
College of Science, Nanjing Forestry University, Nanjing 200037, China
3.
School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China

Funds: The National Natural Science Foundation of China (62373252, 62273174)

摘要

摘要: 随着信息物理系统(CPS)在关键基础设施中的广泛部署，其面临的安全威胁日益严峻，特别是虚假数据注入攻击对系统感知与控制能力构成了实质性挑战。针对这一问题，该文提出了一种融合攻击检测、状态估计与控制策略学习的安全控制框架。该方法通过构建传感器数据的安全评估指标，实现对潜在虚假观测数据的实时检测，并在无攻击先验信息的条件下，动态估计可能存在的攻击信号。在此基础上，进一步提出融合多源传感器观测的状态估计策略，以提高对系统真实状态的重构精度。此外，该文还提出了一种基于动态权重在线更新的自适应学习控制方法，利用梯度下降法逼近最优控制策略，从而增强系统在复杂环境中的稳态性能与抗攻击能力。仿真实验结果验证了该方法在虚假数据注入攻击环境下的有效性与安全性能。
- 信息物理系统 /
- 虚假数据检测 /
- 安全控制 /
- 网络攻击
Abstract: Objective Cyber-Physical Systems (CPS) constitute the backbone of critical infrastructures and industrial applications, but the tight coupling of cyber and physical components renders them highly susceptible to cyberattacks. False data injection attacks are particularly dangerous because they compromise sensor integrity, mislead controllers, and can trigger severe system failures. Existing control strategies often assume reliable sensor data and lack resilience under adversarial conditions. Furthermore, most conventional approaches decouple attack detection from control adaptation, leading to delayed or ineffective responses to dynamic threats. To overcome these limitations, this study develops a unified secure learning control framework that integrates real-time attack detection with adaptive control policy learning. By enabling the dynamic identification and mitigation of false data injection attacks, the proposed method enhances both stability and performance of CPS under uncertain and adversarial environments. Methods To address false data injection attacks in CPS, this study proposes an integrated secure control framework that combines attack detection, state estimation, and adaptive control strategy learning. A sensor grouping-based security assessment index is first developed to detect anomalous sensor data in real time without requiring prior knowledge of attacks. Next, a multi-source sensor fusion estimation method is introduced to reconstruct the system’s true state, thereby improving accuracy and robustness under adversarial disturbances. Finally, an adaptive learning control algorithm is designed, in which dynamic weight updating via gradient descent approximates the optimal control policy online. This unified framework enhances both steady-state performance and resilience of CPS against sophisticated attack scenarios. Its effectiveness and security performance are validated through simulation studies under diverse false data injection attack settings. Results and Discussions Simulation results confirm the effectiveness of the proposed secure adaptive learning control framework under multiple false data injection attacks in CPS. As shown in Fig. 1, system states rapidly converge to steady values and maintain stability despite sensor attacks. Fig. 2 demonstrates that the fused state estimator tracks the true system state with greater accuracy than individual local estimators. In Fig. 3, the compensated observation outputs align closely with the original, uncorrupted measurements, indicating precise attack estimation. Fig. 4 shows that detection indicators for sensor groups 2–5 increase sharply during attack intervals, while unaffected sensors remain near zero, verifying timely and accurate detection. Fig. 5 further confirms that the estimated attack signals closely match the true injected values. Finally, Fig. 6 compares different control strategies, showing that the proposed method achieves faster stabilization and smaller state deviations. Together, these results demonstrate robust control, accurate state estimation, and real-time detection under unknown attack conditions. Conclusions This study addresses secure perception and control in CPS under false data injection attacks by developing an integrated adaptive learning control framework that unifies detection, estimation, and control. A sensor-level anomaly detection mechanism is introduced to identify and localize malicious data, substantially enhancing attack detection capability. The fusion-based state estimation method further improves reconstruction accuracy of true system states, even when observations are compromised. At the control level, an adaptive learning controller with online weight adjustment enables real-time approximation of the optimal control policy without requiring prior knowledge of the attack model. Future research will extend the proposed framework to broader application scenarios and evaluate its resilience under diverse attack environments.
- Cyber-Physical Systems (CPS) /
- False data detection /
- Secure control /
- Network attacks

HTML全文

图 1 虚假数据攻击下的系统状态响应

下载: 全尺寸图片幻灯片

图 2 融合状态估计与真实系统状态的对比

下载: 全尺寸图片幻灯片

图 3 攻击前后观测值以及观测估计值的对比

下载: 全尺寸图片幻灯片

图 4 远程观测数据的攻击检测指标

下载: 全尺寸图片幻灯片

图 5 虚假数据攻击信号及其估计值

下载: 全尺寸图片幻灯片

图 6 不同控制策略下的系统状态响应对比

下载: 全尺寸图片幻灯片

参考文献(20)

[1]	YU Rongrong, ZHAO Xu, LU Si, et al. Intelligent game-theoretic approach for resilient robust control design of cyber-physical systems: Application to intelligent transportation systems[J]. IEEE Transactions on Intelligent Transportation Systems, 2024, 25(11): 16072–16083. doi: 10.1109/TITS.2024.3407721.
[2]	杨挺, 刘亚闯, 刘宇哲, 等. 信息物理系统技术现状分析与趋势综述[J]. 电子与信息学报, 2021, 43(12): 3393–3406. doi: 10.11999/JEIT211135. YANG Ting, LIU Yachuang, LIU Yuzhe, et al. Review on cyber-physical system: Technology analysis and trends[J]. Journal of Electronics & Information Technology, 2021, 43(12): 3393–3406. doi: 10.11999/JEIT211135.
[3]	ZHOU Yuanqiang, VAMVOUDAKIS K G, HADDAD W M, et al. A secure control learning framework for cyber-physical systems under sensor and actuator attacks[J]. IEEE Transactions on Cybernetics, 2021, 51(9): 4648–4660. doi: 10.1109/TCYB.2020.3006871.
[4]	张志鹏, 许倩, 夏承遗. 基于矩阵半张量积的信息物理融合系统状态不透明性分析与控制[J]. 电子与信息学报, 2021, 43(12): 3434–3441. doi: 10.11999/JEIT210492. ZHANG Zhipeng, XU Qian, and XIA Chengyi. Semi-tensor product of matrices-based approach to the opacity analysis of cyber physical systems[J]. Journal of Electronics & Information Technology, 2021, 43(12): 3434–3441. doi: 10.11999/JEIT210492.
[5]	金增旺, 刘茵, 刁靖东, 等. 针对信息物理系统远程状态估计的隐蔽虚假数据注入攻击[J]. 自动化学报, 2025, 51(2): 356–365. doi: 10.16383/j.aas.c240527. JIN Zengwang, LIU Yin, DIAO Jingdong, et al. Stealthy false data injection attacks on remote state estimation of cyber-physical systems[J]. Acta Automatica Sinica, 2025, 51(2): 356–365. doi: 10.16383/j.aas.c240527.
[6]	WANG Zhe, ZHANG Heng, YANG Chaoqun, et al. Improved zero-dynamics attack scheduling with state estimation[J]. IEEE/CAA Journal of Automatica Sinica, 2025, 12(2): 472–474. doi: 10.1109/JAS.2024.124737.
[7]	ZHU Panpan, JIN Shangtai, BU Xuhui, et al. Model-free adaptive control for a class of MIMO nonlinear cyberphysical systems under false data injection attacks[J]. IEEE Transactions on Control of Network Systems, 2023, 10(1): 467–478. doi: 10.1109/TCNS.2022.3203354.
[8]	GAO Yang, MA Jiali, WANG Jiaqi, et al. Event-triggered adaptive fixed-time secure control for nonlinear cyber-physical system with false data-injection attacks[J]. IEEE Transactions on Circuits and Systems II: Express Briefs, 2023, 70(1): 316–320. doi: 10.1109/TCSII.2022.3217823.
[9]	LIU Yifa, CHENG Long, and YE Dan. Stealthy false data injection attacks against the summation detector in cyber-physical systems[J]. IEEE Transactions on Industrial Cyber-Physical Systems, 2024, 2: 391–403. doi: 10.1109/TICPS.2024.3446469.
[10]	GUIBENE K, MESSAI N, AYAIDA M, et al. A pattern mining-based false data injection attack detector for industrial cyber-physical systems[J]. IEEE Transactions on Industrial Informatics, 2024, 20(2): 2969–2978. doi: 10.1109/TII.2023.3297139.
[11]	ZHANG Qiang and HE Dakuo. Adaptive neural control of nonlinear cyber–physical systems against randomly occurring false data injection attacks[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2023, 53(4): 2444–2455. doi: 10.1109/TSMC.2022.3212391.
[12]	LI Tongxiang, CHEN Bo, LIU Shichao, et al. Fast attack detection for cyber–physical systems using dynamic data encryption[J]. IEEE Transactions on Cybernetics, 2024, 54(5): 3251–3264. doi: 10.1109/TCYB.2023.3332079.
[13]	REN Yan, ZHANG Heng, YANG Wen, et al. Transferable adversarial attack against deep reinforcement learning-based smart grid dynamic pricing system[J]. IEEE Transactions on Industrial Informatics, 2024, 20(6): 9015–9025. doi: 10.1109/TII.2024.3379645.
[14]	SOLEIMANI E, SEDIGH A K, and NIKOOFARD A. Data-driven reinforcement learning-based forgetting factor iterative learning control[J]. IEEE Transactions on Automation Science and Engineering, 2025, 22: 12245–12256. doi: 10.1109/TASE.2025.3540699.
[15]	FEI Cheng, SHEN Jun, QIU Hongling, et al. Learning secure control design for cyber-physical systems under false data injection attacks[J]. IEEE Transactions on Industrial Cyber-Physical Systems, 2024, 2: 60–68. doi: 10.1109/TICPS.2024.3373715.
[16]	LI Jinyan, LI Xiaomeng, CHEN Guangdeng, et al. Optimal tracking control for cyber-physical systems under mixed attacks via game-theoretical Q-learning[J]. IEEE Transactions on Automation Science and Engineering, 2025, 22: 11944–11954. doi: 10.1109/TASE.2025.3540401.
[17]	LI Xiaohang, CHADLI M, TIAN Zhaoyang, et al. Resilient-learning control of cyber-physical systems against mixed-type network attacks[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2024, 54(9): 5692–5703. doi: 10.1109/TSMC.2024.3408413.
[18]	SHEN Hao, WANG Yun, WU Jiacheng, et al. Secure control for Markov jump cyber-physical systems subject to malicious attacks: A resilient hybrid learning scheme[J]. IEEE Transactions on Cybernetics, 2024, 54(11): 7068–7079. doi: 10.1109/TCYB.2024.3448407.
[19]	REN Hongru, LONG Yinren, LI Hongyi, et al. Data-driven group formation control of cyber-physical systems via distributed cloud computing[J]. IEEE Transactions on Industrial Cyber-Physical Systems, 2025, 3: 341–350. doi: 10.1109/TICPS.2025.3561726.
[20]	SUO Yuhan, CHAI Runqi, CHAI Senchun, et al. Attack detection and secure state estimation of collectively observable cyber-physical systems under false data injection attacks[J]. IEEE Transactions on Automatic Control, 2024, 69(3): 2067–2074. doi: 10.1109/TAC.2023.3316160.