概率欺骗攻击下的饱和忆阻神经网络稳定性

韩欣佚; 于永斌; 王向向; 蔡竟业; 冯箫; 王靖雅; 钟守铭

doi:10.3969/j.issn.2096-6091.2023.22.002

您当前的位置：

首页 >

文章列表页 >

概率欺骗攻击下的饱和忆阻神经网络稳定性

研究论文 | 更新时间：2024-12-06

- 概率欺骗攻击下的饱和忆阻神经网络稳定性
- Stability of Saturated Memristive Neural Networks Under Probabilistic Deception Attacks
- 新一代信息技术 2023年6卷第22期页码：10-19
- 作者机构：
  
  1.电子科技大学信息与软件工程学院，四川成都 610054
  2.电子科技大学数学科学学院，四川成都 611731
- 作者简介：
  
  [ "韩欣佚（2001—），女，博士研究生在读。研究方向：神经网络动力学。E-mail： xinyihancd@163.com" ]
  [ "于永斌（1975—），男，副教授。研究方向：神经网络动力学、自然语言处理、群体智能。E-mail： ybyu@uestc.edu.cn" ]
  [ "王向向（1992—），男，讲师。研究方向：神经网络动力学。E-mail： xxwang@uestc.edu.cn" ]
  [ "蔡竟业（1963—），男，教授，研究方向：智能计算、信息工程、数字信息处理。E-mail： jycai@uestc.edu.cn" ]
  [ "冯箫（1993—），男，博士研究生在读。研究方向：模糊系统。E-mail： fengxiaocd@gmail.com" ]
  [ "王靖雅（1997—），女，博士研究生在读。研究方向：神经网络动力学。E-mail： jingya.wang@std.uestc.edu.cn" ]
  [ "钟守铭（1955—），男，教授，研究方向：微分方程稳定性理论与应用、神经网络理论、生物数学和鲁棒控制。E-mail： zhongsm@uestc.edu.cn" ]
- 基金信息：
  
  国家自然科学基金(62276055);四川省科技计划项目(23ZDYF0755;24NSFSC5679);新一代人工智能国家科技重大专项(2022ZD0116100)
- DOI：10.3969/j.issn.2096-6091.2023.22.002
  中图分类号： TP13
- 纸质出版日期：2023-11-30，
- 稿件说明：
移动端阅览
韩欣佚, 于永斌, 王向向, 等. 概率欺骗攻击下的饱和忆阻神经网络稳定性[J]. 新一代信息技术, 2023, 6(22): 10-19

HAN Xin-yi, YU Yong-bin, WANG Xiang-xiang, et al. Stability of Saturated Memristive Neural Networks Under Probabilistic Deception Attacks[J]. New Generation of Information Technology, 2023, 6(22): 10-19
韩欣佚, 于永斌, 王向向, 等. 概率欺骗攻击下的饱和忆阻神经网络稳定性[J]. 新一代信息技术, 2023, 6(22): 10-19 DOI： 10.3969/j.issn.2096-6091.2023.22.002.

HAN Xin-yi, YU Yong-bin, WANG Xiang-xiang, et al. Stability of Saturated Memristive Neural Networks Under Probabilistic Deception Attacks[J]. New Generation of Information Technology, 2023, 6(22): 10-19 DOI： 10.3969/j.issn.2096-6091.2023.22.002.

摘要

本文聚焦一类饱和忆阻神经网络模型，利用攻击驱动的事件触发控制策略，研究了其在概率欺骗攻击下的均方稳定问题。首先，针对器件物理限制或设备损耗导致的执行器饱和现象，本文建立了饱和忆阻神经网络模型，解决了执行器与控制器间的信号传输存在较大偏差造成的控制失效问题。其次，网络的强开放性和不可预测性使信号在传输时容易受到攻击，本文结合伯努利过程设计了具有时变时延的概率欺骗攻击模型和攻击驱动的事件触发控制策略，提升了系统应对复杂环境的抗干扰能力，并且减少了数据传输量，节省了通信资源。此外，本文构建了一种依赖于事件触发时刻的新型李雅普诺夫泛函，加强了对系统状态信息的利用，并通过李雅普诺夫稳定性理论和不等式技术，推导出系统均方稳定的充分条件。最后，通过数值仿真验证了结果的可行性和有效性。

Abstract

This paper focuses on a class of saturated memristive neural network models and investigates its mean square stable problem under probabilistic deception attacks with an attack-driven event-triggered control strategy. First

aiming at the actuator saturation phenomenon under device physical limitations or equipment loss

this paper establishes the saturated memristive neural networks

which resolve the issue of control failure caused by significant deviations in signal transmission between actuators and controllers. Next

the strong openness and unpredictability of the network make signals susceptible to attacks during transmission

this paper combines Bernoulli processes to design a probabilistic deception attack model with time-varying delays and an attack-driven event-triggered control strategy

which enhances the anti-interference ability of the system in complex environments

reduces data transmission

and conserves communication resources. Furthermore

this paper constructs a novel Lyapunov functional dependent on event-triggered instants

reinforcing the utilization of system state information. Then

through Lyapunov stability theory and inequality techniques

sufficient conditions for the mean square stability of the system are derived. Ultimately

the feasibility and effectiveness of the results are verified through numerical examples.

关键词

忆阻神经网络执行器饱和概率欺骗攻击事件触发控制

Keywords

memristive neural networksactuator saturationprobabilistic deception attacksevent-triggered control

references

CHUA L O. Memristor-The missing circuit element[J]. IEEE Transactions on Circuit Theory, 1971, 18(5): 507-519.

STRUKOV D B, SNIDER G S, STEWART D R, et al. The missing memristor found[J]. Nature, 2008, 453(7191): 80-83.

XIA Q F, YANG J J. Memristive crossbar arrays for brain-inspired computing[J]. Nature Materials, 2019, 18(4): 309-323.

ZHAO H, LIU Z W, TANG J S, et al. Energy-efficient high-fidelity image reconstruction with memristor arrays for medical diagnosis[J]. Nature Communications, 2023, 14(1): 2276.

CAO Z L, SUN B, ZHOU G D, et al. Memristor-based neural networks: A bridge from device to artificial intelligence[J]. Nanoscale Horizons, 2023, 8(6): 716-745.

ADHIKARI S P, YANG C J, KIM H, et al. Memristor bridge synapse-based neural network and its learning[J]. IEEE Transactions on Neural Networks and Learning Systems, 2012, 23(9): 1426-1435.

ZHANG W, BRANICKY M S, PHILLIPS S M. Stability of networked control systems[J]. IEEE Control Systems Magazine, 2001, 21(1): 84-99.

JIANG B Y, HU Q L, FRISWELL M I. Fixed-time attitude control for rigid spacecraft with actuator saturation and faults [J]. IEEE Transactions on Control Systems Technology, 2016, 24(5): 1892-1898.

WANG J L, JIANG H J, MA T L, et al. <math id="M390"><msub><mrow><mi mathvariant="normal">H</mi></mrow><mrow><mi mathvariant="normal">∞</mi></mrow></msub></math>https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=70173784&type=https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=70173783&type=3.725333212.79399991 control of memristive neural networks with aperiodic sampling and actuator saturation[J]. International Journal of Robust and Nonlinear Control, 2018, 28(8): 3092-3111.

YAN Z L, HUANG X, LIANG J L. Aperiodic sampled-data control for stabilization of memristive neural networks with actuator saturation: A dynamic partitioning method[J]. IEEE Transactions on Cybernetics, 2023, 53(3): 1725-1737.

FAN Y J, HUANG X, WANG Z, et al. Discontinuous event-triggered control for local stabilization of memristive neural networks with actuator saturation: Discrete-and continuous-time Lyapunov methods[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023, 34(4): 1988-2000.

YUAN Y, ZHANG P, GUO L, et al. Towards quantifying the impact of randomly occurred attacks on a class of networked control systems[J]. Journal of the Franklin Institute, 2017, 354(12): 4966-4988.

CHAO Z, WANG C H, YAO W. Quasi-synchronization of stochastic memristive neural networks subject to deception attacks[J]. Nonlinear Dynamics, 2023, 111(3): 2443-2462.

YAN L S, WANG Z, ZHANG M G, et al. Sampled-data control for mean-square exponential stabilization of memristive neural networks under deception attacks[J]. Chaos, Solitons & Fractals, 2023, 174: 113787.

FAN Y J, HUANG X, WANG Z. Memory-dependent event-trigger scheme for secure control of memristive neural networks: Dealing with deception attacks[J]. IEEE Transactions on Industrial Informatics, 2024, 20(4): 5562-5571.

CAO Y T, WANG S B, GUO Z Y, et al. Synchronization of memristive neural networks with leakage delay and parameters mismatch via event-triggered control[J]. Neural Networks, 2019, 119: 178-189.

张仁磊, 童东兵, 陈巧玉, 等.事件触发下带有Lévy噪声的神经网络指数同步[J/OL]. 控制工程, 2024: 1-6. https://doi.org/10.14107/j.cnki.kzgc.20220359https://doi.org/10.14107/j.cnki.kzgc.20220359.

刘飞扬, 李兵. 混合时滞复值神经网络的事件触发状态估计[J].应用数学和力学, 2022, 43(8): 911-919.

王娟, 朱帅东, 朱全民. 基于CNF的饱和切换系统的事件触发控制[J]. 控制与决策, 2024, 39(8): 2631-2636.

GAO H J, MENG X Y, CHEN T W. Stabilization of networked control systems with a new delay characterization[J]. IEEE Transactions on Automatic Control, 2008, 53(9): 2142-2148.

LIU J L, WEI L L, XIE X P, et al. Quantized stabilization for T-S fuzzy systems with hybrid-triggered mechanism and stochastic cyber-attacks[J]. IEEE Transactions on Fuzzy Systems, 2018, 26(6): 3820-3834.

WANG X X, YU Y B, SHI K B, et al. Membership-mismatched impulsive exponential stabilization for fuzzy unconstrained multilayer neural networks with node-dependent delays[J]. IEEE Transactions on Fuzzy Systems, 2023, 31(4): 1214-1228.

LI X, DE SOUZA C E. Delay-dependent robust stability and stabilization of uncertain linear delay systems: a linear matrix inequality approach[J]. IEEE Transactions on Automatic Control, 1997, 42(8): 1144-1148.

CAO Y Y, SUN Y X, CHENG C W. Delay-dependent robust stabilization of uncertain systems with multiple state delays[J]. IEEE Transactions on Automatic Control, 1998, 43(11): 1608-1612.

DING S B, WANG Z S, ZHANG H G. Dissipativity analysis for stochastic memristive neural networks with time-varying delays: A discrete-time case[J]. IEEE Transactions on Neural Networks and Learning Systems, 2018, 29(3): 618-630.

DONG H L, WANG Z D, HO D W C, et al. Robust <math id="M391"><msub><mrow><mi mathvariant="normal">H</mi></mrow><mrow><mi mathvariant="normal">∞</mi></mrow></msub></math>https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=70173784&type=https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=70173783&type=3.725333212.79399991 fuzzy output-feedback control with multiple probabilistic delays and multiple missing measurements[J]. IEEE Transactions on Fuzzy Systems, 2010, 18(4): 712-725.

DA SILVA J M G, TARBOURIECH S. Antiwindup design with guaranteed regions of stability: An LMI-based approach[J]. IEEE Transactions on Automatic Control, 2005, 50(1): 106-111.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据