arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2510.02183eess.SYcs.SY
收藏

利用部分无攻击数据检测和识别传感器攻击

Detection and Identification of Sensor Attacks Using Partially Attack-Free Data

KTH Royal Institute of Technology(皇家理工学院);Digital Futures(数字未来)

Takumi Shinohara, Karl H. Johansson, Henrik Sandberg

AI总结:

本文提出了一种利用部分无攻击数据检测和识别传感器攻击的方法,能够在攻击区间足够长的情况下有效检测和识别攻击。

摘要:

在本文中,我们研究了在无模型设定下数据驱动的攻击检测和识别问题。我们考虑了一个实际 motivated 的场景,其中可用的数据集可能被恶意传感器攻击破坏,但包含一个未知的、连续的、部分无攻击区间。控制输入被假设包含一个小的随机水印信号。在这些假设下,我们建立了从部分无攻击数据中检测和识别攻击的充分条件。我们还开发了数据驱动的检测和识别程序,并 characterization 它们的计算复杂性。值得注意的是,所提出的框架不限制被破坏传感器的数量;因此,即使所有传感器输出在攻击区间外被破坏,只要攻击区间足够长,它仍然可以检测和识别攻击。最后,我们通过数值模拟展示了所提出框架的有效性。

英文摘要:

In this paper, we investigate data-driven attack detection and identification in a model-free setting. We consider a practically motivated scenario in which the available dataset may be compromised by malicious sensor attacks, but contains an unknown, contiguous, partially attack-free interval. The control input is assumed to include a small stochastic watermarking signal. Under these assumptions, we establish sufficient conditions for attack detection and identification from partially attack-free data. We also develop data-driven detection and identification procedures and characterize their computational complexity. Notably, the proposed framework does not impose a limit on the number of compromised sensors; thus, it can detect and identify attacks even when all sensor outputs are compromised outside the attack-free interval, provided that the attack-free interval is sufficiently long. Finally, we demonstrate the effectiveness of the proposed framework via numerical simulations.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑