Stealthy Measurement-Aided Pole-Dynamics Attacks with Nominal Models

TL;DR

This paper introduces a novel measurement-aided pole-dynamics attack method that maintains stealthiness despite model mismatch, using adaptive control and only measurement data, validated on a networked inverted pendulum system.

Contribution

The paper proposes a new MAPDAs method with adaptive control that ensures stealthiness under model mismatch, requiring only measurements and validated experimentally.

Findings

MAPDAs maintain stealthiness despite model mismatch.

MAPDAs are easier to implement, needing only measurements.

Experimental results confirm effectiveness on a networked inverted pendulum.

Abstract

When traditional pole-dynamics attacks (TPDAs) are implemented with nominal models, model mismatch between exact and nominal models often affects their stealthiness, or even makes the stealthiness lost. To solve this problem, our current paper presents a novel stealthy measurement-aided pole-dynamics attacks (MAPDAs) method with model mismatch. Firstly, the limitations of TPDAs using exact models are revealed, where exact models help ensure the stealthiness of TPDAs but model mismatch severely influences its stealthiness. Secondly, to handle model mismatch, the proposed MAPDAs method is designed by using a model reference adaptive control strategy, which can keep the stealthiness. Moreover, it is easier to implement as only the measurements are needed in comparison with the existing methods requiring both the measurements and control inputs. Thirdly, the performance of the proposed MAPDAs method is explored using convergence of multivariate measurements, and MAPDAs with model mismatch have the same stealthiness and similar destructiveness as TPDAs. Specifically, MAPDAs with adaptive gains will remain stealthy at an acceptable detection threshold till destructiveness occurs. Finally, experimental results from a networked inverted pendulum system confirm the feasibility and effectiveness of the proposed method.