* Upcoming papers  
Subject Keyword Abstract Author
A Probing Signal-based Replay Attack Detection Method Avoiding Control Performance Degradation

Gyujin Na and Yongsoon Eun*
International Journal of Control, Automation, and Systems, vol. 20, no. 11, pp.3637-3649, 2022

Abstract : This paper proposes a probing signal-based replay attack detection method that avoids control performance degradation. Employing probing signals in actuators to detect replay attacks is a well-known and effective strategy: the replay attack replaces the sensor reading with stored sensor data, and thus, no response to the probing signal is present at the sensor. Applying the probing signal, however, introduces a perturbation to the actual system output, which is either regulated to a reference value or controlled to track a desired trajectory. Therefore, the probing signal enables attack detection but simultaneously yields control performance degradation. Clearly, a trade-off exists upon determining the probing signal: a larger amplitude increases the detection probability, especially in the presence of measurement noise, but degrades the control performance; a smaller amplitude of probing signal affects the control performance less but lowers the attack detectability. To address this problem, a disturbance observer (DOB) approach is proposed in this work, where the effect of the probing signal is compensated at the output and the anomaly is detected by looking at the output of the DOB instead of the system. In this way, probing is still effective for replay attack detection, but the regulation and/or tracking performance of the system is compromised much less. An optimization of DOB parameters is presented to satisfy specifications for both attack detection probability and control performance. Simulation results on vehicle platooning and experiment results using unmanned ground vehicle system are presented that validate the efficacy of the proposed method.

Keyword : Anomaly detector, disturbance observer, probing signal, replay attack, unmanned ground vehicle, vehicle platooning.

Download PDF : Click this link

Business License No.: 220-82-01782