Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Feb 22, 2026

A Real-Time Interactive System for Studying Confrontational Pursuit Behavior in Rodents
06:25

A Real-Time Interactive System for Studying Confrontational Pursuit Behavior in Rodents

Published on: May 16, 2025

1.4K

Implementation of an RL-based cyberattack detector using VaR thresholding approach.

Ali Khoshlahjeh Sedgh1, Omid Payam1, Hamid Reza Chavoshi1

  • 1The Department of Systems and Control, Faculty of Electrical Engineering, K. N. Toosi University of Technology, Tehran 1631714191, Iran.

ISA Transactions
|February 20, 2026
PubMed
Summary

Related Concept Videos

Difference from Background: Limit of Detection01:05

Difference from Background: Limit of Detection

8.6K
The limit of detection (LOD) is the smallest amount of analyte that can be distinguished from the background noise. The LOD value corresponds to the concentration at which the analyte signal is three times larger than the standard deviation of the blank signal. Below this value, the analyte signal cannot be differentiated from the background noise. It is calculated by dividing the calibration slope by 3 times the standard deviation of the blank signals.
The LOD indicates the presence or absence...
8.6K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Modeling and Analysis of COVID-19 Spread: The Impacts of Nonpharmaceutical Protocols.

Computational and mathematical methods in medicine·2022
Same author

Tail conditional moment for generalized skew-elliptical distributions.

Journal of applied statistics·2022
Same author

Adaptive state-dependent impulsive observer design for nonlinear deterministic and stochastic dynamics with time-delays.

ISA transactions·2019
Same author

On the design of human immunodeficiency virus treatment based on a non-linear time-delay model.

IET systems biology·2014
Same author

Optimal drug regimens in cancer chemotherapy: a multi-objective approach.

Computers in biology and medicine·2013
Same author

Estimation of region of attraction for polynomial nonlinear systems: a numerical method.

ISA transactions·2013

This study introduces a novel, model-free cyberattack detection method for liquid-level control systems using Reinforcement Learning. It effectively identifies sophisticated threats like FDI, DoS, and MitM attacks in real-time.

Area of Science:

  • Cyber-Physical Systems (CPS) security
  • Networked Control Systems (NCS) vulnerability analysis
  • Machine Learning applications in industrial control

Background:

  • Increasing connectivity in CPS and NCS raises significant cybersecurity concerns.
  • Existing detection methods often require system models, limiting their applicability.
  • Sophisticated cyber threats necessitate advanced, real-time detection strategies.

Purpose of the Study:

  • To develop and validate a model-free, real-time cyberattack detection method for a Liquid-Level Control CPS (LLC-CPS).
  • To leverage Reinforcement Learning (RL) and Value-at-Risk (VaR) for robust anomaly detection.
  • To address challenges posed by system dynamics and non-Gaussian noise in attack detection.

Main Methods:

  • Utilized Q-learning to estimate a Bellman deviation sequence for anomaly detection, eliminating the need for an explicit system model.
Keywords:
Cyber-physical systemsCyberattack detectionLiquid-level controlModel-free detectionReinforcement learningVaR

Related Experiment Videos

Last Updated: Feb 22, 2026

A Real-Time Interactive System for Studying Confrontational Pursuit Behavior in Rodents
06:25

A Real-Time Interactive System for Studying Confrontational Pursuit Behavior in Rodents

Published on: May 16, 2025

1.4K
  • Implemented a Value-at-Risk (VaR) thresholding approach to handle heavy-tailed residuals and non-Gaussian distributions.
  • Experimentally validated the method against False Data Injection (FDI), Denial-of-Service (DoS), and Man-in-the-Middle (MitM) attacks.

    • Main Results:

    • The proposed method demonstrated reliable real-time detection of various cyber threats on an LLC-CPS.
    • Achieved effective separation of normal operational states from attack conditions, even with system complexities.
    • Performance was evaluated using False Alarm Rate (FAR), Missed Alarm Rate (MAR), Detection Delay (DD), precision, and F1-score, showing promising results.

    Conclusions:

    • The model-free RL-based approach with VaR thresholding offers a viable solution for securing CPS against advanced cyber threats.
    • The method's effectiveness is confirmed through practical experiments, highlighting its robustness and adaptability.
    • This research contributes to enhancing the cybersecurity posture of industrial control systems.