Security of Machine Learning-Based Anomaly Detection in Cyber Physical Systems

TL;DR

This paper investigates the vulnerability of deep learning-based anomaly detection in cyber-physical systems to adversarial attacks and proposes a retraining mitigation strategy using adversarial samples to improve robustness.

Contribution

It introduces a method to defend CPS anomaly detection models against adversarial attacks by retraining with adversarial samples generated via FGSM.

Findings

Adversarial samples significantly reduce detection accuracy.

Retraining with adversarial samples restores model performance.

Deep learning models can be fortified against attacks using this approach.

Abstract

In this study, we focus on the impact of adversarial attacks on deep learning-based anomaly detection in CPS networks and implement a mitigation approach against the attack by retraining models using adversarial samples. We use the Bot-IoT and Modbus IoT datasets to represent the two CPS networks. We train deep learning models and generate adversarial samples using these datasets. These datasets are captured from IoT and Industrial IoT (IIoT) networks. They both provide samples of normal and attack activities. The deep learning model trained with these datasets showed high accuracy in detecting attacks. An Artificial Neural Network (ANN) is adopted with one input layer, four intermediate layers, and one output layer. The output layer has two nodes representing the binary classification results. To generate adversarial samples for the experiment, we used a function called the `fast_gradient_method' from the Cleverhans library. The experimental result demonstrates the influence of FGSM adversarial samples on the accuracy of the predictions and proves the effectiveness of using the retrained model to defend against adversarial attacks.