An Ensemble Deep Learning-Based Cyber-Attack Detection in Industrial Control System

Abstract

The integration of communication networks and the Internet of Things (IoT) in Industrial Control Systems (ICSs) increases their vulnerability towards cyber-attacks, causing devastating outcomes. Traditional Intrusion Detection Systems (IDSs), which are mainly developed to support information technology systems, count vastly on predefined models and are trained mostly on specific cyber-attacks. Besides, most IDSs do not consider the imbalanced nature of ICS datasets, thereby suffering from low accuracy and high false-positive when being put to use. In this paper, we propose a deep learning model to construct new balanced representations of the imbalanced datasets. The new representations are fed into an ensemble deep learning attack detection model specifically designed for an ICS environment. The proposed attack detection model leverages Deep Neural Network (DNN) and Decision Tree (DT) classifiers to detect cyber-attacks from the new representations. The performance of the proposed model is evaluated based on 10-fold crossvalidation on two real ICS datasets. The results show that the proposed method outperforms conventional classifiers, including Random Forest (RF), DNN, and AdaBoost, as well as recent existing models in the literature. The proposed approach is a generalized technique, which can be implemented in existing ICS infrastructures with minimum effort. INDEX TERMS Cyber-attacks, critical infrastructure, industrial control system, integrity attack, operation technology, information technology, deep learning, neural network.

CPS Attacks Mitigation Approaches on Power Electronic Systems With Security Challenges for Smart Grid Applications: A Review

“…Performing investigation operations targeting a nation's CI [136], [137] Secrecy violations of critical data…”

“…Initiating a cyberwar from a nation against another nation [54], [136], [137] System's shutdown, damage in components, or environmental pollution…”

Real-Time Intrusion Detection in Wireless Network: A Deep Learning-Based Intelligent Mechanism

“…Other machine learning based methods, e.g., Peng proposed an IDS based on Mini Batch K-means combined with Principal Component Analysis (PCA), and this method can be used over big data environment [11]. Farnaaz built a model for intrusion detection system using RF classifier, the model was efficient with low false alarm rate and high detection rate [12]. Besides, the deep learning based methods, e.g., Khan proposed a novel two-stage deep learning (TSDL) based approach to detect intrusion and the authors investigate the impacts on the performance of the proposed model [13].…”

“…Kasongo proposed a Feed-Forward Deep Neural Network (FFDNN) based wireless intrusion detection method, and the experimental results show this method can achieve high detection accuracy [15]. At the same time, Al-Abassi proposed an ensemble method using Deep Neural Network (DNN) and Decision Tree (DT), and this method shows it can achieve high detection accuracy with low false-positive [16]. References [17,18] used other deep learning algorithms to detect network intrusion manners, which can effectively identify various attacks.…”

A Weighted Ensemble Learning Algorithm Based on Diversity Using a Novel Particle Swarm Optimization Approach

“…Ensemble learning combines multiple individual learners to improve the generalization performance of an individual learner [1,2]. It is an important and popular branch of machine learning and is widely used in attack detection [3][4][5], fraud recognition [6,7], image recognition [8,9], biomedicine [10,11], intelligent manufacturing [12,13], time series analysis [14,15] and other fields. Ensemble learning usually involves multiple weak classifiers, such as decision trees [16], support vector machines [17], neural networks [18] and k-nearest neighbors [19], to form a strong classifier, multiple strong classifiers [20] or even a combination of multiple machine learners to complete the learning task.…”