Deep Learning-Based Intrusion Detection Methods in Cyber-Physical Systems: Challenges and Future Trends

Umer, Muhammad; Sadiq, Saima; Karamti, Hanen; Alhebshi, Reemah M.; Alnowaiser, Khaled; Eshmawi, Ala’ Abdulmajid; Song, Houbing; Ashraf, Imran

doi:10.3390/electronics11203326

Cited by 14 publications

(6 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, processing a large amount of data on edge IoT devices saves data-transfer bandwidth. Secondly, edge computing offers more data security regarding data loss, denial of service, etc., than cloud computing [30]. Finally, edge computing has a low response and prediction time.…”

Section: A Motivationmentioning

confidence: 99%

Edge ML Technique for Smart Traffic Management in Intelligent Transportation Systems

Hazarika,

Choudhury,

Nasralla

et al. 2024

IEEE Access

View full text Add to dashboard Cite

In urban traffic, a Dynamic Traffic Light System (DTLS) is an important aspect of automatic driving. DTLS estimates the time of the light signal from images of dynamically changing road traffic. In conventional traffic light systems, light signals are enabled at predefined or fixed time intervals without having information on the current traffic density on the road. This static behavior of the traffic light system increases unnecessary waiting time on the road, eventually creating traffic jams, environmental pollution, and other health emergencies. The smart traffic light system addresses these issues with self-learning algorithms and dynamically allows traffic to pass by learning current traffic density. In this paper, a visionbased DTLS is proposed using the YOLO (You Only Look Once) object detection algorithm that detects and counts the total number of vehicles on the roads of a traffic signal junction. The traffic signals are tuned based on the computed traffic to minimize the overall delay at that junction. Moreover, the traffic junctions are facilitated to communicate with the adjacent junctions to transmit the cumulative traffic delay observed. This delay is used to prioritize traffic passing through salient blocks like schools, offices, hospitals, etc. The paper aims to minimize the overhead incurred in both computations of traffic (using approximate computing) and in communication networks (using low-power technologies of IEEE 802.15.4 standard, specifically DSME MAC and/or LoRaWAN). The proposed system accomplishes its objective of smart city infrastructure by optimizing the traffic flow. Further, the paper provides a mechanism for green traffic corridors for emergency vehicles.

show abstract

Section: A Motivationmentioning

confidence: 99%

Edge ML Technique for Smart Traffic Management in Intelligent Transportation Systems

Hazarika,

Choudhury,

Nasralla

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…As discussed previously, traditional IDS techniques rely on predefined rules and are limited to known attacks; however, AI-based IDS utilizes machine learning algorithms to detect anomalies and identify previously unknown threats, making it more effective. In [19], the authors discuss security threats, vulnerabilities, and cyber-physical system attacks and propose a deep learning-based IDS that analyzes malicious URLs. [20] investigates the vulnerabilities of deep learning-based IDS in power grids to adversarial attacks, proposing defense mechanisms to enhance robustness.…”

Section: Background and Literature Reviewmentioning

confidence: 99%

Validation of a Machine Learning-Based IDS Design Framework Using ORNL Datasets for Power System With SCADA

Zaman,

Upadhyay,

Lung

2023

IEEE Access

View full text Add to dashboard Cite

Supervisory Control and Data Acquisition (SCADA) systems are widely used for remote monitoring and control of industrial processes, such as oil and gas production, power generation, transmission and distribution, and water treatment. Despite the enhanced accessibility, control, and data availability afforded by recent advances in communication technologies, the utilization of these technologies exposes critical infrastructures such as power systems to potential cyber threats. A Machine Learning (ML)based Intrusion Detection System (IDS) seems promising; however, the development of ML models often requires custom methodologies for data preprocessing and training. This strategic approach is necessary for creating high-performance models that can be robustly evaluated and seamlessly integrated into real-time systems. As a result, we propose an ML-based IDS design framework for a SCADA-based power system incorporating effective modeling aspects, such as dataset preprocessing to ensure accurate representation, data augmentation for achieving a balanced dataset, automated feature selection to reduce dimensionality, and rigorous model training and testing procedures. To substantiate our proposed design framework, we conducted a series of experiments using a publicly available ORNL (Oak Ridge National Laboratory) dataset for . The evaluation process encompasses efficient validation techniques with unseen data. Furthermore, the augmented dataset emerged through the aggregation of readings from four Phasor Measurement Units (PMUs) collected over a specific time span into a unified dataset. Among the assessed classifiers, the Random Forest (RF) model, trained on an augmented and balanced dataset, outperformed others, yielding an F1 score of 94.09% during testing with unseen data.

show abstract

“…Reviews of LDoS attacks exist [43,1,44,3,45], but none relates to resource-constrained ICPS. Similarly, some surveys review security concerns of the cyber-physical systems using ML/DL [46][47][48][49][50][51]; however, they do not cover LDoS or their detection.…”

Section: Literature Reviewmentioning

confidence: 99%

Light-Weight Slow-Rate Attack Detection Framework for Resource-Constrained Industrial Cyber-Physical Systems

Zahid,

Kuo,

Sinha

2023

Preprint

View full text Add to dashboard Cite

Industrial Cyber-Physical Systems (ICPS) are heterogenous computer systems interacting withphysical processes in an industrial environment. The high degree of distribution and interconnectedness poses significant security threats to ICPS, including Slow−Rate Attacks (SRA ). SRA often exploit system vulnerabilities arising from the inherent limitations of resource-constrained ICPS computers like programmable logic controller (PLC). Existing literature identify several challenges in detecting and classifying SRA in resource-constrained ICPS. In this article, we propose an optimised Online Sequential−Extreme Learning Machine (OSELM)-based, novel light-weight active security framework for SRA detection. We reduce the memory and space footprint of OSELM through optimisation techniques for deployment in resource-constrained ICPS. Also, we introduce a simple stratified k-fold cross training method to improve the binary and multi-class SRA- detection capability of the proposed framework, focusing on performance and accuracy of attack detection. Experimental results demonstrate that our proposed framework can effectively and efficiently detect binary and multi-class SRA with an accuracy of 0.975 and 0.96 with average detection times of 0.03 and 0.04 sec, respectively, using less than 3 MB of memory and 10-12% of CPU usage in PLC-based ICPS.

show abstract

Deep Learning-Based Intrusion Detection Methods in Cyber-Physical Systems: Challenges and Future Trends

Cited by 14 publications

References 43 publications

Edge ML Technique for Smart Traffic Management in Intelligent Transportation Systems

Edge ML Technique for Smart Traffic Management in Intelligent Transportation Systems

Validation of a Machine Learning-Based IDS Design Framework Using ORNL Datasets for Power System With SCADA

Light-Weight Slow-Rate Attack Detection Framework for Resource-Constrained Industrial Cyber-Physical Systems

Contact Info

Product

Resources

About