Temporal-Spatial Feature Extraction in IoT-Based SCADA System Security: Hybrid CNN-LSTM and Attention-Based Architectures for Malware Classification and Attack Detection

This research presents a Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) model developed for malware classification from IoT devices in the SCADA system and for detecting anomalies in the network. The developed model identifies complex attacks in the network by taking advantage...

Full description

Saved in:
Bibliographic Details
Main Authors: Onur Polat, Ali Ayid Ahmad, Saadin Oyucu, Enes Algul, Ferdi Dogan, Ahmet Aksoz
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11028034/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850166108895576064
author Onur Polat
Ali Ayid Ahmad
Saadin Oyucu
Enes Algul
Ferdi Dogan
Ahmet Aksoz
author_facet Onur Polat
Ali Ayid Ahmad
Saadin Oyucu
Enes Algul
Ferdi Dogan
Ahmet Aksoz
author_sort Onur Polat
collection DOAJ
description This research presents a Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) model developed for malware classification from IoT devices in the SCADA system and for detecting anomalies in the network. The developed model identifies complex attacks in the network by taking advantage of the strengths of CNNs that reveal spatial features and LSTMs that detect temporal dependency. CICIoT 2023 is used as the dataset. ADAM optimization algorithm with cross-entropy loss is used to eliminate overfitting and training is performed. Within the scope of the study, the proposed model is compared by applying it with six deep learning architectures (hybrid CNN-LSTM, Non-Local Neural Network (NLNN), Residual Attention Network (RAN), Dual Attention Network (DANet), Transformer-CNN and Attentional CNN). The obtained results show that the proposed CNN-LSTM model outperforms other complex architectures and achieves a high test accuracy of 99.23%. It has demonstrated remarkable performance according to precision, recall and F1 evaluation metrics in detecting distributed denial of service (DDoS) and denial of service (DoS) attacks. The proposed model successfully identifies Mirai botnet variants and fragmentation-based attacks. Although other models, Transformer-CNN (98.81%) and DANet (98.07%), provide high performance, they fall behind the superior temporal modeling capabilities of CNN-LSTM. When the obtained findings are examined, they highlight the relative strengths of various deep learning approaches for IoT security applications. The performances of the implemented deep learning models reached accuracies exceeding 96%, demonstrating the importance of IoT-based SCADA systems against evolving cyber threats. The study revealed the superior successes of deep learning-based approaches for IoT security.
format Article
id doaj-art-1c1e31bae2bc44a783dc248feb608582
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-1c1e31bae2bc44a783dc248feb6085822025-08-20T02:21:34ZengIEEEIEEE Access2169-35362025-01-011310210910213210.1109/ACCESS.2025.357776111028034Temporal-Spatial Feature Extraction in IoT-Based SCADA System Security: Hybrid CNN-LSTM and Attention-Based Architectures for Malware Classification and Attack DetectionOnur Polat0https://orcid.org/0000-0001-9313-4910Ali Ayid Ahmad1https://orcid.org/0000-0002-6031-9414Saadin Oyucu2https://orcid.org/0000-0003-3880-3039Enes Algul3https://orcid.org/0000-0003-1281-053XFerdi Dogan4https://orcid.org/0000-0002-9203-697XAhmet Aksoz5https://orcid.org/0000-0002-2563-1218Department of Computer Engineering, Bingöl University, Bingöl, TürkiyeDepartment of Electrical Engineering, College of Engineering, University of Kirkuk, Kirkuk, IraqDepartment of Computer Engineering, Faculty of Technology, Gazi University, Ankara, TürkiyeDepartment of Computer Engineering, Bingöl University, Bingöl, TürkiyeDepartment of Computer Engineering, Adıyaman Üniversitesi, Adiyaman, TürkiyeDepartment of Electrical and Electronics Engineering, Kayseri University, Kayseri, TürkiyeThis research presents a Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) model developed for malware classification from IoT devices in the SCADA system and for detecting anomalies in the network. The developed model identifies complex attacks in the network by taking advantage of the strengths of CNNs that reveal spatial features and LSTMs that detect temporal dependency. CICIoT 2023 is used as the dataset. ADAM optimization algorithm with cross-entropy loss is used to eliminate overfitting and training is performed. Within the scope of the study, the proposed model is compared by applying it with six deep learning architectures (hybrid CNN-LSTM, Non-Local Neural Network (NLNN), Residual Attention Network (RAN), Dual Attention Network (DANet), Transformer-CNN and Attentional CNN). The obtained results show that the proposed CNN-LSTM model outperforms other complex architectures and achieves a high test accuracy of 99.23%. It has demonstrated remarkable performance according to precision, recall and F1 evaluation metrics in detecting distributed denial of service (DDoS) and denial of service (DoS) attacks. The proposed model successfully identifies Mirai botnet variants and fragmentation-based attacks. Although other models, Transformer-CNN (98.81%) and DANet (98.07%), provide high performance, they fall behind the superior temporal modeling capabilities of CNN-LSTM. When the obtained findings are examined, they highlight the relative strengths of various deep learning approaches for IoT security applications. The performances of the implemented deep learning models reached accuracies exceeding 96%, demonstrating the importance of IoT-based SCADA systems against evolving cyber threats. The study revealed the superior successes of deep learning-based approaches for IoT security.https://ieeexplore.ieee.org/document/11028034/SCADAIoT securitymalware traffic analysiscritical infrastructureshybrid deep learningnetwork intrusion detection
spellingShingle Onur Polat
Ali Ayid Ahmad
Saadin Oyucu
Enes Algul
Ferdi Dogan
Ahmet Aksoz
Temporal-Spatial Feature Extraction in IoT-Based SCADA System Security: Hybrid CNN-LSTM and Attention-Based Architectures for Malware Classification and Attack Detection
IEEE Access
SCADA
IoT security
malware traffic analysis
critical infrastructures
hybrid deep learning
network intrusion detection
title Temporal-Spatial Feature Extraction in IoT-Based SCADA System Security: Hybrid CNN-LSTM and Attention-Based Architectures for Malware Classification and Attack Detection
title_full Temporal-Spatial Feature Extraction in IoT-Based SCADA System Security: Hybrid CNN-LSTM and Attention-Based Architectures for Malware Classification and Attack Detection
title_fullStr Temporal-Spatial Feature Extraction in IoT-Based SCADA System Security: Hybrid CNN-LSTM and Attention-Based Architectures for Malware Classification and Attack Detection
title_full_unstemmed Temporal-Spatial Feature Extraction in IoT-Based SCADA System Security: Hybrid CNN-LSTM and Attention-Based Architectures for Malware Classification and Attack Detection
title_short Temporal-Spatial Feature Extraction in IoT-Based SCADA System Security: Hybrid CNN-LSTM and Attention-Based Architectures for Malware Classification and Attack Detection
title_sort temporal spatial feature extraction in iot based scada system security hybrid cnn lstm and attention based architectures for malware classification and attack detection
topic SCADA
IoT security
malware traffic analysis
critical infrastructures
hybrid deep learning
network intrusion detection
url https://ieeexplore.ieee.org/document/11028034/
work_keys_str_mv AT onurpolat temporalspatialfeatureextractioniniotbasedscadasystemsecurityhybridcnnlstmandattentionbasedarchitecturesformalwareclassificationandattackdetection
AT aliayidahmad temporalspatialfeatureextractioniniotbasedscadasystemsecurityhybridcnnlstmandattentionbasedarchitecturesformalwareclassificationandattackdetection
AT saadinoyucu temporalspatialfeatureextractioniniotbasedscadasystemsecurityhybridcnnlstmandattentionbasedarchitecturesformalwareclassificationandattackdetection
AT enesalgul temporalspatialfeatureextractioniniotbasedscadasystemsecurityhybridcnnlstmandattentionbasedarchitecturesformalwareclassificationandattackdetection
AT ferdidogan temporalspatialfeatureextractioniniotbasedscadasystemsecurityhybridcnnlstmandattentionbasedarchitecturesformalwareclassificationandattackdetection
AT ahmetaksoz temporalspatialfeatureextractioniniotbasedscadasystemsecurityhybridcnnlstmandattentionbasedarchitecturesformalwareclassificationandattackdetection