Article
Chemistry, Multidisciplinary
Marta Catillo, Antonio Pecchia, Umberto Villano
Summary: This paper proposes a novel IoT-driven cross-device method, which allows learning a single IDS model instead of many separate models atop the traffic of different IoT devices. A semi-supervised approach is adopted due to its wider applicability for unanticipated attacks. The results obtained demonstrate the validity of the proposal, which represents a lightweight and device-independent solution with considerable advantages in terms of transferability and adaptability.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Artificial Intelligence
Thanh Thi Nguyen, Vijay Janapa Reddi
Summary: This article presents a survey of DRL approaches developed for cyber security, including vital aspects such as DRL-based security methods for cyber-physical systems and autonomous intrusion detection techniques. It also discusses multiagent DRL-based game theory simulations for defense strategies against cyberattacks. Future research directions and extensive discussions on DRL-based cyber security are provided.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Review
Computer Science, Information Systems
Asma Alotaibi, Ahmed Barnawi
Summary: The Internet of things (IoT) is a rapidly developing technology that enables smart services in various domains. Securing 6G/massive IoT networks against threats, especially novel attacks, is a major challenge. Innovative architectures and paradigms empowered by artificial intelligence and key networking enablers are urgently needed. Researchers are using machine learning and deep learning techniques to improve cyber threat detection. Designing an intrusion detection system (IDS) for massive IoT applications is a challenge that requires consideration of multiple factors. This survey provides a comprehensive study on massive IoT security aspects, particularly IDS systems, in the context of 6G networks.
INTERNET OF THINGS
(2023)
Article
Automation & Control Systems
Jun Zhang, Lei Pan, Qing-Long Han, Chao Chen, Sheng Wen, Yang Xiang
Summary: With the increasing cyber attacks against cyber-physical systems, the use of deep learning methods in detecting these attacks is explored in this survey. A six-step methodology is provided for summarizing and analyzing the literature on applying deep learning methods for cyber attack detection. The survey reveals the great potential of deep learning modules in detecting cyber attacks against CPS systems, with excellent performance achieved partly due to the availability of high-quality datasets. The survey also identifies challenges, opportunities, and future research trends in this area.
IEEE-CAA JOURNAL OF AUTOMATICA SINICA
(2022)
Article
Computer Science, Information Systems
Weiping Ding, Mohamed Abdel-Basset, Reda Mohamed
Summary: Our daily lives have been greatly influenced by the Internet of Things (IoT) in recent years. While IoT brings convenience and efficiency to our lives, it also exposes devices to cyberattacks due to weak security mechanisms. This paper introduces DeepAK-IoT, a deep learning model designed to detect cyberattacks against IoT devices. It utilizes three blocks – RSR, TRB, and DB – to extract features, learn temporal representations, and classify input records. Experimental results on three public datasets demonstrate DeepAK-IoT's high accuracy in detecting cyber threats in IoT systems, making it a powerful alternative model for managing cybersecurity in IoT networks.
INFORMATION SCIENCES
(2023)
Article
Chemistry, Analytical
Latifah Almuqren, Fuad Al-Mutiri, Mashael Maashi, Heba Mohsen, Anwer Mustafa Hilal, Mohamed Ibrahim Alsaid, Suhanda Drar, Sitelbanat Abdelbagi
Summary: In this study, a Sine-Cosine-Adopted African Vultures Optimization with Ensemble Autoencoder-based Intrusion Detection (SCAVO-EAEID) technique is proposed to address the security challenges in CPS environments. It focuses on intrusion detection through feature selection and deep learning modeling.
Article
Computer Science, Artificial Intelligence
Stavros Ntalampiras, Ilyas Potamitis
Summary: This paper proposes a modeling scheme for cyber physical systems operating in non-stationary, small data environments using the few-shot learning paradigm. It introduces a change detection mechanism and evaluates the efficacy of the proposed method through experiments. The paper also addresses the interpretability of AI predictions.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Automation & Control Systems
Imran Ahmed, Marco Anisetti, Awais Ahmad, Gwanggil Jeon
Summary: 5G is the foundation for the Industrial Internet of Things (IIoT) and enables efficient integration of artificial intelligence and cloud computing in a smart IIoT ecosystem. However, it also poses security and privacy risks due to increased complexity and new attack vectors. This article presents a 5G-enabled system using deep learning to classify malware attacks on the IIoT, achieving 97% accuracy by extracting discriminative features through convolutional neural networks.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Computer Science, Information Systems
Ioana Apostol, Marius Preda, Constantin Nila, Ion Bica
Summary: The Internet of Things is a cutting-edge technology that requires security solutions. This paper proposes an anomaly-based detection solution using machine learning methods to protect IoT systems. Experimental results demonstrate the performance of the proposed method.
Article
Chemistry, Analytical
Segun I. Popoola, Bamidele Adebisi, Ruth Ande, Mohammad Hammoudeh, Kelvin Anoh, Aderemi A. Atayero
Summary: This paper proposes an efficient DL-based botnet attack detection algorithm that can handle highly imbalanced network traffic data, using Synthetic Minority Oversampling Technique (SMOTE) to achieve class balance. Experimental results show that this approach outperforms state-of-the-art ML and DL models in detecting botnet attacks in IoT networks.
Article
Chemistry, Multidisciplinary
Hend Khalid Alkahtani, Khalid Mahmood, Majdi Khalid, Mahmoud Othman, Mesfer Al Duhayyim, Azza Elneil Osman, Amani A. Alneil, Abu Sarwar Zamani
Summary: The rapid development and widespread utilization of the Internet of Things (IoT) have brought cybersecurity to the forefront. This article introduces an Optimal Graph Convolutional Neural Network based Ransomware Detection (OGCNN-RWD) technique for IoT security. By utilizing learning enthusiasm for feature subset selection and the GCNN model for ransomware classification, the OGCNN-RWD system outperforms other existing techniques with an accuracy of 99.64% according to simulation experiments on a ransomware database.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Interdisciplinary Applications
Marc Schmitt
Summary: This paper investigates AI-based cyber threat detection and evaluates ML-based classifiers and ensembles for anomaly-based malware detection and network intrusion detection. It discusses the challenges of deploying and integrating these models into network security, mobile security, and IoT security, and provides future research directions to enhance the security and resilience of modern digital industries, infrastructures, and ecosystems.
JOURNAL OF INDUSTRIAL INFORMATION INTEGRATION
(2023)
Article
Computer Science, Artificial Intelligence
Sanjeev Kumar, Kajal Panda
Summary: This paper proposes a novel malware detection and classification architecture based on image visualization using fine-tuned convolutional neural networks. The methodology involves using a pre-trained VGG16 model as a feature extractor and different feature selection methods to construct a feature map. The MLP classifier achieves the best accuracy in detecting malware.
APPLIED SOFT COMPUTING
(2023)
Article
Chemistry, Multidisciplinary
Ammar Odeh, Anas Abu Taleb
Summary: Cybersecurity plays a crucial role in various domains including intelligent industrial systems, residential environments, personal gadgets, and automobiles. IoT intrusion detection, using techniques such as deep learning models and anomaly detection algorithms, is vital for safeguarding data integrity, ensuring privacy, and maintaining reliability and safety of critical systems. Deep learning-based intrusion detection approaches, especially the ensemble deep learning models CNN-LSTM and CNN-GRU, have shown impressive performance in accurately detecting and preventing unauthorized or malicious activities in IoT ecosystems.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Fernando J. Rendon-Segador, Juan A. Alvarez-Garcia, Angel Jesus Varela-Vaca
Summary: Cyber-attacks cause significant financial losses, and they are becoming increasingly sophisticated. As a result, there is a high demand for cybersecurity systems to protect both public and private institutions. This study focuses on developing a deep learning model for detecting different cyber-attacks, exploring the relevance of feature selection, and analyzing the importance of attention mechanisms in improving feature assessment. Comparative experiments were conducted using benchmark datasets in the field of cybersecurity.
COMPUTERS & SECURITY
(2023)
Article
Computer Science, Software Engineering
MohammadReza HoseinyFarahabady, Javid Taheri, Albert Y. Zomaya, Zahir Tari
Summary: This article presents a CPU throttling control strategy to optimize the energy consumption of the Apache Storm platform, and validates its effectiveness in a multi-core system.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2023)
Article
Computer Science, Information Systems
Muhammad A. Alsherbiny, Ibrahim Radwan, Nour Moustafa, Deep Jyoti Bhuyan, Muath El-Waisi, Dennis Chang, Chun Guang Li
Summary: In this paper, a deep learning-based model, SynPredict, is proposed to effectively predict the synergy and sensitivity score of chemotherapeutic drug combinations. By fusing the gene expression data of cancer cells and the chemical features of drugs, SynPredict evaluates the combinations in five synergy metrics. The experimental results demonstrate that SynPredict outperforms existing predictive models with a 74% decrease in mean square error. Moreover, the study highlights the importance of considering multiple synergy metrics and combination sensitivity in predictive models.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2023)
Article
Computer Science, Theory & Methods
Wael Issa, Nour Moustafa, Benjamin Turnbull, Nasrin Sohrabi, Zahir Tari
Summary: The Internet of Things (IoT) ecosystem connects physical devices to the internet, offering agility, responsiveness, and potential environmental benefits. Deep learning (DL) algorithms are integrated into IoT applications to learn and infer patterns. However, current IoT paradigms rely on centralized storage and computing, causing scalability, security threats, and privacy breaches. Federated learning (FL) helps preserve data privacy, but faces challenges related to vulnerabilities and attacks. This study reviews blockchain-based FL methods for securing IoT systems, addressing security issues and open research questions, and discussing challenges and risks associated with integrating blockchain and FL in IoT.
ACM COMPUTING SURVEYS
(2023)
Review
Computer Science, Theory & Methods
Muhammad Akbar Husnoo, Adnan Anwar, Nasser Hosseinzadeh, Shama Naz Islam, Abdun Naser Mahmood, Robin Doss
Summary: This paper presents a comprehensive survey of recent advances in False Data Injection (FDI) attacks within active distribution systems in smart grids. It proposes a taxonomy to classify the FDI threats and summarizes the related studies in terms of attack methodologies and implications on electrical power distribution networks. It also identifies research gaps and recommends future research directions.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2023)
Article
Computer Science, Information Systems
Mohanad Sarhan, Siamak Layeghy, Nour Moustafa, Marius Portmann
Summary: In this paper, a collaborative cyber threat intelligence sharing scheme is proposed to design and evaluate a robust ML-based network intrusion detection system using heterogeneous network data samples from different organisations. The scheme utilizes a common format for network data traffic and a federated learning mechanism to protect sensitive users' information. The proposed framework is able to effectively classify various traffic types from multiple organisations without the need for inter-organisational data exchange.
JOURNAL OF NETWORK AND SYSTEMS MANAGEMENT
(2023)
Article
Computer Science, Theory & Methods
Haftu Tasew Reda, Adnan Anwar, Abdun Naser Mahmood, Zahir Tari
Summary: This article presents a comprehensive review of defense countermeasures against false data injection attacks in the Smart Grid. The theoretical and practical significance of relevant existing literature in Smart Grid cybersecurity is evaluated and compared. The study identifies technical limitations of existing false data attack detection research and recommends future research directions.
ACM COMPUTING SURVEYS
(2023)
Article
Computer Science, Software Engineering
Faisal Alam, Adel N. Toosi, Muhammad Aamir Cheema, Claudio Cicconetti, Pablo Serrano, Alesandru Iosup, Zahir Tari, Majid Sarvi
Summary: Rapid growth in popularity of smart vehicles and increasing demand for vehicle autonomy presents new opportunities for vehicular edge computing (VEC). However, VEC offloading poses resource management challenges and is largely inaccessible to automotive companies. This work proposes serverless VEC as an execution paradigm for Internet of Vehicles applications and analyzes its benefits, drawbacks, and technology gaps. Emulation is proposed as a methodology for designing and evaluating serverless VEC solutions, and our toolkit validates the feasibility of serverless VEC for real-world traffic scenarios.
IEEE INTERNET COMPUTING
(2023)
Article
Engineering, Civil
Francesco Schiliro, Nour Moustafa, Imran Razzak, Amin Beheshti
Summary: This paper presents a deep learning-based human cognitive privacy framework called DeepCog, which protects user privacy through feature transforming normalization. Experimental results show that the framework achieves an accuracy of 93.4%, outperforming other techniques.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Information Systems
Marwa Keshk, Nickolaos Koroniotis, Nam Pham, Nour Moustafa, Benjamin Turnbull, Albert Y. Zomaya
Summary: Although XAI has gained significant interest, its implementation in cyber security applications needs further investigation. This paper proposes a novel explainable intrusion detection framework for IoT networks, using a LSTM model and a novel SPIP framework for training and evaluating the model. The SPIP framework achieves high detection accuracy, processing time, and interpretability of data features and model outputs.
INFORMATION SCIENCES
(2023)
Proceedings Paper
Computer Science, Information Systems
Ziaur Rahman, Xun Yi, Ibrahim Khalil, Adnan Anwar, Shantanu Pal
Summary: In recent years, false data injection attacks on intelligent connected vehicles have caused significant industrial losses and loss of lives. Conventional centralized techniques can be misused to maliciously update the legitimate status of vehicles. However, the combination of blockchain and fuzzy logic intelligence shows potential in solving localization issues, trust, and false data detection challenges in autonomous vehicular systems.
THIRD INTERNATIONAL WORKSHOP ON ADVANCED SECURITY ON SOFTWARE AND SYSTEMS, ASSS 2023
(2023)
Article
Engineering, Electrical & Electronic
Haftu Tasew Reda, Adnan Anwar, Abdun Mahmood, Naveen Chilamkurti
Summary: In a smart grid, state estimation is an important component for energy management system, including system SE and anomaly detection. Conventional SE techniques are vulnerable to FDI attack, but this paper proposes a new FDI attack detection technique using a data-driven SE model, which has advantages in maintaining temporal correlations, learning actual power system states, and detecting stealthy attacks. Experimental results show that this technique has a higher detection rate with reduced false alarms.
JOURNAL OF MODERN POWER SYSTEMS AND CLEAN ENERGY
(2023)