Article
Computer Science, Hardware & Architecture
Omar Elnakib, Eman Shaaban, Mohamed Mahmoud, Karim Emara
Summary: Internet of Things (IoT) is a disruptive technology with significant importance for Intrusion Detection Systems (IDSs). This paper proposes an enhanced anomaly-based Intrusion Detection Deep learning Multi-class classification model (EIDM) that can accurately classify 15 traffic behaviors including 14 attack types with 95% accuracy. Comparative study shows EIDM outperforms other state-of-the-art deep learning-based IDSs in terms of accuracy and efficiency.
JOURNAL OF SUPERCOMPUTING
(2023)
Article
Computer Science, Information Systems
Mohammed Saleh Ali Muthanna, Reem Alkanhel, Ammar Muthanna, Ahsan Rafiq, Wadhah Ahmed Muthanna Abdullah
Summary: The Internet of Things (IoT) is a multibillion-dollar business that has become mainstream, but its widespread nature makes it vulnerable to cyber-attacks. This research proposes an intelligent, deep learning-based hybrid framework for efficient threat detection in IoT environments. Experimental results demonstrate that the proposed model outperforms other models in terms of accuracy, speed, and other standard evaluation metrics.
Article
Computer Science, Theory & Methods
Souradip Roy, Juan Li, Bong-Jin Choi, Yan Bai
Summary: The increasing popularity of the Internet of Things has led to more security breaches associated with vulnerable IoT devices, emphasizing the importance of employing intrusion detection techniques. Traditional intrusion detection mechanisms may not work well for IoT environments, leading to the proposal of a novel intrusion detection model utilizing machine learning. Through optimizations such as removal of multicollinearity and dimensionality reduction, the model shows promising results with high detection rates and low false alarm rates in experiments on popular datasets.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2022)
Article
Health Care Sciences & Services
Eman Ashraf, Nihal F. F. Areed, Hanaa Salem, Ehab H. Abdelhay, Ahmed Farouk
Summary: This paper proposes a lightweight artificial neural network solution based on blockchain for protecting the privacy of medical data in IoT healthcare applications. By leveraging edge computing and federated learning, the solution conducts detection on smaller datasets, enhancing security performance and achieving higher accuracy for the heterogeneity of data in IoT devices.
Article
Mathematical & Computational Biology
Noor Wali Khan, Mohammed S. Alshehri, Muazzam A. Khan, Sultan Almakdi, Naghmeh Moradpoor, Abdulwahab Alazeb, Safi Ullah, Naila Naz, Jawad Ahmad
Summary: The Internet of Things (IoT) is a technology with vast potential applications, but the security of IoT networks is still a major concern. This paper proposes an intelligent intrusion detection system (IDS) for IoT networks based on deep learning algorithms. The proposed system achieves high accuracy rates of 99% for network flow datasets and 98% for application layer datasets, surpassing previous IDS models.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2023)
Article
Computer Science, Hardware & Architecture
Muneeba Nasir, Abdul Rehman Javed, Muhammad Adnan Tariq, Muhammad Asim, Thar Baker
Summary: This article focuses on intrusion attacks on edge IoT devices and proposes a model named DF-IDS for detecting intrusions in traffic. The model achieves high accuracy and F1 score through feature selection and training of a deep neural network, showing improvement compared to other models and existing studies.
JOURNAL OF SUPERCOMPUTING
(2022)
Article
Multidisciplinary Sciences
Adeel Abbas, Muazzam A. Khan, Shahid Latif, Maria Ajaz, Awais Aziz Shah, Jawad Ahmad
Summary: The IoT domain has evolved significantly in recent years, transforming human lives through automation of daily tasks. In response to the increasing cyber threats in IoT networks, there is a need to enhance intrusion detection systems. This study proposes an ensemble-based intrusion detection model leveraging machine learning techniques, which shows significant improvements in performance compared to existing models.
ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING
(2022)
Article
Computer Science, Information Systems
Youcef Djenouri, Asma Belhadi, Gautam Srivastava, Jerry Chun-Wei Lin, Anis Yazidi
Summary: This paper presents a new framework for intrusion detection in the next-generation Internet of Things. MinMax normalization strategy is used to collect and preprocess data. The Marine Predator algorithm is then used to select relevant features, which are trained with an advanced recurrent neural network. Shapely values are calculated to determine the contribution of each feature, and the proposed system achieved a rate of more than 94%.
COMPUTER COMMUNICATIONS
(2023)
Article
Computer Science, Information Systems
Minxiao Wang, Ning Yang, Ning Weng
Summary: Machine learning-based Network Intrusion Detection Systems (NIDSs) can classify network flow behavior as benign or malicious. However, current ML-based NIDSs are insufficient to generalize in changing network environments like IoT-based smart homes. This paper proposes a novel Transformer-based IoT NIDS method that combines network traffic-based and telemetry data-based NIDSs to achieve improved intrusion detection performance.
Article
Computer Science, Information Systems
Celestine Iwendi, Joseph Henry Anajemba, Cresantus Biamba, Desire Ngabo
Summary: This study focused on intrusion detection systems in smart healthcare, proposing a machine learning support system that combined RF and genetic algorithm for building new intrusion detection systems. The research emphasized the importance of optimizing functionality in achieving better precision, detection rate, and F1 metrics, with the combination of genetic algorithm and RF models achieving high detection and low false alarm rates.
Article
Computer Science, Information Systems
Amir Haider, Muhammad Adnan Khan, Abdur Rehman, Muhib Ur Rahman, Hyung Seok Kim
Summary: The rapid growth of the Internet of Things and computer infrastructure has led to a significant interest in cybersecurity. Researchers have developed an intrusion detection model based on machine learning techniques and demonstrated its superiority through experiments, showing both research and practical significance.
CMC-COMPUTERS MATERIALS & CONTINUA
(2021)
Article
Computer Science, Information Systems
Adel Abusitta, Glaucio H. S. de Carvalho, Omar Abdel Wahab, Talal Halabi, Benjamin C. M. Fung, Saja Al Mamoori
Summary: Internet of Things (IoT) systems are widely used in various industries and government services. However, they are vulnerable to security attacks targeting data integrity and service availability. Detecting anomalous behavior and compromised nodes in IoT systems is more challenging compared to traditional IT networks. Therefore, effective and reliable anomaly detection is essential to ensure the security of IoT-driven decision support systems. In this paper, we propose a deep learning-based anomaly detection method for IoT that can learn robust features in unstable environments. Experimental results demonstrate the effectiveness of our approach in enhancing the accuracy of detecting malicious data compared to existing IoT-based anomaly detection models.
INTERNET OF THINGS
(2023)
Article
Chemistry, Analytical
Mohammed M. Alani, Ali Miri
Summary: With the rapid growth of IoT devices' adoption, security has become increasingly important. In order to counter security threats, we propose an explainable and efficient method to select the most effective features for building highly accurate intrusion detection systems in IoT.
Article
Computer Science, Artificial Intelligence
Romany F. Mansour
Summary: Industry 4.0 has the potential to revolutionize industries by offering intelligent, secure, independent, and self-adaptive Industrial Internet of Things (IIoT) networks. However, security remains a major concern, which can be addressed through the use of Intrusion Detection Systems (IDS) and blockchain technology. This study proposes a Blockchain-Assisted Cluster-based Intrusion Detection System (BAC-IDS) for IIoT, which clusters devices to detect intrusions and enables secure data transmission through blockchain. The experimental results demonstrate the superiority of the proposed BAC-IDS technique over current state-of-the-art techniques.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Chemistry, Analytical
Ketan Kotecha, Raghav Verma, Prahalad Rao, Priyanshu Prasad, Vipul Kumar Mishra, Tapas Badal, Divyansh Jain, Deepak Garg, Shakti Sharma
Summary: In order to predict anomalies more accurately, a reasonably good network intrusion detection system requires high detection rate and low false alarm rate. This paper operates on the UNSW-NB15 Dataset to suggest various models for modern attacks. Alongside detailed modeling, comprehensive data analysis on dataset features is done for better modeling. Moreover, hypothetical ponderings on potential network intrusion detection systems including suggestions on prospective modeling and dataset generation are discussed.