The interconnected and heterogeneous nature of the next-generation Electrical Grid (EG), widely known as Smart Grid (SG), bring severe cybersecurity and privacy risks that can also raise domino effects against other Critical Infrastructures (CIs). In this paper, we present an Intrusion Detection System (IDS) specially designed for the SG environments that use Modbus/Transmission Control Protocol (TCP) and Distributed Network Protocol 3 (DNP3) protocols. The proposed IDS called MENSA (anoMaly dEtection aNd claSsificAtion) adopts a novel Autoencoder-Generative Adversarial Network (GAN) architecture for (a) detecting operational anomalies and (b) classifying Modbus/TCP and DNP3 cyberattacks. In particular, MENSA combines the aforementioned Deep Neural Networks (DNNs) in a common architecture, taking into account the adversarial loss and the reconstruction difference. The proposed IDS is validated in four real SG evaluation environments, namely (a) SG lab, (b) substation, (c) hydropower plant and (d) power plant, solving successfully an outlier detection (i.e., anomaly detection) problem as well as a challenging multiclass classification problem consisting of 14 classes (13 Modbus/TCP cyberattacks and normal instances). Furthermore, MENSA can discriminate five cyberattacks against DNP3. The evaluation results demonstrate the efficiency of MENSA compared to other Machine Learning (ML) and Deep Learning (DL) methods in terms of Accuracy, False Positive Rate (FPR), True Positive Rate (TPR) and the F1 score.

The smart grid (SG) paradigm is the next technological leap of the conventional electrical grid, contributing to the protection of the physical environment and providing multiple advantages such as increased reliability, better service quality, and the efficient utilization of the existing infrastructure and the renewable energy resources. However, despite the fact that it brings beneficial environmental, economic, and social changes, the existence of such a system possesses important security and privacy challenges, since it includes a combination of heterogeneous, co-existing smart, and legacy technologies. Based on the rapid evolution of the cyber-physical systems (CPS), both academia and industry have developed appropriate measures for enhancing the security surface of the SG paradigm using, for example, integrating efficient, lightweight encryption and authorization mechanisms. Nevertheless, these mechanisms may not prevent various security threats, such as denial of service (DoS) attacks that target on the availability of the underlying systems. An efficient countermeasure against several cyberattacks is the intrusion detection and prevention system (IDPS). In this paper, we examine the contribution of the IDPSs in the SG paradigm, providing an analysis of 37 cases. More detailed, these systems can be considered as a secondary defense mechanism, which enhances the cryptographic processes, by timely detecting or/and preventing potential security violations. For instance, if a cyberattack bypasses the essential encryption and authorization mechanisms, then the IDPS systems can act as a secondary protection service, informing the system operator for the presence of the specific attack or enabling appropriate preventive countermeasures. The cases we study focused on the advanced metering infrastructure (AMI), supervisory control and data acquisition (SCADA) systems, substations, and synchrophasors. Based on our comparative analysis, the limitations and the shortcomings of the current IDPS systems are identified, whereas appropriate recommendations are provided for future research efforts.

The rise of the Internet of Medical Things introduces the healthcare ecosystem in a new digital era with multiple benefits, such as remote medical assistance, real-time monitoring, and pervasive control. However, despite the valuable healthcare services, this progression raises significant cybersecurity and privacy concerns. In this article, we focus our attention on the IEC 60 870-5-104 protocol, which is widely adopted in industrial healthcare systems. First, we investigate and assess the severity of the IEC 60 870-5-104 cyberattacks by providing a quantitative threat model, which relies on Attack Defence Trees and Common Vulnerability Scoring System v3.1. Next, we introduce an intrusion detection and prevention system (IDPS), which is capable of discriminating and mitigating automatically the IEC 60 870-5-104 cyberattacks. The proposed IDPS takes full advantage of the machine learning (ML) and software defined networking (SDN) technologies. ML is used to detect the IEC 60 870-5-104 cyberattacks, utilizing 1) Transmission Control Protocol/Internet Protocol network flow statistics and 2) IEC 60 870-5-104 payload flow statistics. On the other side, the automated mitigation is transformed into a multiarmed bandit problem, which is solved through a reinforcement learning method called Thomson sampling and SDN. The evaluation analysis demonstrates the efficiency of the proposed IDPS in terms of intrusion detection accuracy and automated mitigation performance. The detection accuracy and the F1 score of the proposed IDPS reach 0.831 and 0.8258, respectively, while the mitigation accuracy is calculated at 0.923.

Despite the fact that Artificial Intelligence (AI) has boosted the achievement of remarkable results across numerous data analysis tasks, however, this is typically accompanied by a significant short-coming in the exhibited transparency and trustworthiness of the developed systems. In order to address the latter challenge, the so-called eXplainable AI (XAI) research field has emerged, which aims, among others, at estimating meaningful explanations regarding the employed model’s reasoning process. The current study focuses on systematically analyzing the recent advances in the area of Multimodal XAI (MXAI), which comprises methods that involve multiple modalities in the primary prediction and explanation tasks. In particular, the relevant AI-boosted prediction tasks and publicly available datasets used for learning/evaluating explanations in multimodal scenarios are initially described. Subsequently, a systematic and comprehensive analysis of the MXAI methods of the literature is provided, taking into account the following key criteria: a) The number of the involved modalities (in the employed AI module), b) The processing stage at which explanations are generated, and c) The type of the adopted methodology (i.e. the actual mechanism and mathematical formalization) for producing explanations. Then, a thorough analysis of the metrics used for MXAI methods’ evaluation is performed. Finally, an extensive discussion regarding the current challenges and future research directions is provided.