IMIoannis Makris

Ioannis Makris

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Ioannis Makris has a total of 28 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    StatAvg: Mitigating Data Heterogeneity in Federated Learning for Intrusion Detection Systems

    by Pavlos S. Bouzinis, Panagiotis Radoglou-Grammatikis, Ioannis Makris, Thomas Lagkas, Vasileios Argyriou, Georgios Th. Papadopoulos, Panagiotis Sarigiannidis, George K. Karagiannidis

    Federated learning (FL) enables devices to collaboratively build a shared machine learning (ML) or deep learning (DL) model without exposing raw data. Its privacy-preserving nature has made it popular for intrusion detection systems (IDS) in the field of cybersecurity. However, data heterogeneity across participants poses challenges for FL-based IDS. This paper proposes statistical averaging (StatAvg) method to alleviate non-independently and identically (non-iid) distributed features across local clients’ data in FL. In particular, StatAvg allows the FL clients to share their individual local data statistics with the server. These statistics include the mean and variance of each client’s feature vector. The server then aggregates this information to produce global statistics, which are shared with the clients and used for universal data normalization, i.e., common scaling of the input features by all clients. It is worth mentioning that StatAvg can seamlessly integrate with any FL aggregation strategy, as it occurs before the actual FL training process. The proposed method is evaluated against well-known baseline approaches that rely on batch and layer normalization, such as FedBN, and address the non-iid features issue in FL. Experiments were conducted using the TON-IoT and CIC-IoT-2023 datasets, which are relevant to the design of host and network IDS, respectively. The experimental results demonstrate the efficiency of StatAvg in mitigating non-iid feature distributions across the FL clients compared to the baseline methods, offering a gain in IDS accuracy ranging from 4% to 17%.


  • Open Access

    AI4FIDS: Multimodal Federated Intrusion Detection

    by Panagiotis Radoglou-Grammatikis, Pavlos S. Bouzinis, Ioannis Makris, Thomas Lagkas, Vasileios Argyriou, Georgios Th. Papadopoulos, Panagiotis Fouliras, George Seritan, Panagiotis Sarigiannidis

    The rapid progression of smart technologies creates several advantages like enhanced connectivity, personalisation solutions and environmental sustainability. However, this revolution creates also several cyber risks. In particular, the attackers have the ability to synthesise and automate advanced attack scenarios over time, while it is evident that Artificial Intelligence (AI) allows the composition of intelligent attack vectors that can adapt in real-time to conventional countermeasures. Despite the fact that AI can also benefit defensive mechanisms, there are still functional and privacy issues that need to be resolved. First, AI requires appropriate datasets that can differ from environment to environment. In addition, these datasets usually are not available due to privacy issues. Finally, adversarial attacks have the ability to target and affect the AI-based decision-making process. Therefore, in light of the previous remarks, we provide AI4FIDS, a multimodal Intrusion Detection System (IDS) for critical infrastructures. AI4FIDS leverages Federated learning (FL) and combines multiple data sources, thus allowing cooperative intelligence across multiple domains in a private manner and minimising the impact of potential adversarial attacks. In this paper, we present in detail the architectural design and specifications of AI4FIDS, while the evaluation results demonstrate their detection performance, taking into account several datasets and aggregation strategies. Finally, based on the evaluation results, we discuss how the overall reliability and detection capabilities (in terms of detecting multi-step attack scenarios) of AI4FIDS can be improved by combining the detection outcomes of the components behind AI4FIDS.


  • Open Access

    A comprehensive survey of Federated Intrusion Detection Systems: Techniques, challenges and solutions

    by Ioannis Makris, Aikaterini Karampasi, Panagiotis Radoglou-Grammatikis, Nikolaos Episkopos, Eider Iturbe, Erkuden Rios, Nikos Piperigkos, Aris Lalos, Christos Xenakis, Thomas Lagkas, et al.


  • Open Access

    NERO: Advanced Cybersecurity Awareness Ecosystem for SMEs

    by Charalambos Klitis, Ioannis Makris, Pavlos Bouzinis, Dimitrios Christos Asimopoulos, Wissam Mallouli, Kitty Kioskli, Eleni Seralidou, Christos Douligeris, Loizos Christofi

    NERO represents a sophisticated Cybersecurity Ecosystem comprising five interconnected frameworks designed to deliver a Cybersecurity Awareness initiative, as advocated by ENISA as the optimal method for cultivating a security-centric mindset among employees to mitigate the impact of cyber threats. It integrates activities, resources, and training to nurture a culture of cybersecurity. NERO primarily equips SMEs with a repository of Cyber Immunity Toolkits, a Cyber Resilience Program, and Gamified Cyber Awareness Training, all accessible through a user-friendly Marketplace. The efficacy and performance of this concept will be affirmed through three distinct use case demonstrations across various sectors: Improving Patient Data Security in Healthcare with Cybersecurity Tools, Enhancing Supply Chain Resilience in the Transportation and Logistics Industry through Cybersecurity Awareness, and Elevating Financial Security via Enhanced Cybersecurity Awareness and Tools.


  • Open Access

    Channel-Aware QUIC Control for Enhanced CAM Communications in C-V2X Deployments Over Aerial Base Stations

    by George Amponis, Thomas Lagkas, Vasileios Argyriou, Panagiotis Radoglou-Grammatikis, Konstantinos Kyranou, Ioannis Makris, Panagiotis Sarigiannidis

    The proliferation of vehicular ad hoc networks necessitates efficient data transfer protocols, particularly in the context of Cellular Vehicle-to-Everything (C-V2X) communications. This paper focuses on enhancing the performance of the Quick UDP Internet Connections (QUIC) protocol, focusing on cooperative vehicular networks supported by aerial drone relays. While QUIC outperforms traditional protocols, its default congestion and flow control mechanisms do not adequately address the unique challenges posed by volatile networks spanning the terrestrial and aerial domains, as they are characterized by frequent topology changes, and high propagation delay volatility. We analyse QUIC's congestion and flow control and propose enhancements to optimize its performance in such networks, specifically designed for C-V2X communications in Open Radio Access Networks (O-RAN). Our proposal adjusts connections' congestion window size and individual streams' flow control windows in a channel-aware manner. Simulation experiments assess the performance of our proposal, comparing it with QUIC's default mechanisms. Our proposal can be seamlessly integrated into existing implementations, making it a viable approach for improving performance and addressing the challenges specific to vehicle-to-drone communications. By addressing QUIC's limitations and optimizing its performance for C-V2X applications in O-RAN, our enhancement offers a valuable contribution towards enabling low-latency, and resource-aware vehicular communications for the realization of autonomous driving and advanced vehicular services.


  • Open Access

    Breaching the Defense: Investigating FGSM and CTGAN Adversarial Attacks on IEC 60870-5-104 AI-enabled Intrusion Detection Systems

    by Dimitrios Christos Asimopoulos, Panagiotis Radoglou-Grammatikis, Ioannis Makris, Valeri Mladenov, Konstantinos E. Psannis, Sotirios Goudos, Panagiotis Sarigiannidis

    In the digital age of the hyper-connected Critical Infrastructures (CIs), the role of the smart electrical grid is crucial, providing several benefits, such as improved grid resilience, efficient energy distribution and smart load and response management. However, despite the several advantages, the rapid evolution of the heterogeneous technologies involved in the smart electrical grid increases the attack surface. In this paper, we focus first our attention on how Artificial Intelligence (AI) can be used to protect the smart electrical grid in terms of detecting efficiently potential cyberattacks and anomalies. Secondly, we investigate how AI can be used to trick AI-enabled detection services, thus resulting in false alarms. In particular, we emphasise on cyberattacks against IEC 60870-5-104, an industrial communication protocol which is widely used in the energy domain. Therefore, a relevant AI-powered Intrusion Detection System (IDS) is provided, utilising strong Machine Learning (ML)/Deep Learning (DL) methods, such as Decision Tree, Random Forest, XGBOOST and deep MultiLayer Perceptron (MLP). On the other hand, we investigate how adversarial attacks can affect the detection performance of the previous IDS. For this purpose, the Fast Gradient Signed Method (FGSM) is examined, and a Conditional Tabular Generative Adversarial Network (CTGAN) adversarial attack generator is implemented. The evaluation results demonstrate the efficiency of the proposed IDS and the aforementioned adversarial attacks.