ΔΗΜΟΣΙΕΥΣΕΙΣ

Στα πλαίσια του έργου έγιναν οι παρακάτω διεθνείς δημοσιεύσεις σε συνέδρια και περιοδικά:

• • Emmanouel Michailidis, Demosthenes Vouyioukas, A Review on Software-Based and Hardware-Based Authentication Mechanisms for the Internet of Drones, Drones, MDPI Journal of Drones, Special Issue "Security, Privacy and Reliability of Drone Communications for beyond 5G Networks",  February 2022, https://doi.org/10.3390/drones6020041

During the last few years, a wide variety of Internet of Drones (IoD) applications have emerged with numerous heterogeneous aerial and ground network elements interconnected and equipped with advanced sensors, computation resources, and communication units. The evolution of IoD networks presupposes the mitigation of several security and privacy threats. Thus, robust authentication protocols should be implemented in order to attain secure operation within the IoD. However, owing to the inherent features of the IoD and the limitations of Unmanned Aerial Vehicles (UAVs) in terms of energy, computational, and memory resources, designing efficient and lightweight authentication solutions is a non-trivial and complicated process. Recently, the development of authentication mechanisms for the IoD has received unprecedented attention. In this paper, up-to-date research studies on authentication mechanisms for IoD networks are presented. To this end, the adoption of conventional technologies and methods, such as the widely used hash functions, Public Key Infrastructure (PKI), and Elliptic-Curve Cryptography (ECC), is discussed along with emerging technologies, including Mobile Edge Computing (MEC), Machine Learning (ML), and Blockchain. Additionally, this paper provides a review of effective hardware-based solutions for the identification and authentication of network nodes within the IoD that are based on Trusted Platform Modules (TPMs), Hardware Security Modules (HSMs), and Physically Unclonable Functions (PUFs). Finally, future directions in these relevant research topics are given, stimulating further work.

• • Kansizoglou, I., Misirlis, E., & Gasteratos, A., “Learning Long-Term Behavior through Continuous Emotion Estimation” In The 14th PErvasive Technologies Related to Assistive Environments Conference (pp. 502-506), June 2021.

Emotions convey concise information regarding an individual’s internal state, while in the long-term they can be used to form an opinion about his/her overall personality. The latter can be proved particularly vital in many human-robot interaction tasks, like in the case of an assisted living robotic agent, where the human’s mood may in turn require the adaptation of the robot’s behavior. As a result, the paper at hand proposes a novel approach enabling an artificial agent to conceive and gradually learn the personality of a human, by tracking his emotional variations throughout their interaction time. To achieve that, the facial landmarks of the subject are extracted and fed into a Deep Neural Network architecture that estimates the two coefficients of human emotions, viz., arousal and valence, as introduced by the broadly known Russell’s model. Finally, by creating a dashboard for user-friendly display of our results, we present both momentarily and in the long-term the monitored fluctuations of a person’s emotional state.

• • V. Liagkou, P.E. Nastou, P. Spirakis and Y.C. Stamatiou, “On the undecidability of the Panopticon detection problem”,  The 6th International Symposium on Cyber Security, Cryptology and Machine Learning (CSCML 2022), June 2022, Israel.

In this paper we provide a theoretical framework for studying the detectability status of Panopticons based on two theoretical definitions. We show, using Oracle Turing Machines, that detecting modern day, ICT-based, Panopticons is an undecidable problem.

• • Dong Wei, Shan An, Xiajie Zhang, Jiayi Tian, Konstantinos A. Tsintotas, Antonios Gasteratos, and Haogang Zhu, “Dual Regression for Efficient Hand Pose Estimation”, the 2022 IEEE International Conference on Robotics and Automation, May 2022, USA.

Hand pose estimation constitutes prime attainment for human-machine interaction-based applications. Realtime operation is vital in such tasks. Thus, a reliable estimator should exhibit low computational complexity and high precision at the same time. Previous works have explored the regression techniques, including the coordinate regression and heatmap regression methods. Primarily incorporating ideas from them, in this paper, we propose a novel, fast and accurate method for hand pose estimation, which adopts a lightweight network architecture and a post-processing scheme. Hence, our architecture uses a Dual Regression strategy, consisting of two regression branches, namely the coordinate and the heatmap ones, and we refer to the proposed method as DRHand. By carefully selecting the branches' characteristics, the proposed structure has been designed to exploit the benefits of the two methods mentioned above while impoverishing their weaknesses to some extent. The two branches are supervised separately during training, and a post-processing module estimates their outputs to boost reliability. This way, our novel pipeline is considerably faster, reaching 44.39 frames-per-second on an NVIDIA Jetson TX2 graphics processing unit, offering a beyond real-time performance for any custom robotics application. Lastly, extensive experiments conducted on two publicly-available datasets demonstrate that the proposed framework outperforms previous state-of-the-art techniques and can generalize on various hand pose scenarios.

• • V. Liagkou, P.E. Nastou, P. Spirakis and Y.C. Stamatiou, “How hard is it to detect surveillance? A formal study of Panopticons and their detectability problem”,  MDPI Journal of Cryptography, 2022, 6, 42, https://doi.org/10.3390/cryptography6030042.

The Panopticon (which means “watcher of everything”) is a well-known prison structure of continuous surveillance and discipline studied by Bentham in 1785. Today, where persistent, massive scale, surveillance is immensely facilitated by new technologies, the term Panopticon vaguely characterizes institutions with a power to acquire and process, undetectably, personal information. In this paper we propose a theoretical framework for studying Panopticons and their detectability status. We show, based on the Theory of Computation, that detecting Panopticons, modeled either as a simple Turing Machine or as an Oracle Turing Machine, is an undecidable problem. Furthermore, we show that for each sufficiently expressive formal system, we can effectively construct a Turing Machine for which it is impossible to prove, within the formal system, its Panopticon status. Finally, we discuss how Panopticons can be physically detected by the heat they dissipate each time they acquire, effortlessly, information in the form of an oracle and we investigate their detectability status with respect to a more powerful computational model than classical Turing Machines, the Infinite Time Turing Machines (ITTMs).

• • D. Dermanis, A. Bogris, P. Rizomiliotis and C. Mesaritakis, "Photonic Physical Unclonable Function based on Integrated Neuromorphic Devices," in Journal of Lightwave Technology, 2022, doi: 10.1109/JLT.2022.3200307.

In this work, we present a physical unclonable function, implemented using an integrated photonic neuromorphic device. The physical security feature in this case relies on the complex and unpredictable relation between hardware implemented complex weights at the hidden layer of a photonic reservoir computing scheme and the digital trainable weights at the output layer. Numerical simulations confirm that the neural weights are significantly affected by inevitable fabrication related imperfections of the silicon photonic platform. These features can be utilized as a physical root of trust, suitable for authentication/cryptographic applications. The proposed neuromorphic physical unclonable function concept, can be based on different types of neural networks, thus it paves the way for a wide range of photonic devices, able to simultaneously perform efficient non von Neumann computation and security related operations.

• • Vassilis Papataxiarhis, Marilena Athanasiou, Dimitris Karadimas, Katerina Lamprakopoulou, Christos Panagiotou, Alexandros Spournias, Christos Antonopoulos, Nikolaos Voros, “Automated Machine Learning Framework towards Predictive Maintenance for Manufacturing SMEs”, Embedded world Conference 2023, 14 - 16 March 2023, Nuremberg, Germany.

Realization of efficient predictive maintenance techniques through the use of Machine Learning and Artificial Intelligence is promoted as one of the next big breakthroughs in manufacturing industries of different domains. However, although this vision has been endorsed by high interest from both academia and industry alike due to the impressive potential gains, critical challenges remain especially related to the high demands and requirements in terms of investment, complexity, training and time required to actually achieve it.  Driven by such observations this work puts forward a new paradigm proposing a cloud-based service framework allowing the application of ML/AI to develop tailor fitted models for specific scenarios posing the minimum required to the interested SME or industry. By analyzing the architecture and services offered we aim to demonstrate that the framework equally empowers professional and non-professional data scientists indicating specific scenarios. Finally, by adopting adequate services’ model and collaborating with a very active Digital Innovation Hub (DIH) the high impact and viability of the proposed solution is also underpinned.