SECURENET 2024

Prof. Liam Murphy received a B.E. in Electrical Engineering from University College Dublin in 1985, and an M.Sc. and Ph.D. in Electrical Engineering and Computer Sciences from the University of California, Berkeley in 1988 and 1992 respectively. He is currently a Full Professor in Computer Science & Informatics at University College Dublin, where he is Director of the Performance Engineering Laboratory. Prof. Murphy has published over 150 refereed journal and conference papers on various topics, including multimedia transmissions, dynamic and adaptive resource allocation algorithms in computer/communication networks, and software performance. His current research projects involve computer network convergence and software performance engineering. Prof. Murphy is a Member of the IEEE (Communications, Broadcasting, and Computer societies) and a Fellow of the Irish Computer Society.

Assoc. Prof. Madhusanka Liyanage

Associate Professor / Ad Astra Fellow and Director of Graduate Research, School of Computer Science, University College Dublin, Ireland

Madhusanka Liyanage is an Assistant Professor/Ad Astra Fellow and Director of Graduate Research at University College Dublin, Ireland. He is also a Docent/Adjunct Professor at the University of Oulu, Finland, the University of Ruhuna, Sri Lanka, and the University of Sri Jarawardhenepura, Sri Lanka. He holds a Doctor of Technology degree from the University of Oulu, Finland (2016) and prestigious fellowships during 2018-2020. Madhusanka has been a Visiting Research Fellow at various renowned institutions globally. He received the "2020 IEEE ComSoc Outstanding Young Researcher" award and was ranked among the World's Top 2% of Scientists in 2021 and 2022. He has over 150+ publications, authored books, edited books, and two patents. Additionally, he serves as an expert consultant at European Union Agency for Cybersecurity (ENISA) and has secured over 5 Million Euro research funding. Currently, he leads three large EU H2020/Horizon Europe projects and is the director of the Netslab team at University College Dublin, Ireland. More info: www.madhusanka.com

Luis Muñoz-González, Telefonica Spain

Dr Luis Muñoz-González is a senior research scientist at Telefónica Research in Barcelona, Spain. Before that, he worked as a research associate in the Department of Computing at Imperial College London, being part of the Resilient Information Systems Security (RISS) group. Dr Muñoz-González obtained a PhD on machine learning at University Carlos III of Madrid, Spain. His PhD thesis on Gaussian Process models for nonstationary regression was recognized with the Extraordinary Doctorate Award. His current research interests lie at the intersection of machine learning and cyber security, including the security of machine learning, federated learning, and machine learning for cyber security.

Abstract: At training time, Federated Learning algorithms can be vulnerable to data and model poisoning attacks, where attackers can manipulate the parameters of the global model to produce indiscriminate or targeted errors when the model is deployed. In this talk we present, adaptive federated averaging, a technique for mitigating the effect of such attacks and detect the malicious participants trying to compromise a federated learning task, including a robust aggregation algorithm capable of ignoring malicious updates and a Hidden Markov Model to model the participants’ behaviour during training.

Sandra Scott-Hayward, Queen's University Belfast, UK

Sandra Scott-Hayward is an Associate Professor with the School of Electronics, Electrical Engineering and Computer Science, and a Member of the Centre for Secure Information Technologies at Queen’s University Belfast (QUB). She began her career in industry and became a Chartered Engineer in 2006. Since joining academia, she has contributed security designs and solutions for softwarized networks based on her research on network security architectures and security functions for emerging networks, specifically considering threat detection and protection mechanisms in programmable networks. She received Outstanding Technical Contributor and Outstanding Leadership awards from the Open Networking Foundation in 2015 and 2016, respectively, having been elected and serving as the Vice-Chair of the ONF Security Working Group from 2015 to 2017. Dr. Scott-Hayward serves on the editorial board for IEEE Transactions on Network and Service Management, and IEEE Transactions on Machine Learning in Communications and Networking. She is Vice-Chair of the IEEE NetSoft Steering Committee. She is Director of the QUB Academic Centre of Excellence in Cyber Security Education (ACE-CSE), co-lead of the QUB Leverhulme Interdisciplinary Network on Algorithmic Solutions (LINAS) doctoral training programme, and a Polymath Fellow of the Global Fellowship Initiative at the Geneva Centre for Security Policy (GCSP) from 2021 to 2023. With LINAS and GCSP, she explores the impact of ML and AI technologies on security and society.

Abstract: Space-air-ground integrated networks (SAGINs) extend terrestrial network infrastructure to non-ter­restrial networks (NTNs), integrating satellite and unmanned aerial vehicle communications, high-altitude platforms, and space systems. These integrated networks can provide connectivity to remote locations, enhance service delivery, and support new ventures. NTN integration is an ambitious mission in 3GPP to extend the Internet to sky and outer space. Machine learning (ML) and artificial intelli­gence (AI) solutions are being explored to advance cooperative and autonomous behaviour, self-organisation, and optimisation in SAGINs with associated security challenges. In this talk, we will explore the security, intelligence, and programmable aspects of SAGIN.

Abstract: Federated learning provides a new way to use Machine Learning models as there is no need to exchange data. The only information exchange between the nodes regards the ML model’s parameters like the weights of a deep neural network. Although FL makes it possible to extract knowledge of training data without direct data exchange, transmitting the model updates could enable an attacker to infer private data. Our proposal for secure FL leverages the (Fully) Homomorphic Encryption (FHE) approach. FHE enables computation on encrypted data without the need for decryption, thus offering a robust solution for privacy-preserving FL. In this presentation we will examine the design considerations for integrating FHE into the Federated Learning model built for CONFIDENTIAL 6G use cases.

Abstract:The sixth generation (6G) of cellular networks is expected to provide a reliable, trustworthy, and resilient service with substantial increase of coverage and network heterogeneity. The emerging technologies such as post-quantum cryptography (PQC), artificial intelligence (AI), machine learning (ML) and distributed ledger technology (DLT) are playing a significant role in achieving the 6G objectives. This talk focuses on confidential networking in 6G, referring to the protection of data transmitted over networks with encryption and secure protocols such as Transport Layer Security (TLS) with regards to new requirements and architectures for the 6G security and privacy such as authentication, encryption, access control and communication.