Seminar Series on Advances in Telecommunications, Networking and Computing

This seminar series is organized by the Department of Networked Systems and Services to provide unique opportunities to meet internationally recognized leading experts in the fields of Telecommunications, Networking and Computing. The lectures will be given in English by prominent foreign researchers from academia and industry, and they will be open to ALL interested colleagues, PhD students, and students. There is no fixed schedule for the seminar, but the lectures will be organized based on the availability of the invited lecturers, and they will be announced via various mailing lists. In addition, slides and up-to-date information on the program will be published on this site.

For further information, please, contact Dr. Levente Buttyán, Program Chair by e-mail (buttyan (at) hit.bme.hu) or telephone (+36 1 463 1803).

Past years: 2009, 2010, 2011, 2012, 2013, 2014, 2015-2016

Program in 2017

Date Lecturer Title Time Location
Feb 16, 2017 (Thu) Iraklis Symeonidis
COSIC, KU Leuven, Belgium
Car sharing without a physical key: a security and privacy perspective 14:00 IB 110
Sep 18, 2017 (Mon) Prof. Ho Tu Bao
School of Knowledge Science, Japan Advanced Institute of Science and Technology
The infancy of EMR data analytics 10:00 IB 110
Nov 13, 2017 (Mon) Prof. Sebastian Ferrando
Department of Mathematics, Ryerson University (Canada)
Trajectorial Models based on Operational Assumptions 13:00 IB 110

Program in details

Car sharing without a physical key: a security and privacy perspective

Speaker   Iraklis Symeonidis, COSIC, KU Leuven, Belgium
Date and time   Feb 16, 2017, 14:00
Location   BME, Informatics Building, IB 110
Download    

Abstract: This presentation focuses on key-less car sharing systems enabling users to share their cars conveniently without the need of physical keys. In spite of many advantages, physical key-less car sharing systems come with substantial security and privacy challenges. This work provides a comprehensive security and privacy analysis and proposes a high-level model for such a system. Based on this model, we present a methodological threat analysis of the system. This analysis results in a specification of an extensive set of security and privacy requirements. To fulfill the aforementioned security and privacy requirements, we propose a novel physical key-less car sharing protocol. Within this work, we design a decentralized and Privacy-Enhancing Protocol for (Temporary) Car Access Provision named PePTCAP. PePTCAP uses Multi-Party Computation allowing the generation, update, revocation and distribution of an access token for a car in a privacy-preserving manner. In order to solve disputes and to deal with law enforcement requests, our protocol provides forensic evidence of car incidents based on threshold secret sharing. We perform a complexity and practical efficiency analysis for the Multi-Party Computation elements of PePTCAP under realistic scenarios, which demonstrates the feasibility of our solution.

Short bio: Iraklis Symeonidis is a Ph.D. candidate at COSIC (COmputer Security and Industrial Cryptography) at KU Leuven, Belgium. Within COSIC, he is advised by Prof. Bart Preneel. His research interests fall in the intersection between information privacy and security with a specific focus on Online Social Networks and Connected Cars. Iraklis holds a master’s degree in Digital Systems Security from the University of Piraeus, and an engineering degree in Information and Communication Engineering from the University of the Aegean.



The infancy of EMR data analytics

Speaker   Prof. Ho Tu Bao, School of Knowledge Science, Japan Advanced Institute of Science and Technology
Date and time   Sep 18, 2017, 10:00
Location   BME, Informatics Building, IB 110
Download    

Abstract: An electronic medical record (EMR) is the digital version of a patient medical record during his/her hospitalization. It is expected that EMRs will play an evolutional role in medical care and research when large amounts of EMRs can be collected and analyzed. In this talk, we show typical big data properties of EMRs and address the challenges of EMR data analytics. We present problems of our current project and the work on EMR clinical texts in drug safety and effectivenes study.

Short bio: Ho Tu-Bao is currently a professor at Japan Advanced Institute of Science and Technology (JAIST). He also work with John von Neumann (JVN) Institue at Vietnam National University Ho Chi Minh city and Vietnam Institute for Advanced Study in Mathematics (VIASM). He received his B.Tech. from Hanoi University of Technology, master and PhD from Pierre and Marie Curie University. His research interests include AI, machine learning, data mining and bio-medicine informatics. More information: http://www.jaist.ac.jp/~bao



Trajectorial Models based on Operational Assumptions

Speaker   Prof. Sebastian Ferrando, Department of Mathematics, Ryerson University (Canada)
Date and time   Nov 13, 2017, 13:00
Location   BME, Informatics Building, IB 110
Download    

Abstract: We illustrate by example the construction of one-dimensional models for option pricing based on operational and observable features of a single class of investors and a risky asset. Market models are defined based on a class of investors characterized by how they operate on financial data leading to potential portfolio re-balances. Once observable variables are selected for modeling, necessary conditions constraining these variables and resulting from the operational setup are derived. Future uncertainty is then reflected in the construction of combinatorial trajectory spaces satisfying such constraints. In the absence of probability assumptions, a minmax methodology is available to price option contracts; numerical results are presented based on worst case estimation of parameters.

Short bio: Sebastian Ferrando is a Professor in the Department of Mathematics at Ryerson University, Toronto. During 2005-2010 Prof. Ferrando served as Chair of the Department. He received his undergraduate degree in Physics in Argentina and his MSc and PhD in Mathematics at the University of Toronto. His Research Areas include: Real Analysis and Ergodic Theorems, Computational Signal Processing and Wavelets, and Financial Mathematics.




buttyan (at) hit.bme.hu