Mahdi Jafari Siavoshani
On information theoretic secrecy
Halls department, Hall 2
Thursday, 28 December 2017
10:00 - 11:00
In this presentation, we talk about "information theoretic secrecy" in general and the problem of secret key agreement among multiple parties in the presence of an eavesdropper in particular. The notion of information theoretic secrecy is much stronger than the classical approach in cryptography which is mainly based on unproven assumptions about computational hardness of some problems. This means that if a crypto system is proven to be information theoretic secure, no matter how computationally powerful an adversary, she cannot find any information about the message.In this talk, we will first review some of the seminal works on this subject, e.g., we will mention to the negative result of perfect secrecy shown by Shannon, and discuss about the value of noise and feedback in increasing the shared secret key generation rate. Then, more specifically, we will consider a group of m trusted and authenticated nodes that aim to create a shared secret key K over a wireless channel in the presence of an eavesdropper Eve. For this setup, we develop information-theoretically secure secret key agreement algorithms and talk about the optimality of such algorithms.
Mahdi Jafari Siavoshani is an Assistant Professor in the Department of Computer Engineering, Sharif University of Technology (SUT), Tehran, Iran, since 2013. He received his B.Sc. degrees in Communication Systems as well as Physics both from SUT in 2005. He was awarded an Excellency scholarship from Ecole Polytechnique Federale de Lausanne (EPFL), Switzerland, where he received the M.Sc. degree in 2007 and the Ph.D. degree in 2012, both in Computer, Communication, and Information Sciences. After completing his Ph.D., he joined the Institute of Network Coding (INC) at the Chinese University of Hong Kong (CUHK) as a postdoctoral fellow from 2012 to 2013. His research interests include Information Processing, Information Theory, Large Scale Networks, Machine Learning, and Graphical Models.