User Tools

Site Tools

Writing /home/users/ashutosn/public_html/CommNetS2016/dokuwiki/data/cache/d/deca48443c1f0086bfda184b7806fe4e.i failed
Unable to save cache file. Hint: disk full; file permissions; safe_mode setting.
Writing /home/users/ashutosn/public_html/CommNetS2016/dokuwiki/data/cache/d/deca48443c1f0086bfda184b7806fe4e.metadata failed
Writing /home/users/ashutosn/public_html/CommNetS2016/dokuwiki/data/cache/d/deca48443c1f0086bfda184b7806fe4e.i failed
Unable to save cache file. Hint: disk full; file permissions; safe_mode setting.
Writing /home/users/ashutosn/public_html/CommNetS2016/dokuwiki/data/cache/d/deca48443c1f0086bfda184b7806fe4e.xhtml failed

Three talks for the price of one: 1. Concatenated polar codes 2. Coding for causal adversaries 3. Robust network tomography

Mar 17th, 2:00pm-3:00pm. Room TBD

Abstract: These three mini-talks aim to initiate conversations (at the cost of detailed proofs):
1. Polar codes have excited much recent interest as the first provably capacity achieving channel codes with near linear computational complexity. One drawback has been that the probability of success decays sub-exponentially in the block-length. We show how a simple code-concatenation idea improves this to near-exponential decay in the probability of error with no penalty in the computational complexity.
2. Optimal code design in the presence of jamming adversaries is in general a hard problem, and for which many interesting variants have been open for several decades. In this work we relax the problem by considering an adversary who is causal, in the sense that he must decide when and how to jam based on past and current observations. We show new upper bounds in this model, and for several variants demonstrate codes with performance that matches these bounds (in some cases via computationally efficient code designs).
3. Network tomography aims to determine the interior structure of a network via end-to-end measurements. Often, however, the network may contain faulty or malicious nodes. In the setting where each node performs random linear network coding we consider the problems of topology determination, and “error localization”. We give necessary conditions (which are often also sufficient conditions) for these problems, for both random and adversarial errors and erasures.

Biography: Sidharth (Sid) Jaggi is an Assistant Professor at the Dept. of Information Engineering, Chinese University of Hong Kong. He received his Bachelor of Technology degree from the Indian Institute of Technology in 2000, and his Master of Science and Ph.D. degrees from the California institute of Technology in 2001 and 2006 respectively, all in electrical engineering. He was awarded the Caltech Division of Engineering Fellowship 2001-'02, and the Microsoft Research Fellowship for the years 2002-'04. He interned at Microsoft Research, (Redmond, WA, USA) in the summers of 2002-'03 and engaged in research on network coding. He spent 2006 as a Postdoctoral Associate at the Laboratory of Information and Decision Systems at the Massachusetts Institute of Technology. He joined the Department of Information Engineering, the Chinese University of Hong Kong in 2007.

Sidharth's research interests lie at the intersection of information theory, algorithms, and networking. He is currently particularly interested in the field of network coding which neatly merges practice and theory in all three of these fields. However, his interests are eclectic (above all, he likes a good challenge) and he has dabbled in communication complexity, quantum computation, coding theory, random matrix theory, and signal processing for vision. His name (liberally) translated from Sanskrit means “one who proves theorems”, and he intends to keep trying to live up to it.

Host: Alex Dimakis, dimakis [at]

Back to CommNetS Seminar Page

three_talks_for_the_price_of_one/start.txt · Last modified: 2016/09/01 19:15 (external edit)