Recent Changes - Search:

MLFL Home

University of Massachusetts

MLFL Wiki

Low Probability Of Detection Communication Over Optical Channels Theory And Practice

Abstract:

The ubiquity of high-speed optical networks as well as the specific breaches in privacy of these networks described in the documents released by Edward Snowden necessitates the study of low probability of detection (LPD) communication systems in this domain. Such systems would not just protect the information content of the user's message from being accessed, but prevent the detection of its transmission in the first place. Square root law has been proven achievable for optical LPD communications systems if the adversary's channel measurements are subject to noise (either from the environment or his detection equipment), that is, a transmitter can transmit no more than O(\sqrt{n}) bits in n uses of the channel to the legitimate receiver reliably (i.e. with arbitrarily low probability of decoding error) and covertly (i.e. limiting the adversary’s probability of detecting the transmission arbitrarily close to zero). However, these proofs rely on the on-off keying modulation and random coding schemes that are impractical. We present analysis of LPD communication using pulse position modulation (PPM) and Reed-Solomon error correction codes, which are widely deployed in modern optical communication systems. We first prove that the square root law holds under these constraints and then demonstrate practical viability of optical LPD communication systems by presenting the experimental results from a physical implementation of such a system in a laboratory.

Bio:

Boulat Bash is currently pursuing his Ph.D. in Computer Science from the University of Massachusetts, Amherst. He received his B.A. in Economics from Dartmouth College in 2001 and his M.S. in Computer Science from the University of Massachusetts, Amherst in 2008. He spent the summer of 2013 as a graduate intern and visiting scientist with the Quantum Information Processing Group at Raytheon BBN Technologies. His research interests include signal processing, communications, information theory and privacy.

Edit - History - Print - Recent Changes - Search
Page last modified on March 27, 2014, at 10:12 PM