Randomization bounds on Gaussian arbitrarily varying channels
Proceedings of the 2006 Infernational Symposium on Information Theory (ISIT 2006), Seattle, WA, July 2006.
Download
Adobe Portable Document Format - [PDF]
PostScript - [PS]
Slides from talk - [PDF]
Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.
Abstract
The random coding capacity of the Gaussian arbitrarily varying channel (GAVC) under a maximal probability of error criterion is equal to that of an additive white Gaussian noise (AWGN) channel with the interference power as additional noise. The deterministic coding capacity under an average probability of error criterion is zero if the transmitter power does not exceed the interference power. We show that the random coding capacity for average probability error is the same as that for maximal probability of error and that the randomization need only be over a sub-exponential number of codebooks. The achievable error exponent is related to the amount of randomization. An application of these results to the degraded broadcast channel is discussed.
Reference
A.D. Sarwate and M. Gastpar, Randomization bounds on Gaussian arbitrarily varying channels, Proceedings of the 2006 International Symposium on Information Theory (ISIT 2006), Seattle, WA, July 2006.
BibTeX
@INPROCEEDINGS(ads_mg_isit06, AUTHOR = "A.~D.~Sarwate and M.~Gastpar", TITLE = "Randomization bounds on Gaussian arbitrarily varying channels", BOOKTITLE = "Proceedings of the 2006 International Symposium on Information Theory (ISIT 2006)", MONTH = "July", ADDRESS = "Seattle, WA", YEAR = "2006")