Compute-and-Forward: Harnessing Interference with Structured Codes
Proceedings of the IEEE International Symposium on Information Theory (ISIT 2008), Toronto, Canada, July 2008.
Download
Adobe Portable Document Format - [PDF]
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
For a centralized encoder and decoder, a channel matrix is simply a set of linear equations that can be transformed into parallel channels. We develop a similar approach to multi-user networks: we view interference as creating linear equations of codewords and that a receiver's goal is to collect a full rank set of such equations. Our new relaying technique, compute-and-forward, uses structured codes to reliably compute functions over channels. This allows the relays to efficiently recover a linear functions of codewords without recovering the individual codewords. Thus, our scheme can work with the structure of the interference while removing the effects of the noise at the relay. We apply our scheme to a Gaussian relay network with interference and achieve better rates than either compress-and-forward or decode-and-forward for certain regimes.
Reference
B. Nazer and M. Gastpar, Computation over Gaussian Multiple-Access Channels, Proceedings of the IEEE International Symposium on Information Theory (ISIT 2008), Toronto, Canada, July 2008.
BibTeX
@INPROCEEDINGS(bn_mg_isit08, AUTHOR = "B.~Nazer and M.~Gastpar", TITLE = "Compute-and-Forward: Harnessing Interference with Structured Codes", BOOKTITLE = "Proceedings of the IEEE International Symposium on Information Theory (ISIT 2008)", MONTH = "July", ADDRESS = "Toronto, Canada", YEAR = "2008")