Joint source channel coding with side information using hybrid digital analog codes
IEEE Transactions on Information Theory
The Gaussian watermarking game
IEEE Transactions on Information Theory
Channel capacity and state estimation for state-dependent Gaussian channels
IEEE Transactions on Information Theory
| Hi-index | 754.84 |
Two new schemes are proposed for transmitting a Gaussian source over a Gaussian channel. These schemes directly generalize previous results of Bross et al. and Puri et al. A scaled version of either the source itself or the quantization error is superimposed on the digital information, and thus serves as effective channel state information (CSI) unknown to the receiver. It is shown that for any power allocation between the coded and uncoded components of transmission, optimal distortion can be achieved by a continuum of auxiliary random variables (rather than only Costa's) if the decoded auxiliary codeword is properly used. This observation provides a new degree of freedom in point-to-point transmission. This freedom, in turn, can be utilized in multiterminal scenarios, as is demonstrated with an example.