Distributed Source Coding of Correlated Gaussian Remote Sources

Abstract

We consider the distributed source coding system for $L$ correlated Gaussian observations $Y_i, i=1,2, ..., L$. Let $X_i,i=1,2, ..., L$ be $L$ correlated Gaussian random variables and $N_i,$ $i=1,2,... L$ be independent additive Gaussian noises also independent of $X_i, i=1,2,..., L$. We consider the case where for each $i=1,2,..., L$, $Y_i$ is a noisy observation of $X_i$, that is, $Y_i=X_i+N_i$. On this coding system the determination problem of the rate distortion region remains open. In this paper, we derive explicit outer and inner bounds of the rate distortion region. We further find an explicit sufficient condition for those two to match. We also study the sum rate part of the rate distortion region when the correlation has some symmetrical property and derive a new lower bound of the sum rate part. We derive a sufficient condition for this lower bound to be tight. The derived sufficient condition depends only on the correlation property of the sources and their observations.