Coeecient Rate and Lossy Source Coding

Campbelll1] derived and deened a quantity called the coeecient rate of a random process that involves the process spectral entropy. Two new derivations of Campbell's coeecient rate are given here. One derivation solidiies the interpretation of this quantity as a coeecient rate, while the second derivation implies a new approach for source compression. We investigate this new approach for transform coding of images and compare its performance to the classical methods in terms of performance for a particular image-coding example. This new compression approach is also contrasted with rate-distortion-theoretic optimal coding with respect to a mean squared error delity criterion when the transform coeecients are Gaussian. It is further shown that this new compression method can have an advantage over rate distortion optimal approaches if the actual bandwidth available is less than the design bandwidth.