Title :
What is the interpretation of spectral entropy?
Author :
Gibson, Jerry D.
Author_Institution :
Dept. of Electr. Eng., Texas A&M Univ., College Station, TX, USA
fDate :
27 Jun-1 Jul 1994
Abstract :
In 1960 Campbell used a version of the asymptotic equipartition property (AEP) to derive a relationship between the entropy of the source power spectral density and the minimum coefficient rate required to encode the source. Abramson (1985) subsequently provided a brief discussion of Campbell´s coefficient rate and a qualitative comparison to the mean squared prediction error of samples taken at the Nyquist rate. Mester and Franke (1992) investigated the spectral entropy as an alternative measure of energy compaction in transform coding. Otherwise there has been very little interest in spectral entropy in the literature. We analyse the spectral entropy and compare it to the familiar entropy rate power for autoregressive processes. We also discuss a possible limitation on its utility for a class of random processes
Keywords :
autoregressive processes; entropy; prediction theory; random processes; signal sampling; source coding; spectral analysis; transform coding; Nyquist rate; asymptotic equipartition property; autoregressive processes; energy compaction; entropy rate power; mean squared prediction error; minimum coefficient rate; random processes; samples; source coding; source power spectral density; spectral entropy; transform coding; Autoregressive processes; Compaction; Eigenvalues and eigenfunctions; Energy measurement; Entropy; Frequency; Random processes; Random variables; Rate-distortion; Spectral analysis;
Conference_Titel :
Information Theory, 1994. Proceedings., 1994 IEEE International Symposium on
Conference_Location :
Trondheim
Print_ISBN :
0-7803-2015-8
DOI :
10.1109/ISIT.1994.395055