Minimumcrossentropy spectral analysis of multiple signals
Abstract
This paper presents a new informationtheoretic method for simultaneously estimating a number of power spectra when a prior estimate of each is available and new information is obtained in the form of values of the autocorrelation function of their sum. A derivation of the method from the principle of minimum cross entropy is given, and the method is compared with maximumentropy spectral analysis and with minimumcrossentropy spectral analysis, of which it is a generalization. Some basic mathematical properties are discussed. Two illustrative numerical examples are included, one based on synthetic spectra, and one based on actual speech data.
 Publication:

Naval Research Lab. Report
 Pub Date:
 April 1981
 Bibcode:
 1981nrl..reptR....J
 Keywords:

 Information Theory;
 Minimum Entropy Method;
 Signal Analysis;
 Spectrum Analysis;
 Autocorrelation;
 Monotone Functions;
 Noise Reduction;
 Optimization;
 Power Spectra;
 Probability Distribution Functions;
 Speech;
 Electronics and Electrical Engineering