Title :
Time-frequency modelling as a mixture of X2 distributions
Author :
Hory, C. ; Martin, N.
Author_Institution :
Lab. des Images et des Signaux, St. Martin d´´Heres, France
fDate :
28 Sept.-1 Oct. 2003
Abstract :
This communication presents a new method for non-stationary signal interpretation, which consists of extracting and characterizing time-frequency structures. It is based on a modelling of a time-frequency representation probability density function as a mixture of central and noncentral chi-square distributions. The problem of interpretation is reduced to a problem of estimation and classification solved by a specific formulation of the EM algorithm for such mixture.
Keywords :
signal representation; statistical distributions; time-frequency analysis; chi-square distributions; non-stationary signal interpretation; probability density function; time-frequency characterization; time-frequency extraction; time-frequency modelling; time-frequency representation; Computer science; Decision making; Gaussian distribution; Parameter estimation; Probability density function; Signal analysis; Skeleton; Time frequency analysis; USA Councils; White noise;
Conference_Titel :
Statistical Signal Processing, 2003 IEEE Workshop on
Print_ISBN :
0-7803-7997-7
DOI :
10.1109/SSP.2003.1289390
