DocumentCode :
294756
Title :
Statistical study of reflection coefficients derived from a random AR process
Author :
Tourneret, Jean-Yues ; Lacaze, Bernard
Author_Institution :
GAPSE, ENSEEIHT, Toulouse, France
Volume :
3
fYear :
1995
fDate :
9-12 May 1995
Firstpage :
1709
Abstract :
Being linked by non-linear relations, the AR parameters and reflection coefficients cannot both be Gaussian. However, a statistical study can show that these two sets of parameters are Gaussian asymptotically. The aim of this paper is to show that the convergence rate of the reflection coefficient distribution to the Gaussian one depends on the position of the AR model poles in the unit circle. An analysis of the reflection coefficient Taylor expansion around the AR parameters is proposed to determine this convergence rate
Keywords :
Gaussian distribution; autoregressive processes; convergence of numerical methods; random processes; signal processing; statistical analysis; AR model poles; AR parameters; Gaussian distribution; Taylor expansion; convergence rate; nonlinear relations; random AR process; reflection coefficient distribution; reflection coefficients; signal processing; statistical study; unit circle; Convergence; Estimation theory; Gaussian processes; Parameter estimation; Pattern recognition; Quantization; Reflection; Signal processing; Statistics; Taylor series;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1995. ICASSP-95., 1995 International Conference on
Conference_Location :
Detroit, MI
ISSN :
1520-6149
Print_ISBN :
0-7803-2431-5
Type :
conf
DOI :
10.1109/ICASSP.1995.479935
Filename :
479935
Link To Document :
بازگشت