structcov.3

speech signal process tools

.\" Copyright (c) 1986-1990 Entropic Speech, Inc.
.\" Copyright (c) 1991 Entropic Research Laboratory, Inc.; All rights reserved
.\" @(#)structcov.3	1.3 02 May 1997 ESI/ERL
.ds ]W (c) 1991 Entropic Research Laboratory, Inc.
.TH STRUCT_COV 3\-ESPSsp 02 May 1997
.SH NAME
struct_cov \- compute a structured covariance estimate of a Toeplitz autocorrelation matrix
.SH SYNOPSIS
.ft B
int 
.br
extern int debug_level;

struct_cov(S, R, distance, size, cvgtst, max_iter)
.br
double **S, *R, *distance, cvgtst;
.br
int size;
.SH DESCRIPTION
.PP
\fIStruct_cov\fR estimates the autocorrelation coefficients, \fIR\fR,
(equivalent to a Toeplitz autocorrelation matrix) that corresponds to
the input sampled covariance matrix (\fIS\fR).  \fISize\fR is the
number of rows (or columns) in the input covariance matrix, and the
number of returned autocorrelation values.  \fIDistance\fR is the
final distortion between the input covariance matrix and the resulting
autocorrelation matrix.  \fIstruct_cov\fP estimates \fIR\fP
iteratively; the estimator terminates after \fImax_iter\fP iterations,
or after the relative change in autocorrelation values falls below the
convergence threshold \fIcvgtst\fR.
.PP
Input sample covariance matrices can be obtained from
\fIestimate_covar\fP (3\-\s-1ESPS\s+1sp).  
.SH DIAGNOSTICS
.PP
\fIStruct_cov\fR returns 0 if no problem is discovered in the process
of estimating the autocorrelation values, and it returns 1, if a
problem arises.  If the external \fIdebug_level\fP is positive, 
debugging output is printed on stderr, with higher values yielding
more verbose output.  
.SH SEE ALSO
.PP
\fIestimate_covar\fP(3\-\s-1ESPS\s+1sp),
\fIstrcov_auto\fP(3\-\s-1ESPS\s+1sp),
.br
\fIgenburg\fP(3\-\s-1ESPS\s+1sp),
\fIget_vburg\fP(3\-\s-1ESPS\s+1sp), 
\fIget_fburg\fP(3\-\s-1ESPS\s+1sp)
.SH REFERENCES
.TP
[1]
J.P.Burg, D.G.Luenberger, D.L.Wenger, "Estimation of Structured
Covariance Matrices" \fIProceedings of the IEEE\fP, Vol. 70, No. 9
September 1982
.TP
[2] T.W. Anderson, "Estimation for Autoregressive Moving Average
Models in the Timne and Frequency Domain," \fIThe Annals of
Statistics\fP, 1977, Vol. 5, No. 5, 842-865.
.TP
[3]
J.E. Shore, "On a Relation Between Maximum Liklihood Classification 
and Minimum Relative-Entropy Classification, \fIIEEE Transactions on 
Information Theory\fP, Vol. IT-30, No. 6, Nov. 1984, pp. 851-854.
.SH AUTHOR
Code by B. Frankel; ESPS modification by D. Burton;
manual page by D. Burton. Additional mods by John Shore