getvburg.3

speech signal process tools

.\" Copyright (c) 1987-1990 Entropic Speech, Inc.
.\" Copyright (c) 1991-1997 Entropic Research Laboratory, Inc. All rights reserved.
.\" @(#)getvburg.3	1.4 18 Apr 1997 ESI/ERL
.ds ]W (c) 1997 Entropic Research Laboratory, Inc.
.TH  GET_VBURG 3\-ESPSsp 18 Apr 1997
.SH NAME
get_vburg \- Compute reflection coefficients using a vector Burg method.
.SH SYNOPSIS
.nf
.ft B

get_vburg (x, lnt, r, order, c, rc, pgain, matsiz)
.br
float     *x;
int      lnt;
double    *r;
int    order;
float     *c;
float    *rc;
float *pgain;
int   matsiz;

.fi
.SH DESCRIPTION
.PP
This routine computes the reflection coefficients and the lpc
filter coefficients directly from the data using a Burg-like
algorithm. The resulting reflection coefficients are found to be
very close to those obtained using structured covariance method, 
so we think of this as a fast approximation to structured covariance.
.PP
\fIget_vburg\fP takes data of size \fIlnt\fP in array \fIx\fP as input
and computes \fIorder\fP reflection coefficients, \fIrc\fP[1] to
\fIrc\fP[\fIorder\fP], \fIorder\fP lpc filter coefficients, \fIc\fP[1]
to \fIc\fP[\fIorder\fP], the signal \fIpower\fP *r, and the lpc filter
gain, *\fIpgain\fP.  The parameter \fImatsiz\fP specifies the size of
the sample covariance matrix that is to be used in obtaining the
reflection coefficients (when \fIget_vburg\fP is used in \fIrefcof\fP,
\fImatsiz\fP = \fIorder\fP + 1).  For more details on this algorithm,
see [1].
.SH BUGS
None
.SH SEE ALSO
.PP
.nf
\fIstrcov_auto\fP(3\-\s-1ESPS\s+1sp), \fIgenburg\fP(3\-\s-1ESPS\s+1sp),
\fIestimate_covar\fP(3\-\s-1ESPS\s+1sp), \fIstruct_cov\fP(3\-\s-1ESPS\s+1sp),
\fIget_fburg\fP(3\-\s-1ESPS\s+1sp), \fIget_burg\fP(3\-\s-1ESPS\s+1sp),
\fIcompute_rc\fP(3\-\s-1ESPS\s+1sp), \fIrefcof\fP(1\-\s-1ESPS\s+1) 
.fi
.SH REFERENCES
.\" an internal reference is Shankar Narayan, ETM-S-86-49:ssn, 
.\" Spectrum Analysis of periodic data,
.PP
.TP
[1]
Shankar Narayan and J.P. Burg, "Spectral Estimation of Quasi-Periodic
Data", \fIProceedings ICASSP 87\fP, pp. 944-947.  
.SH AUTHOR
Shankar Narayan