Signal Toolkit - corrmtx

Function File: H = corrmtx (x, m)

Function File: H = corrmtx (x, m, method)

Function File: [H, R] = corrmtx (…)

Build a data matrix for autocorrelation matrix estimation.

Given a vector x of length N and a model order m, compute the rectangular Toeplitz matrix H such that H’*H is a biased estimate of the autocorrelation matrix. The size of H depends on the selected method:

’autocorrelation’ (default)

Uses both prewindowed and postwindowed data.

’prewindowed’

Uses prewindowed data only.

’postwindowed’

Uses postwindowed data only.

’covariance’

Uses nonwindowed data.

’modified’

Uses forward and backward prediction error estimates.

The optional second output R is the biased autocorrelation matrix estimate H’*H.

See also: aryule.