WaveThresh Help

wp objects

Wavelet Packet decomposition objects.

DESCRIPTION

These are objects of classes

wp

They represent a decomposition of a function with respect to a set of wavelet packet functions.

GENERATION

This class of objects is returned from the wp function to represent a wavelet packet decomposition of a function. Many other functions return an object of class wp.

METHODS

The wp class of objects has methods for the following generic functions: InvBasis, MaNoVe, accessC, accessD, basisplot, draw, getpacket, nlevels, plot, print, putC, putD, putpacket, summary, threshold.

INHERITANCE

None (yet).

STRUCTURE

The following components must be included in a legitimate `wp' object.

wp: a matrix containing the wavelet packet coefficients. Each row of the matrix contains coefficients with respect to a particular resolution level. There are nlevels(wp)+1 rows in the matrix. Row nlevels(wp)+1 (the ``bottom'') row contains the ``original'' data used to produce the wavelet packet coefficients. Rows nlevels(wp) to row 1 contain coefficients at resolution levels nlevels(wp)-1 to 0 (so the first row contains coefficients at resolution level 0).
The columns contain the coefficients with respect to packets. A different packet length exists at each resolution level. The packet length at resolution level i is given by 2^i. However, the getpacket.wp function should be used to access individual packets from a wp object.
nlevels: The number of levels in the wavelet packet decomposition. If you raise 2 to the power of nlevels you get the number of data points used in the decomposition.
filter: a list containing the details of the filter that did the decomposition (equivalent to the return value from the filter.select function).
date: The date that the transform was performed or the wp was modified.

DETAILS

To retain your sanity we recommend that wavelet packets be extracted in one of two ways:

use getpacket.wp to obtain individual packets.
use accessD.wp to obtain all coefficients at a particular resolution level.

You can obtain the coefficients directly from the wp$wp component but you have to understand their organization described above.