wp.table(x, cost.fun="entropy", wavelet="s8", n.levels=6, boundary=NULL, precondition=F, pdeg=0, pfrac=0, dual=F, analysis.filter=NULL, synthesis.filter=NULL, scale=NULL, thresh=NULL, p=2, prob=.5) as.ptable(x)
Associated with a wavelet packet table is a table of wavelet packet costs, stored in the attribute pcosts. Additional information about the transform is stored in attribute dictionary.
as.ptable returns x, if x is an object of class which inherits from ptable, wpt, or wp.molecule.
The wavelet packet table has n.levels resolution levels. At resolution level j, a table has n coefficients. When you stack the J+1 resolution levels on top of one another, you get the (J+1)*n table of coefficients.
Each resolution level is divided into 2^j coefficient blocks corresponding to oscillations 0, 1, ..., 2^j-1. The coefficient blocks are called crystals, and have names of the form wj.b where j is the resolution level and b is the oscillation number.
The wp.table function stores a wavelet packet table by level in "sequency" order (increasing oscillation numbers): see Wickerhauser (1994) for details.
You can select wavelet transforms from a wavelet packet table using the best.basis, best.level, as.dwt functions and the Subscript operators.
Refer to the section "Wavelet Packet Tables" in the S+WAVELETS User's Manual for more details about wavelet packet tables.
The default optional arguments for wavelet, n.levels, boundary, precondition, pdeg, pfrac, dual can be reset using function wavelet.options. See wavelet.options for details.
Wickerhauser, M. V. (1994). Adapted Wavelet Analysis -- from theory to software.
xx <- make.signal("twochirp", n=1024) wp0 <- wp.table(xx, wavelet="s8", boundary="zero") plot(wp0)wpi <- wp.table(xx, wavelet="s8", boundary="periodic") plot(wpi)