www.pudn.com > multiwavelet.rar > coef_thadj.m
function thadj=coef_thadj(pflt)
%thadj=coef_thadj(pflt)
%
% This function returns matrix of mean variances of a given transform.
% Each element of this matrix corresponds to certain length of the input
% signal and certain depth of wavelet decomposition. For details see [SW].
%
% Input:
% pflt string of characters, name of the transform; for admissible
% names and short descriptions see below.
%
% Output:
% thadj 7 by 7 real array, mean variances of the transform;
% data in thadj is organized as follows:
% each row of thadj corresponds to the certain length
% of the signal (32, 64, ..., 2048, from top to bottom),
% each column of thadj corresponds to certain number of
% levels of wavelet decomposition (2, 3, ..., 7, from
% left to right)
%
% Admissible Names of the Transforms:
% (all names of wavelet transforms are from coef.m,
% all names of prefilters are from coef_prep.m)
%
% 'ghmrr' 'ghm' multiwavelet transform with oversampled preprocessing,
% 'ghmghmap' 'ghm' multiwavelet transform with with 'ghmap' prefiltering,
% 'ghmghmorap' 'ghm' multiwavelet transform with with 'ghmorap'
% prefiltering,
% 'clrr' 'cl' multiwavelet transform with oversampled preprocessing,
% 'clclap' 'cl' multiwavelet transform with with 'clap' prefiltering,
% 'sa4rr' 'sa4' multiwavelet transform with oversampled preprocessing,
% 'sa4sa4ap' 'sa4' multiwavelet transform with 'sa4ap' prefiltering,
% 'd4scalar' 'd4' scalar wavelet transform,
% 'la8scalar' 'la8' scalar wavelet transform,
% 'bih54nrr' 'bih54n' multiwavelet transform with oversampled
% preprocessing,
% 'bih54nbih5ap' 'bih54n' multiwavelet transform with 'bih5ap' prefiltering,
% 'bih52srr' 'bih52s' multiwavelet transform with oversampled
% preprocessing,
% 'bih52sbih5ap' 'bih54n' multiwavelet transform with 'bih5ap' prefiltering,
% 'bi9scalar' 'bi9' scalar biorthogonal wavelet transform,
%
%
% Example of Usage:
% thadj=coef_thadj('bih552sbih5ap')
% Author: Vasily Strela
% COPYRIGHT 1997,98 by Vasily Strela
a=ones(7);
if strcmp(pflt,'ghmrr')
thadj=sqrt(3/4)*a;
elseif strcmp(pflt,'ghmghmap')
thadj=0.980*a;
elseif strcmp(pflt,'ghmghmorap')
thadj=a;
elseif strcmp(pflt,'clrr')
thadj=sqrt(1/2)*a;
elseif strcmp(pflt,'clclap')
thadj=0.3711*a;
elseif strcmp(pflt,'sa4rr')
thadj=sqrt(1/2)*a;
elseif strcmp(pflt,'sa4sa4ap')
thadj=a;
elseif strcmp(pflt,'d4scalar')
thadj=a;
elseif strcmp(pflt,'la8scalar')
thadj=a;
elseif strcmp(pflt,'bih54nrr')
thadj=[2.4993 0 0 0 0 0 0;
2.4685 3.0360 0 0 0 0 0;
2.5113 3.1514 3.4429 0 0 0 0;
2.4935 3.1412 3.5917 4.0982 0 0 0;
2.4887 3.0965 3.6248 4.1330 4.4911 0 0;
2.4946 3.0963 3.6439 4.0049 4.5797 4.7930 0;
2.4867 3.0894 3.6174 4.0308 4.5419 5.0452 5.3843];
elseif strcmp(pflt,'bih54nbih5ap')
thadj=[1.4128 0 0 0 0 0 0;
1.4379 1.7458 0 0 0 0 0;
1.4624 1.7665 2.0159 0 0 0 0;
1.4913 1.8085 2.0560 2.2913 0 0 0;
1.4630 1.8159 2.0879 2.3826 2.5088 0 0;
1.4885 1.8060 2.0923 2.3942 2.5418 2.8059 0;
1.4873 1.8235 2.0989 2.3659 2.6201 2.9911 3.1561];
elseif strcmp(pflt,'bih52srr')
thadj=[3.0722 0 0 0 0 0 0;
3.0190 2.7322 0 0 0 0 0;
3.0945 2.8258 2.3783 0 0 0 0;
3.0807 2.8146 2.4429 2.0525 0 0 0;
3.0708 2.8130 2.3904 2.1213 1.9204 0 0;
3.0717 2.7951 2.4074 2.1107 1.9203 1.8158 0;
3.0648 2.8064 2.3992 2.0959 1.9120 1.8181 1.7606];
elseif strcmp(pflt,'bih52sbih5ap')
thadj=[1.4009 0 0 0 0 0 0;
1.3620 1.2982 0 0 0 0 0;
1.4346 1.3634 1.1865 0 0 0 0;
1.4471 1.3677 1.1765 1.0112 0 0 0;
1.4244 1.3742 1.1962 1.0403 0.9428 0 0;
1.4441 1.3569 1.1864 1.0420 0.9469 0.8991 0;
1.4474 1.3706 1.1872 1.0386 0.9473 0.8975 0.8655];
elseif strcmp(pflt,'bi9scalar')
thadj=[1.0130 0 0 0 0 0 0;
1.0055 1.0136 0 0 0 0 0;
1.0087 1.0123 1.0113 0 0 0 0;
1.0093 1.0114 1.0109 1.0127 0 0 0;
1.0096 1.0113 1.0112 1.0120 1.0126 0 0;
1.0098 1.0119 1.0120 1.0130 1.0129 1.0129 0;
1.0095 1.0118 1.0126 1.0123 1.0129 1.0131 1.0128];
end