bestlevt (Wavelet Toolbox)

Best level tree (wavelet packet).

[T,D] = bestlevt(T,D)
[T,D,E] = bestlevt(T,D)

Description

bestlevt is a one- or two-dimensional wavelet packet analysis function.

bestlevt computes the optimal complete sub-tree of an initial tree with respect to an entropy type criterion. The resulting complete tree may be of smaller depth than the initial one.

[T,D] = bestlevt(T,D) computes the modified tree structure T and data structure D, corresponding to the best level tree decomposition.

[T,D,E] = bestlevt(T,D) returns the best tree T, data structure D, and in addition, the best entropy value E.

Examples

% Load signal. 
    load noisdopp; 
    x = noisdopp;

% Decompose x at depth 3 with db1 wavelet packets, using 
% default entropy (shannon) and decompose the packet [3 0]. 
    [wpt,wpd] = wpdec(x,3,'db1'); 
    [wpt,wpd] = wpsplt(wpt,wpd,[3 0]);

% Plot wavelet packet tree structure wpt. 
    plottree(wpt) 




% Compute best level tree. 
    [blt,bld] = bestlevt(wpt,wpd);

% Plot best level tree structure blt. 
    plottree(blt)

Algorithm

See besttree algorithm section. The only difference is that the optimal tree is searched among the complete sub-trees of the initial tree.