Wavelets nonlinear for complex numbers, multiple wavelets are slow

Hi,

There seem to be an error in the idwt function in the way it handles complex numbers. I.e. idwt(1i*a,1i*d,wavelet) is not equal to 1i*idwt(a,d,wavelet). Thus, the wavelet transform is not linear for complex numbers. I've traced the problem to the use of commands like y(:)' in filtering and up-sampling subroutins. I've been evaluating the wavelet toolbox in a 2010 version. Does anyone know if this problem still persists?

I second issue that I have is that I will use n-D wavelets. I can construct these from the 1D wavelets but I cannot find a way to process an array of vectors in one go, like u can with the fft. This makes the transform really slow. Has this feature been implemented now or is there some easy way to get around it.

Best, Anders