Assessing wavelet decomposition computational complexity

Hello, need the help of experienced wavelet processing gurus here...
I have certain project that is aimed to accomplish a noise filtration
by means of wavelet decomposition, subband processing and then wavelet
composiiton to reconstruct the processed signal.
Once the algorithm will be established, I'll have to try its real-time
implementation thus will have to assess wavelet
decomposition/composition (i.e. DWT and IDWT) computational complexity
to get an estimation whether the entire approach will fit into
real-time processing constrains of the particular system.
What I'm willing to understand how to approach such assessement in
terms of necessary amount of multiplications/summations and other
elementary operations (probably similar to operations count per
DFT/FFT).

Will be grateful for any advises or useful links in this regard.

Thanks, Alex Start a New Thread