If I am not mistaken it's a technique introduced by Prof. Ingrid Daubechies in which the WT coefficients are compressed in order to obtain a better time-frequency resolution. From what I understood, it is somehow done by integrating the WT phase x WT coefficient over time. (But I am not sure, the math doesn't seem that hard but I still have some problems understanding it). Please correct me if I am wrong and elaborate some more so that I can better understand it. I would also appreciate your thoughts and advice about the utility and benefits of this technique.