Wavelet transform and its application to the identification of short duration pulses

The authors provide quantitative characterization of individual radar pulses. They chose to use wavelets as an instrument of analysis. Discrete implementation of the wavelet transform (WT) is well known and utilized in many applications. The paper discusses some properties of the WT which are important in the context of the time-frequency signal characterization, but often seem to be omitted or ignored. These properties are related to frequency and phase resolution for signals with several frequency components. The issue which is particularly sensitive is the corruption of higher scale wavelets response to lower frequency components of a signal. This feature of WT is presented in detail with examples demonstrating the problem