The signal understanding can not be achieved only by the numerical data itself. Therefore, the proper description suitable for the signal recognition should be required. The extended-scale description of signal have been already discussed by using several methods in many fields. In considering practical uses, the authors have a special attention to the scale-space filtering (SSF) by using Gaussian smoothing because of its wide applications. This paper describes the basic treatment of SSF using Gaussian convolution and the discussion on availability of SSF for the underwater signal processing. From the simulation results, it was found that when the some necessary conditions for the signal are satisfied, the SSF has some advantages and it can be effective to recognize the underwater acoustic data.
