Convolution now gets easier with these two methods for a practical input
signal, which has a quite big length. The OA and OS methods are implemented
using FFT. The larger length input signal gets decomposed into desired length
smaller signals.
For OAM, we use linear convolution using circular convolution. The OSM first overlaps the smaller input signals and then, after obtaining individual smaller output signals, y1, y2,.., etc., the overlapped part is neglected and the remaining elements of the output signals are then clubbed to obtain the actual output signal. Thus, a conclusion can be made that OAM and OSM are block processing techniques and are suitable for real-time implementation of signal processing.
For OAM, we use linear convolution using circular convolution. The OSM first overlaps the smaller input signals and then, after obtaining individual smaller output signals, y1, y2,.., etc., the overlapped part is neglected and the remaining elements of the output signals are then clubbed to obtain the actual output signal. Thus, a conclusion can be made that OAM and OSM are block processing techniques and are suitable for real-time implementation of signal processing.
