Definition: Zero padding consists of appending zeros to a signal. It maps a length signal to a length signal, but need not be an integer multiple of :
The above definition is natural when represents a signal starting at time and extending for samples. If, on the other hand, we are zero-padding a spectrum, or we have a time-domain signal which has nonzero samples for negative time indices, then the zero padding is normally inserted between samples and for odd (note that ), and similarly for even . I.e., for spectra, zero padding is inserted at the point ( ). Figure 8.5 illustrates this second form of zero padding. It is also used in conjunction with zero-phase FFT windows(discussed a bit further below).
Figure 8.5:Illustration of frequency-domain zero padding: a) Original spectrum plotted over the domain where (i.e., as the spectral array would normally exist in a computer array). b) . c) The same signal plotted over the domain which is more natural for interpreting negative frequencies. d) .
Using Fourier theorems, we will be able to show that zero padding in the time domain gives bandlimited interpolation in the frequency domain. Similarly, zero padding in the frequency domain gives bandlimited interpolation in the time domain. This is how ideal sampling rate conversion is accomplished.
It is important to note that bandlimited interpolation is idealinterpolation in digital signal processing.