ABSTRACT
Power spectral estimation is a process of estimating the different frequency components contained in a signal. Spectral estimation aims at finding the PSD or spectral information of a signal on an average basis, such that the estimate has less bias and variance, apart from having the required frequency resolution. The periodogram estimate of a random process that is windowed with any general window is called modified periodogram. The resolution can be defined as the 3 dB bandwidth of the data window since it transforms into the 6 dB bandwidth of the lag window. The STFT can be interpreted in two different ways: either as the Fourier transform of a windowed sequence or as a linear filtering operation. The segments are allowed to overlap, and hence the number of segments over which averaging is done can be increased, thereby giving a better estimate with reduced variance. The effect of applying a data window is to suppress the effects of side lobes.
