Signal Processing for Intelligent Sensor Systems with MATLAB®, 2nd Edition

7    Spectral Density

The spectral density of a signal is a statistical quantity very useful in determining the mean-square value, or power spectral density (PSD), for frequency-domain signals. Strictly speaking, the spectral density, as its name implies, is an expected power density per Hertz. As the time integral in the forward Fourier transform increases, so does the amplitude of the spectral density, while the resolutions of the Fourier transform narrows. Therefore, the expected signal power per Hertz in the frequency domain stays the same, matching the expected signal power at that frequency in the time domain. However, many current signal processing texts define the spectral density with some subtle differences which depend on whether ...