Differentiating random amplitude harmonics from constant amplitude harmonics

Periodograms are useful tools to reveal hidden periodicities in a given time series but the resulting spectral lines have often been associated with constant amplitude harmonics. Possibilities exist where the harmonics actually have non-zero mean random (as opposed to constant) amplitudes because the two can have identical periodograms. Applications exist to support the random amplitude models. Cyclic statistics are employed as effective tools to distinguish constant from random amplitude harmonic models. The algorithms are FFT based and are easy to implement as illustrated by numerical examples.