A cyclostationary feature detector

Cyclostationary models for communications signals have been shown in recent years to offer many advantages over stationary models. Stationary models are adequate in many situations, but they cause important features of the signal to be overlooked. One such important feature is the correlation between spectral components that many signals exhibit. Cyclostationary models allow this spectral correlation to be exploited. This paper presents a signal detector that exploits spectral correlation to determine the presence or absence of a cyclostationary signal in noise. The detector's probability of false alarm is analytically derived. Computer simulations verify that the analytical derivation is correct. The detector's receiver operating characteristic curves are determined from the simulation data and the analytical expression for the probability of false alarm.<<ETX>>