Analysis and design of an optimum detector for weak sinusoidal signals

A novel detector for detecting sinusoidal signals with unknown parameters in additive Gaussian noise is proposed. The detector is based on the locally most powerful test (LMP), which is an asymptotic version of the uniformly most powerful test (UMP). It can maximize the probability of detection for a given false alarm rate, especially when the signal is weak. We adopt the estimate-and-plug detector method to complete the design. We carry out the performance analysis for our LMP detector. It is shown that the detector approaches a constant false-alarm rate (CFAR) property for large data records.