Application of characteristic function to detection in sinusoidal interference plus Gaussian noise

In this work, a detector scheme for the detection of signal in a group of non-Gaussian narrowband interferences and white Gaussian noise is developed. Since there exists no closed-form probability distribution for this type of disturbance modeling, the key innovation lies in the use of characteristic function rather than the probability distribution to both design and implement the detector. Parameter estimation is performed at first step to find the unknown disturbance parameters. The utilized detector uses these parameters to form an approximately Gaussian distributed test statistic based on the empirical characteristic function of received data. Performance of the detector is investigated by means of both analytical and Monte Carlo simulations.