Estimation and detection in a mixture of symmetric alpha stable and Gaussian interference

We address the problem of detecting a known signal in a mixture of independent symmetric alpha stable (SαS) and Gaussian interference. The detector is based on the optimal likelihood ratio receiver, but uses parameters estimated from a training sample. To this end, we propose a parameter estimator based on nonlinear weighted least squares (NWLS) in the characteristic function domain. We compare the NWLS estimator to the distance and moment methods, against which the NWLS estimator performs well. Finally we demonstrate the near optimal performance of the detector based on the NWLS estimator