BROAD-BAND DETECTION BASED ON 2-DIMENSIONAL MIXED AUTOREGRESSIVE MODELS

Two new detectors are described to detect a broadband plane wave in a colored noise field using a line array. The signal spectrum and arriva l angle as well as the noise field coloration are not known a priori. Using a two-dimensional mixed autoregressive model, a generalized like lihood ratio test and its weak-signal equivalent, the Rao test, are de rived. The performances of the proposed detectors are shown by compute r simulations to be superior to conventional Fourier-based methods. Fi nally, the detection results based on real data are presented.