Likelihood ratios and signal detection for nonGaussian processes

摘要: Abstract : The emphasis is on development of likelihood ratios and detection algorithms for problems involving nonGaussian data. first problem considered that detecting a signal in Gaussian noise. This frequently arises active sonar; it could also be important passive sonar. General results are presented nonsingular ratio. A recursive discrete-time algorithm obtained shown to ratio detector when the signal-plus-noise Gaussian. second major spherically-invariant noise (SIN). model which has been proposed some impulsive-plus-Gaussian environments, closely linked encountered sonar applications. obtained. For known signal, behavior analyzed as sample size increases. Constant-false-alarm-probability detectors given, an example based data illustrates potential loss due using actually SIN.