Passive Sonar Detection and Localization by Matched Filtering

operation that estimates the number of targets. This paper presents a new method of detecting and tracking low signal-to-noise ratio (SNR) wide-band targets on a constant course and velocity trajectory. A track-before-detect strategy is adopted using spatial images constructed from conventional beamformer power bearing maps and a discrete bank of three dimensional matched velocity filters. A Neyman-Pearson detector forms a key feature of this technique, allowing the selection of trajectory solutions to be automated. Theoretical receiver operating characteristic curves show the increase in detection gain under low SNR conditions for matched velocity filtering in comparison to detect-before-track methods. Notationally, boldfaced lower-case and upper-citse symbols denote vectors and matrices, respectively. 11. Background We model the ocean as a non-dispersive, homogeneous propagation medium. The wavefield consists of Ns independent wide-band point sources of acoustic energy located in the far field of a horizontally oriented linear array of M equi-spaced sensors. At time t we denote Ns 4 2 ) = CSl(t) + n(t) (1) 3=1