PASSIVE SONAR DETECTION AND LOCALIZATION BY MATCHED VELOCITY FILTERING

This paper presents a new bearings-only method of detecting and tracki ng low signal-to-noise ratio (SNR) wide-band targets on a constant cou rse and velocity trajectory. A track-before-detect strategy based on m atched velocity filtering is adopted using spatial images constructed from a sequence of power bearing map (PBM) estimates accumulated durin g a track. To lower the threshold SNR for detection, a discrete bank o f matched velocity filters integrates the PBM images over a range of h ypothesized trajectories. Such an approach eliminates the need to esti mate the number of target since signal detection is determined by comp aring the output of each matched filter (MF) to a decision threshold. The distribution of the MF output is derived based on a single point t arget in diffuse noise assumption. Receiver operating characteristic c urves show a definite detection gain under low SNR conditions for matc hed velocity filtering (track-before-detect) over detection from a sin gle PBM.