Algorithms for Rapid and Accurate Depth Localization of Targets for Mine Avoidance

The US Navy's new DD(X) destroyer currently under design features amongst its extensive sensor suite an innovative dual (high frequency and mid-frequency) frequency, active bow sonar. One of the primary missions of the high frequency (HF) component of the bow-sonar will be in-stride mine avoidance. Advanced signal processing algorithms, developed for predecessor sonars aboard modern US submarines, allow high-resolution localization (range, depth and relative azimuth) of mine-like objects. However, the performance of these algorithms is degraded under multipath conditions typical of shallow water (<200m) environments. Here we propose to investigate, develop, demonstrate and implement modifications to the existing signal processing sequence to improve the sonar's mine-hunting capabilities in refractive, multipath environments. These modifications will be implemented in a robust, post-monopulse correlated-field-processing algorithm. The proposed algorithm will endeavor to simultaneously estimate both target location and relevant environmental parameters by comparing the beam-time structure of detections on both the real and imaginary parts of the complex monopulse output to efficiently computed, model estimations of the same. Multi-ping trends in beam-time space of the measured and modeled multipath echo differences will facilitate efficient corrections to the assumed sound speed structure and bathymetry via low order empirical and analytical parametric perturbations