Signal detection and localization using the Fourier series method (FSM) and cross-sensor data

Abstract

A significant amount of acoustic information that is available in the individual hydrophone or cross-sensor measurements of line array systems may be irrevocably lost in the conventional beamforming (CBF) process. The Fourier series method (FSM) solution is presented for individual hydrophone pairs or cross-sensor measurements and compared to the CBF output for simulated deterministic acoustic fields consisting of signals embedded in directional noise background. By using the acoustic information lost in CBF, potentially significant improvements in the detection and localization of signals are realized by using FSM. For near broadside, design frequency conditions, FSM gives significant peak signal to background noise enhancement over CBF for a 64-element equally spaced line array. FSM can also resolve two signals in azimuth separated by less than a conventional beamwidth. These array performance improvements result from the accurate estimation of the high order spatial Fourier coefficients of the acoustic field. It is emphasized that FSM requires no a priori assumptions about the unknown acoustic fields required by adaptive beamforming methods and the FSM solution has no mathematical ill conditioning associated with many of the inverse beamforming methods.