Synthesis of adaptive processing of a passive fathometer.

Abstract

By using ambient noise array-processing techniques, it has been demonstrated that a passive array can be used as a fathometer [Siderius et al., JASA (2006)] and that adaptive processing gives higher temporal resolution. We show that the waveform extracted with the adaptive algorithm has a phase error of pi relative to that of conventional processing (independently observed at SIO and NURC). For the case of an array with regularly spaced hydrophone elements, the cross-spectral density (CSD) matrix is Toeplitz and the symmetry of Toeplitz matrices allows a simple analytic expression for the phase of the fathometer output. This expression is used to characterize the introduction of phase errors by adaptive processing. The effect of this phase error is investigated with simulated and real data on a vertical hydrophone array.