Abstract:

In the realm of acoustic signal detection, the identification of weak signals, particularly in the presence of negative signal-to-noise ratios, poses a significant challenge. This challenge is further heightened when signals are acquired through fiber-optic hydrophones, as these signals often lack physical significance and resist clear systematic modeling. Conventional processing methods, e.g., low-pass filter (LPF), require a thorough understanding of the effective signal bandwidth for noise reduction, and may introduce undesirable time lags. This paper introduces an innovative feedback control method with dual Kalman filters for the demodulation of phase signals with noises in fiber-optic hydrophones. A mathematical model of the closed-loop system is established to guide the design of the feedback control, aiming to achieve a balance with the input phase signal. The dual Kalman filters are instrumental in mitigating the effects of signal noise, observation noise, and control execution noise, thereby enabling precise estimation for the input phase signals. The effectiveness of this feedback control method is demonstrated through examples, showcasing the restoration of low-noise signals, negative signal-to-noise ratio signals, and multi-frequency signals. This research contributes to the technical advancement of high-performance devices, including fiber-optic hydrophones and phase-locked amplifiers.

Key words: feedback control method, fiber-optic hydrophone, acoustic signal detection, phase signal