The Sagnac interferometer-based fiber-optic hydrophone (S-FOH) exhibits a frequency-dependent response, causing the output signal to deviate from the original acoustic signal, with severe cases leading to signal distortion. A response equalization demodulation algorithm is demonstrated to recover high-fidelity acoustic signals from interference phase signals. An eight-sensor S-FOH array featuring an alternative quadrature phase bias scheme is demonstrated, and experimental verification of the response equalization demodulation algorithm is performed. The temporal relationship of phase modulation pulses and sampling light pulses is analyzed, and a demodulation algorithm is introduced to obtain the phase difference of the Sagnac interferometer. The acoustic pressure sensitivity is equalized to be flat with an average of −135.0 ± 0.4 dB from 10 to 2032 Hz. The pulse response of the S-FOHA after the equalization algorithm is highly similar to the PZT hydrophone output signal, with a correlation coefficient of 0.987.
A novel underwater noise learning model with a local-global convolution-enhanced Transformer(NL-LGCET) is proposed to underwater acoustic signal denoising within complex marine environments.The model introduces a Local-Global Convolution-Enhanced Transformer (LGCET) with a dual-branch mechanism for parallel extraction of local and global temporal features, surpassing traditional serial structures. A Convolution-Enhanced Transformer (CET) block integrates multi-head self-attention with an advanced depthwise separable convolution to capture intricate time-frequency features. Replacing the traditional STFT, a learnable one-dimensional convolution module preserves phase characteristics and adapts to environmental noise for precise feature extraction. Additionally, a noise learning module and time-domain residual mechanism ensure accurate noise estimation and signal reconstruction. Experimental results shows that the proposed model achieves 19.11 dB scale-invariant signal-to-noise ratio (SISNR) and 20.40 dB signal-to-distortion ratio (SDR) on the ShipsEar dataset, representing significant improvements over the optimal baseline. Furthermore, the model exhibits robust generalization capabilities in unknown ship scenarios and real scenarios.
A fibre optic curvature sensor is proposed with the enhanced backscattering sensitive to the deflection of thin structures. A short piece of single mode fibre attached on 1mm thickness titanium substrate is polished with rough sand paper. The rough surface and micro-cracks near fibre core dramatically enhance the backscatter in the fibre. The sensor is experimentally researched with self-pulse reference intensity demodulation method. The minimum detectable curvature is as small as 3.0 × 10−4/m, which means the strain resolution is 0.1με . The backscatter optical fibre curvature senor can be much more compact and convenient to use in smart structure.