Wenrui Wang

Optical sparse aperture (OSA) imaging systems synthesize a large effective aperture with reduced volume and weight, but their sparse pupil configurations attenuate the mid-frequency modulation transfer function and degrade image quality. Existing learning-based restoration methods mainly optimize spatial-domain reconstruction and provide limited explicit supervision for the frequency-dependent degradation characteristics of OSA systems. To address this issue, we propose a frequency-domain-guided deep Wiener deconvolution network (FDWDN), which incorporates the point spread function as an explicit physical prior and introduces complex-spectrum supervision into the reconstruction objective. This formulation enables the restoration process to be directly constrained in the frequency domain and naturally supports frequency-selective weighting. At an SNR of 40 dB, FDWDN improves PSNR from 25.16 to 33.43 dB, SSIM from 0.6381 to 0.8767, and mean mid-band Fourier ring correlation from 0.5376 to 0.7375. FDWDN also achieves better overall restoration performance than conventional methods and representative deep restoration networks. These results demonstrate that FDWDN provides accurate spatial reconstruction and improved mid-frequency spectral consistency for OSA image restoration.

We propose and experimentally demonstrate a new polarization-reciprocal angular velocity measurement scheme based on an optoelectronic oscillator (OEO) with a Sagnac loop. The scheme employs a dual parallel Mach-Zehnder modulator (DPMZM) to generate the single-sideband modulated signal containing only the optical carrier (OC) and the +1st-order sideband, which are then separated by a delay interferometer (DI) and propagate in opposite directions in the Sagnac loop. The Sagnac-effect-induced phase difference between them is converted into the frequency shift of the OEO, proportional to the angular velocity. By maintaining the same polarization state for all signals, the system enhances polarization reciprocity and avoids polarization interference. The experiment demonstrates a sensitivity of 43.13 kHz/(rad/s).

Large dynamic range (DR) is one of the primary requirements in fiber-optic acoustic sensing systems, wherein acoustic signals are converted into phase-modulated (PM) signals for detection. In the signal transduction stage, research has been conducted to increase the phase sensitivity. This allows for the conversion of acoustic signals into PM signals with higher phase modulation indices (PMIs), therefore improving the detection of weak acoustic signals. However, higher PMIs induced by strong acoustic signals correspondingly broaden the signal bandwidth. A phase demodulation structure that provides both high-bandwidth capability and low-noise performance is needed to improve both the upper and lower limits of DR. To solve the contradiction between phase demodulation noise and bandwidth, we proposed a novel phase demodulation structure featuring an enhanced phase acquisition method and an enhanced feedback control algorithm. The enhanced phase acquisition method combines high-precision analog-to-digital converter (ADC) sampling data with high-frequency comparator data, enhancing the capability to capture dynamic phase variations. The enhanced feedback control algorithm incorporates a predictive control method using high-order signal models and Kalman Filter (KF) iteration to optimize phase tracking performance. With the proposed demodulation structure, phase variation in the residual signal is reduced and the residual signal bandwidth is compressed before sampling. This combined approach effectively balances noise and bandwidth. The performance of the proposed structure is validated through simulations and experiments, achieving a DR of 170.1 dB@1 kHz with a phase resolution of 4×10 ^-6 rad/ √ Hz in practical tests. Based on our estimation, the optimal DR is approximately 211.5 dB@1 kHz.

Noise floor is an important metric for fiber optic sensors. In particular, common-mode noise (CMN) suppression is critical to improving the system’s ability to detect weak signals. In this paper, an innovative method for reducing noise in white-light-driven sensors is proposed. This structure fully utilizes the broad spectrum of white light and the wavelength selectivity of fiber gratings. The proposed CMN suppression method is capable of suppressing the phase noise of the light source, as well as the noise due to environmental disturbances. The experiments demonstrate that the noise floor of this vibration sensor is about -88 dB/Hz above 30Hz with 10km disturbed transmission fiber. The max 1/f noise reduction can reach approximately 60 dB near 5 Hz, while the max transmission path noise suppression exceeds 103 dB at hundreds of Hz. The proposed structure has the potential to achieve higher resolution in wavelength division multiplexing (WDM) sensor arrays without excessive cost.

A highly sensitive method utilizing coarse frequency sampling and Vernier effect for optoelectronic oscillator (OEO) based fiber Bragg grating (FBG) sensing system is proposed and experimentally demonstrated. To generate the Vernier effect, the frequency sampling interval is set close to the free spectral range of the OEO. Cubic spline fitting and power normalization were applied to obtain smooth sampling curve. The peak frequency of the fitted curve was chosen to measure the spectral shift. In the temperature test experiment, a controllable sensitivity was achieved by changing the sampling interval. The sensitivity of 222.84kHz $/^{\circ }$ C was obtained, which was 127.26 times higher than that of traditional single-loop OEO. Our method requires no modifications to the sensing circuit. Furthermore, it can be used in other OEO-based sensors, indicating promising application prospects.

In advanced sensor system designs, one of the common methods involves converting target physical quantities into phase/frequency-modulated signals. Then, the in-phase/quadrature-phase (I/Q)-based phase demodulation structure is mainly employed to enhance flexibility and efficiency in obtaining the phase information. However, the presence of I/Q imbalance and additive noise in analog I/Q channels greatly degrade the overall system detection performance. Specifically, we investigated the detailed degradation of signal detection, including the introduction of unwanted interference frequency components, elevated noise floor, and distortion of the target signal. To solve this problem, we proposed a novel controlled-carrier closed-loop structure to realize the phase demodulation. This structure, built upon the zero-error closed-loop structure, further mitigates the imbalance distortion and optimizes the noise performance by modifying the control method. The detailed structure and key issues are provided. Further, the effectiveness and enhanced performance of the proposed structure are verified by simulation and experiments. The experimental results align with the simulations and it shows that the proposed structure can achieve enhanced imbalance-distortion mitigation and noise suppression without the need for complex I/Q estimation and compensation.

The white light interferometer is advantageous for wavelength division multiplexing (WDM), but the excessive noise floor limits its application in practicality. In this Letter, we propose a fiber-optic sensor driven by a broadband light source, which uses a fiber-optic Fabry-Perot cavity and a reference interferometer to enhance strain resolution. In the experiment, the strain resolution of a 5.86 m resonant sen root sor is 18.5 f epsilon/ Hz at 1.5 kHz, while the maximum detectable signal is over 230 rad at 1 kHz. With low cost, this method provides a new, to the best of our knowledge, solution for WDM sensing arrays with a large dynamic range. (c) 2024 Optica Publishing Group