This paper proposes and demonstrates a cilia-inspired fiber-optic inclinometer (CFOI) to achieve high-precision inclination measurement. This sensor uses a reflective polarization-maintaining optical fiber (PMF) embedded with a long-period fiber grating (LPFG) as the "cilia" sensing unit. It is fixed at the upper hinge with ultraviolet (UV) adhesive, which acts as a biomimetic "hair follicle" in the structure, achieving mechanical fixation and deformation transmission. The sensor is integrated into a mechanical structure similar to a hinge. The design of this structure enables the cilia to bend as the hinge rotates, converting the change in inclination into a change in curvature. Thus, the inclination can be measured through the wavelength shift of the reflected spectrum. The experimental results show that the maximum inclination sensitivity of the CFOI sensor within the range of-45 to 45 degrees is-410.81 pm/degrees. Moreover, the temperature-inclination crosstalk is only-0.0438 degrees/degrees C. This structure not only realizes two-dimensional inclination measurement, but also achieves biomimicry of the biological cilia system in mechanical packaging, providing a stable, compact and environmentally interference-resistant optical measurement new solution for inclination sensing.

A scanning-based method is proposed for high-precision two-dimensional shape reconstruction of tubular structures using a compact reflective fiber sensor. Different from conventional distributed or quasi-distributed fiber shape sensing, this method reconstructs the overall shape by moving a compact sensing unit along the target structure. It is especially suitable for narrow spaces and for cases with very small curvature changes or curvature reversals. To realize this method, a phase-shifted reflective long-period fiber grating (PS-RLPFG) is designed and fabricated. A reflective metal film is deposited on the fiber end face, and a π phase shift is introduced to generate two resonance dips for curvature and temperature decoupling. Due to the asymmetric refractive-index modulation induced by CO2 laser irradiation, the sensor exhibits a vector bending response with a maximum curvature sensitivity of 10.18 nm/m−1. The proposed method reconstructs tubular structures with an overall arc length of 175 mm, including constant-curvature cases at 0.1, 1, and 5 m−1, as well as S-shaped geometries. The system achieves a mean absolute error (MAE) as low as 0.1944 mm at a curvature of 0.1 m−1. Across the tested curvature range, the MAE varies from 0.1944 to 1.3002 mm. For the S-shaped tube, stable reconstruction is maintained over 25–70 °C, with MAEs of 0.5642–0.7164 mm. To further evaluate the capability for more complex planar shapes, a multi-segment S-shaped tube with an overall arc length of 435 mm is also reconstructed. The tube consists of four curvature sections of −1, 1.56, −2, and 3 m−1, and the reconstruction achieves an MAE of 0.8541 mm and a maximum error of 1.5690 mm. These results show that the proposed temperature-decoupled scanning-based method provides a compact and practical solution for high-precision static 2-D shape reconstruction of tubular structures, with potential use in endoscope-assisted applications.

This article proposes and demonstrates an all-fiber vector magnetic field sensor based on a cascade structure of single-mode fiber (SMF)-multimode fiber (MMF)- no-core fiber (NCF)-SMF. The sensor employs an axial-offset splicing technique to asymmetrically connect the NCF to both the MMF and the output SMF, thereby constructing a composite waveguide structure with strong modal coupling characteristics. This asymmetric geometry effectively modulates the mode-field distribution of the propagating light, enhances the interaction between the evanescent field in the fiber cladding and the external magnetic fluid (MF), and enables significant responses to both magnetic field intensity and direction. Experimental results show that with a 6 mu m axial offset applied to the NCF, the sensor exhibits excellent sensing performance within a magnetic field range of 8-24 mT, achieving a magnetic field intensity sensitivity of -974 pm/mT and a directional sensitivity of -821 pm/degrees. It also reaches a temperature sensitivity of -1205 pm/degrees C in the 25 degrees C-55 degrees C range. The sensor offers advantages of simple fabrication, low cost, and high sensitivity, demonstrating promising potential for multidisciplinary sensing applications.

Objective Distributed fiber-optic vibration sensing (DFOS) must simultaneously address two open-world challenges in long-distance security monitoring: "unknown-event intrusion" and "extreme imbalance among known classes." Traditional supervised models can only perform closed-set classification and label any disturbance absent from the training set as a generic "anomaly," failing to distinguish risk levels among events such as manual digging, rain washing, or cable cutting. This leads to high false-alarm and missed-alarm rates in the field. This paper aims to build a zero-shot deep-learning framework that combines "high-precision known-class classification" with "fine-grained unknown-class recognition," enabling DFOS systems to automatically discriminate among multiple potential threats without retraining, and providing reliable monitoring for pipelines, railways, borders, and other complex scenarios. Methods The framework consists of three core modules. 1) Multi-scale feature extraction: parallel 1-D CNN branches with small, medium, and large kernels capture fine, medium, and coarse-grained local features from phi-OTDR back-scattered spectra; each branch is followed by ReLU and max-pooling, after which a feature-pyramid network fuses multi-scale information into a unified representation that retains both high-frequency transients and low-frequency trends. 2) Bagging ensemble classification: to counter on-site class imbalance, independent CNN base classifiers are trained, and their posterior probabilities are averaged at inference, markedly reducing environmental noise and overfitting risks for rare classes. 3) Zero-shot prototype network: class-prototype vectors are constructed from the feature means of known classes in the embedding space; for unknown events, Euclidean distances to these prototypes are computed. An adaptive distance threshold implements a dual decision-below the threshold the sample is assigned to the nearest known class, above it is labeled unknown, and further distance ranking provides sub-class ordering of unknown types, achieving "open-set + fine classification." The entire network is trained end-to-end with a composite loss that weights cross-entropy and prototype contrastive loss. The optimizer is Adam with an initial learning rate of 0.001; if validation loss does not improve significantly within 20 epochs, the learning rate is multiplied by 0.75 until a minimum of 0.00001 is reached, and an early-stopping mechanism prevents overfitting. Results and Discussions Experiments are conducted on a real-world dataset collected along a 12 km buried pipeline using our self-developed phi-OTDR system. Four classes-"ambient, car, excavation, pedestrian"-are designated as known, and three additional unknown classes are introduced, maintaining significant time-frequency differences. Without using any labels from the unknown classes, the proposed framework achieves an overall accuracy of 97.6 % on known events and an average accuracy of 58.33 % on the three unknown events. Conclusions The proposed "multi-scale CNN-Bagging ensemble-prototype network" hybrid framework realizes, for the first time in the DFOS domain, zero-shot event recognition. It maintains approximately 98 % precision on known threats and distinguishes multiple unknown disturbance types without retraining, offering operators dual information-"risk level + event category. " This work overcomes the limitations of traditional closed-set recognition, allowing a single sensing system to adapt to open, dynamic, and imbalanced real-world scenarios, markedly reducing on-site false alarms and maintenance costs. Future efforts will introduce incremental learning and cross-domain transfer to shorten adaptation cycles for unknown classes and explore extensions to multimodal sensing networks such as acoustic emission and MEMS.

This work presents a long-period fiber grating (LPFG) curvature sensor fabricated via overlapping discharge. The sensor integrates single-mode and multimode fibers into an interleaved bent-core structure, where structural curvature enhances bending sensitivity. Experiments show maximum sensitivities of 125.23 and -76.60 nm/m-1 in the +/- X directions, with minimal temperature crosstalk (2.3 & times; 10-4 m-1/degrees C) and immunity to strain and torsion. Combining robustness, compactness, and high sensitivity, the proposed sensor demonstrates strong potential for vector curvature sensing with effective crosstalk suppression, offering a novel approach to core-modulated fiber optic sensing.

A Swin-ReconGAN network model is proposed to address the instability in MMF imaging caused by environmental disturbances. Speckle feature transfer is performed to adapt the pretrained reconstruction network, mitigating the impacts of fiber perturbations, speckle drift, and varying scattering media on image reconstruction. The proposed network achieves effective feature transfer imaging for individual fiber bending states using only 200 image-speckle pairs, significantly reducing the required training dataset size and data acquisition costs compared to traditional neural networks. After 50 independent runs in the cross-bending states feature transfer imaging, the Swin-ReconGAN achieved an average structural similarity index measure (SSIM) of 0.705, outperforming both the Transfer Learning U-Net (0.565) and the Scratch U-Net (0.620). Similarly, in the cross-medium feature transfer imaging, the Swin-ReconGAN achieved an average SSIM of 0.680 with a coefficient of variation (CV) of only 1.27%, significantly outperforming the Transfer Learning U-Net (average SSIM 0.497, CV 1.46%) and the Scratch U-Net (average SSIM 0.573, CV 9.84%). Beyond individual states, the Swin-ReconGAN also demonstrates robust capability in multiple discrete bending states feature transfer imaging. By directly performing feature transfer on speckle patterns, this model provides a practical approach for robust speckle reconstruction in small-sample scenarios.

With the rapid development of society, the demand for public health care has increased significantly. Among them, heart rate and respiration, as the core parameters of basic vital signs, are closely related to the occurrence of chronic diseases such as cardiovascular diseases and respiratory diseases. In this article, an optical fiber sensor based on mode interference mechanism is proposed for simultaneous monitoring of human respiration rate and heartbeat signals (including heart rate and ballistocardiogram (BCG) signal). The sensor uses a single-mode fiber-multimode fiber (MMF)-single-mode fiber structure, bent into a balloon-like shape and coated with polydimethylsiloxane (PDMS). The sensor is sensitive to weak vibrations, enabling the monitoring of respiratory, pulse, and heartbeat signals via the vibrations generated by physiological activities. The sensor is placed in the chest to achieve simultaneous monitoring of respiration rate and heartbeat signals. In addition, the sensor is placed on the arm and abdomen, respectively, to achieve separate monitoring of pulse and respiration. Compared with traditional electronic sensors, the sensor has a simple structure, convenient fabrication, stable performance, and anti-electromagnetic interference capability. It has great potential application value in the prevention and monitoring of respiratory diseases and cardiovascular diseases.

In this paper, a polarization-multiplexed long-period fiber grating (LPFG) sensor capable of simultaneous curvature, torsion, and temperature measurements is proposed for fiber-based shape sensing. The sensor is fabricated by CO2 laser inscription on a pre-twisted composite fiber structure comprising a polarization-maintaining fiber (PMF) fusion-spliced between two single-mode fibers (SMFs). The synergistic effects of pre-twist and CO2 laser induced thermal stress significantly modify the internal stress distribution of the PMF, thereby enhancing its elliptical birefringence. This enhancement results in pronounced polarization-dependent resonant wavelengths and sensitivities. The fabricated PMF-LPFG exhibits distinct resonant wavelengths of 1547.4 nm and 1525.6 nm when the input light is aligned with the slow axis (0 degrees) and fast axis (90 degrees), respectively. Experimental results demonstrate that the proposed sensor achieves a maximum torsion sensitivity of 2.24 nm/(rad/m) and a maximum curvature sensitivity of 45.68 nm/m(-1) at 0 degrees polarization, with a corresponding temperature sensitivity of 108.9 p.m./degrees C. Owing to its compact structure, high sensitivity, and polarization-multiplexing capability, the proposed sensor demonstrates strong potential for practical three-dimensional shape sensing applications.

In this paper, we propose a multiplexing method for fiber Bragg gratings (FBGs) and Fabry-Perot interferometers (FPIs) based on optical frequency-domain reflectometry (OFDR) technology and the Vernier effect, this approach establishes a pressure sensing system with temperature compensation. We address the demodulation challenge associated with multiplexing a FBG and a Vernier-effect-enhanced FPI by employing a dedicated digital signal processing (DSP) approach. The system enables high-spatial-resolution multiplexing of sensors, allowing for simultaneous measurement of distance, pressure, and temperature. The specially designed configuration, which combines an FBG with an FPI featuring the Vernier effect, exhibits a pressure sensitivity of 58.517 nm/MPa and a temperature sensitivity of-0.128 nm/degrees C, and incorporates temperature compensation. The results demonstrate that our sensors offer advantages in terms of high sensitivity, easy demodulation, and good stability. The fabricated sensors, after being coated with an organic layer or suitably packaged, can be deployed in various complex real-world scenarios. One potential application is layer-specific sensing within lithium-ion batteries.

We demonstrate a miniature optical fiber vibration accelerometer based on cascaded Fabry-Pérot interferometers (FPIs) fabricated by two-photon polymerization (TPP) three-dimensional (3D) printing. The 3D-printed structure consists of a ring base, a cross-shaped support beam, support beam stiffeners, and a central inertial mass. The structure of the central inertial mass is optimized through finite element simulation. The central inertial mass not only improves the acceleration sensitivity, but also adjust the resonant frequency. The maximum acceleration sensitivity of the accelerometer measured by the edge filtering method is 191.76 mV/g at the resonant frequency of 474 Hz. The operating frequency band of the accelerometer is 2–390 Hz. The proposed accelerometer achieves an acceleration sensitivity of 2.20 mV/g at 300 Hz. The accelerometer exhibits good unidirectional characteristics within a mounting angle range of ±30°. The proposed 3D-printed optical fiber vibration accelerometer features a compact structure, small size, and good unidirectional characteristics, showing potential for applications in the aerospace field, such as vibration monitoring on cable-driven parallel robots.

Achieving both high sensitivity and effective temperature compensation is crucial for high-performance humidity sensors. However, integrating temperature compensation often compromises sensitivity. We develop a fiber-optic surface plasmon resonance (SPR) sensor that integrates a cellulose-based humidity-sensing module and a polydimethylsiloxane (PDMS)-based temperature-sensing module on a plastic-coated multimode optical fiber. By utilizing a highly humidity-sensitive cellulose film, we developed a sensor that delivers high sensitivity while incorporating temperature compensation. The sensor exhibits an average humidity sensitivity of -9.15 nm/RH% and a maximum sensitivity of -14.1 nm/RH% within a relative humidity (RH) range of 25%-70%. In addition, it shows a temperature sensitivity of -2.48 nm/degrees C between 25 degrees C and 45 degrees C. By employing a sensitivity coefficient matrix, the system achieves real-time temperature compensation, thereby significantly enhancing humidity measurement accuracy. Due to its high sensitivity, robust performance, and simple structure, the proposed sensor demonstrates strong potential for practical applications, such as human respiration monitoring and health status analysis.

This paper proposes a novel multi-gear fiber-optic sensing strategy to resolve the long-standing trade-off between high sensitivity and wide dynamic range. The approach employs an array of parallel-connected Fabry-P & eacute;rot interferometers (FPIs), whose free spectral ranges (FSRs) are precisely designed to excite a multi-stage optical Vernier effect. This enables a cooperative "coarse-to-fine" measurement mode: the FPI with the largest FSR first operates independently, using its wide unambiguous detection range (UDR) to coarsely locate the target interval. The system then switches to a high-sensitivity mode, leveraging multi-order beat-frequency amplification for precise measurement within that interval. We constructed a sensing system using four intrinsic FPIs with cavity lengths of 11111 mu m, 10000 mu m, 1000 mu m, and 100 mu m, respectively. Experimental results demonstrate simultaneous unambiguous detection and high-sensitivity response across a temperature range of 30-90 degrees C, achieving a maximum amplification factor of 668.5 & times; and a sensitivity of -17.38 nm/ degrees C. This work provides a novel and effective approach to designing high-performance fiber-optic sensors, successfully integrating wide dynamic range with high sensitivity in a single, flexible system.

Hydrogel flexible optical fibers (HFOFs) have great potential in environmental monitoring and biomedical applications. However, current HFOFs face significant challenges, including the presence of large amounts of free water and low fiber crosslinking density, which result in poor water retention and environmental tolerance. These issues hinder the stable use of HFOFs as sensing waveguides. This study proposes for the first time a high-water-retention, fine-core HFOF. HFOFs are spun from a polyacrylamide hydrogel material containing glycerol using the drawing spinning method. During the drawing and stretching process, the fiber network porosity decreases, and self-assembly induced by water evaporation generates numerous hydrogen bonds. The hydroxyl groups in the glycerol molecules form hydrogen bonds with water molecules, while a large amount of bound water is generated, thereby enhancing the water retention performance of the HFOFs. HFOF demonstrates excellent water retention (water loss rate: <0.1%/day), stretching performance (500%), light-guiding ability (1.45 dB/cm), and recovery properties (1 min). The highly water-retentive HFOFs are applied in various scenarios, including bionic sensors for vibration detection and flexible sensors for monitoring human respiration and heartbeat. This work offers a new solution for improving the stable use of HFOFs and provides new inspiration for bionic monitoring and wearable human sensors.

Low-velocity measurement has broad applications in underwater environmental monitoring. Inspired by the exceptional hydrodynamic properties of seal whiskers, we develop a biomimetic streamlined optical fiber flow velocity sensor. Addressing the limitations of conventional designs that suffer from vortex-induced vibrations or fragility, we propose a synergistic integration strategy to effectively solve the trade-off between sensitivity and robustness. Specifically, we employ a fiber Bragg grating Fabry-Perot interferometer (FBG-FPI) as the core sensing element and incorporate the Pound-Drever-Hall (PDH) ultra-narrow linewidth demodulation technique to enable precise tracking of subtle spectral peak shifts. The sensor achieves a highly linear response in the low-velocity range of 0-15 mm/s, as indicated by a coefficient of determination (R2) of 0.999, a resolution of 1.7 mu m/ s, and a minimum detectable flow velocity of 5.5 mu m/s. Owing to its high sensitivity, strong immunity to electromagnetic interference, and low fabrication cost, the proposed sensor holds significant promise for applications in fluid mechanics research, underwater environmental monitoring, and marine resource exploration.

We propose and demonstrate an optical tweezer technique based on a triple-core fiber. The fiber tweezer supports two independent optical potential wells at its fiber tip, enabling the trapping and optical transport of two biological cells. The light field emitted from the central core of the fiber probe is axially focused via a high-refractive-index glass microsphere to form the first optical potential well. The two symmetric side cores of the fiber probe collectively form a second strongly converging optical potential well. Experiments demonstrate that these two optical potential wells can independently trap two yeast cells along the axial direction, and short-distance optical transport of a single yeast cell between the dual traps is realized by varying the core powers. Theoretical simulations further reveal that this transport function arises from the combined action of the two optical potential wells. The proposed scheme can serve as a fundamental building block for all-optical microfluidic chip technology.

In this paper, an optical fiber vector magnetic field and temperature dual-parameter sensor based on multimode intermodal interference (MMI) is proposed and experimentally demonstrated. The sensor head is prepared by using magnetic fluid (MF) encapsulated single-mode fiber (SMF) -six-hole single-core fiber (SHSCF) -noncore fiber (NCF) -SMF structure. Wherein offset splicing between the lead-in SMF and the SHSCF is adopted to obtain a noncircular symmetry structure and effectively excite higher-order modes. In addition, employing a segment of NCF to be sandwiched between the SHSCF and the lead-out SMF is further to generate richer higher-order modes. Experimental results indicate that the transmission spectrum intensity of the sensor is highly sensitive to the magnetic field, while the wavelength shift of the transmission spectrum is extremely sensitive to temperature. Therefore, wavelength shift can be used to monitor temperature, and intensity can be used to measure the magnetic field. In our experiments, magnetic field intensity sensitivity, direction sensitivity and temperature sensitivity of the proposed fiber sensor were 1.34 dB/mT, 0.23 dB/degrees, and 2.31 nm/degrees C, respectively. The proposed optical fiber sensor has a compact structure, is easy to fabricate, and exhibits high sensitivity. It provides a novel sensing solution for the simultaneous measurement of magnetic field and temperature by combining the MF's optical absorption and refractive index tunability through intensity and wavelength demodulation methods.

This paper presents a novel approach for modulating the surface plasmon resonance (SPR) response through optical fiber curvature. A plastic-clad silica fiber (PCSF) is used as the sensing platform, with the cladding removed to expose the core. A 50 nm Au layer is deposited followed by a 30 nm Ge2Sb2Te5 (GST) layer to create a composite plasmonic interface. The inclusion of the GST layer enhances the local dielectric contrast near the metal surface, improving the refractive index (RI) sensitivity of the SPR sensor. Unlike conventional fiber-optic SPR sensors, this method allows continuous and reversible modulation of optical coupling by mechanically adjusting the fiber curvature within a range of -1/70 m(-1) to 1/70 m(-1). This modulation facilitates dynamic tuning of essential resonance characteristics, including resonance wavelength and full width at half maximum (FWHM), without requiring structural modifications or complex functionalization. Experimental results show a resonance wavelength tuning range of 186 nm at a fixed RI of 1.403 RIU. For RI values between 1.333 and 1.403 RIU, the Au/GST-coated sensor exhibits enhanced performance under curvature modulation, with the sensitivity increasing from 5877.1 nm/RIU to 7971.4 nm/RIU and the figure of merit (FOM) improving from 7.9 (-70 m(-1)) to 45.7 (70 m(-1)), accompanied by a reduction in FWHM from 746.3 nm to 174.6 nm. This curvature modulation technique offers a flexible, compact, and cost-effective means of tuning SPR performance, making it a promising solution for miniaturized, high-performance SPR sensing applications.

Tactile perception plays a vital role in artificial finger pulp skin, especially in regions responsible for grasping and touching tasks, where precise sensing of deformation position and applied force is critical. Conventional demodulation methods often fail to fully leverage temporal correlations in data during the pressing process, limiting the accuracy of tactile demodulation. To address this, we propose a tactile sensing system based on quasi-distributed Fiber Bragg Gratings (FBGs) integrated into artificial finger pulp skin, along with a two-stage hybrid LSTM-Transformer neural network (TSH-LTNN) to jointly reconstruct pressing position and force. The network trains a temporal demodulation model by constructing possible data variations over three consecutive time steps, where the LSTM captures short-term continuity, the Transformer extracts long-range dependencies, and an adaptive fusion module integrates their complementary features. Experimental results show that the proposed model outperforms existing methods. In the 0-30 mm pressing range and 0-14.71 N force range, the mean absolute error (MAE) for position prediction is 0.2331 mm (R2 = 0.9971), and for force prediction, it is 0.303 N (R2 = 0.9829). Compared to the Random Forest model, the TSH-LTNN achieves a 2.34% improvement in position R2 and a 66.06% reduction in MAE. For force prediction, it demonstrates a 2.48% improvement in R2 and a 31.94% reduction in MAE. These results confirm that the proposed system offers precise, stable, and real-time pressure-state demodulation, with strong potential for high-precision haptic feedback applications.

Digital holographic interferometry is a high-precision technique for quantitative imaging and measurement. However, the phase retrieved from interference patterns requires correction to obtain the true object phase. Existing numerical methods typically rely on either assumed prior constraints (as in traditional optimization-based approaches) or accurate correction data labels (as in deep learning-based methods), making it difficult to achieve an optimal balance between improving accuracy and reducing dependence on labelled data. To this end, we propose a joint neural network framework comprising a mask neural network and a phase correction neural network (MNN-PCNN) for joint background segmentation and aberration fitting to achieve phase correction in digital holographic interferometry. Inspired by the masked least-squares polynomial fitting, MNN-PCNN first employs a segmentation-based MNN to separate the pure background regions, thereby eliminating interference from the object. Subsequently, PCNN with embedded Zernike polynomial modes is leveraged to achieve accurate and reliable aberration compensation and noise suppression via self-supervised learning. Both simulation and experimental results demonstrate that the proposed MNN-PCNN outperforms classical polynomial fitting methods as well as the representative UNet. Furthermore, the developed network effectively eliminates phase deviations retrieved across multiple projection angles in digital holographic micro-tomography, which plays a critical role in enhancing the accuracy of quantitative three-dimensional measurements.