Search

As an inherent property of optical devices, dispersion plays an important role in the areas of optical communication and nonlinear optics. Traditional dispersion optimization approaches are time-consuming and power-hungry. In this paper, to accelerate the design of dispersive optical devices, an indirect inverse design method based on the long short-term memory forward model combined with gradient-free optimization algorithms is proposed. In the case of photonic crystal fiber, the results show that the forward model can predict the group velocity dispersion (GVD) with an accuracy of up to 99.62%, and the calculation speed is more than one thousand times faster than the conventional numerical simulations. The prediction accuracy of the inverse model is higher than 93%, with a calculation time of less than 20 s. In the case of slot waveguide, the results show that the forward model can predict the GVD with a prediction accuracy of 96.99% and the inverse design accuracy goes to 99%. The proposed machine learning model offers an efficient tool for dispersion optimization in both fiber and waveguide platforms.

Optical fiber distributed acoustic sensing (DAS) based on phase-sensitive optical time domain reflectometry (φ-OTDR) is in great demand in many long-distance application fields, such as railway and pipeline safety monitoring. However, the DAS measurement distance is limited by the transmission loss of optical fiber and ultralow backscattering power. In this paper, a DAS system based on multispan relay amplification is proposed, where the bidirectional erbium-doped fiber amplifier (EDFA) is designed as a relay module to amplify both the probe light and the backscattering light. In the theoretical noise model, the parameters of our system are carefully analyzed and optimized for a longer sensing distance, including the extinction ratio (ER), span number, span length, and gain of erbium-doped fiber amplifiers. The numerical simulation shows that a bidirectional EDFA relay DAS system can detect signals over 2500 km, as long as the span number is set to be more than 100. To verify the effectiveness of the scheme, a six-span coherent-detection-based DAS system with an optimal design was established, where the cascaded acoustic-optic modulators (AOMs) were used for a high ER of 104 dB. The results demonstrate that the signal at the far end of 300.2 km can be detected and recovered, achieving a high signal-to-noise ratio of 59.6 dB and a high strain resolution of 51.8pε/Hz at 50 Hz with a 20 m spatial resolution. This is, to the best of our knowledge, a superior DAS sensing distance with such a high strain resolution.

As a low-coherence interferometry-based imaging modality, optical coherence tomography (OCT) inevitably suffers from the influence of speckles originating from multiply scattered photons. Speckles hide tissue microstructures and degrade the accuracy of disease diagnoses, which thus hinder OCT clinical applications. Various methods have been proposed to address such an issue, yet they suffer either from the heavy computational load, or the lack of high-quality clean images prior, or both. In this paper, a novel self-supervised deep learning scheme, namely, Blind2Unblind network with refinement strategy (B2Unet), is proposed for OCT speckle reduction with a single noisy image only. Specifically, the overall B2Unet network architecture is presented first, and then, a global-aware mask mapper together with a loss function are devised to improve image perception and optimize sampled mask mapper blind spots, respectively. To make the blind spots visible to B2Unet, a new re-visible loss is also designed, and its convergence is discussed with the speckle properties being considered. Extensive experiments with different OCT image datasets are finally conducted to compare B2Unet with those state-of-the-art existing methods. Both qualitative and quantitative results convincingly demonstrate that B2Unet outperforms the state-of-the-art model-based and fully supervised deep-learning methods, and it is robust and capable of effectively suppressing speckles while preserving the important tissue micro-structures in OCT images in different cases.

Combining Gaussian beam and annular phase grating (APG) can highlight the spiral stripes with uniform distribution. Here, it shines new light on ultrahigh-order orbital angular momentum (OAM) states detecting, achieving higher than ± 220.

Rapid coherent Raman hyperspectral imaging shows great promise for applications in sensing, medical diagnostics, and dynamic metabolism monitoring. However, the spectral acquisition speed of current multiplex coherent anti-Stokes Raman scattering (CARS) microscopy is generally limited by the spectrometer integration time, and as the detection speed increases, the signal-to-noise ratio (SNR) of single spectrum will decrease, leading to a terrible imaging quality. In this Letter, we report a dual-comb coherent Raman hyperspectral microscopy imaging system developed by integrating two approaches, a rapid delay-spectral focusing method and deep learning. The spectral refresh rate is exploited by focusing the relative delay scanning in the effective Raman excitation region, enabling a spectral acquisition speed of 36 kHz, ≈4 frames/s, for a pixel resolution of 95 × 95 pixels and a spectral bandwidth no less than 200 cm−1. To improve the spectral SNR and imaging quality, the deep learning models are designed for spectral preprocessing and automatic unsupervised feature extraction. In addition, by changing the relative delay focusing region of the comb pairs, the detected spectral wavenumber region can be flexibly tuned to the high SNR region of the spectrum.

Cavity solitons are shape-preserving waveforms infinitely revolving around a cavity, which have numerous applications from spectroscopy to telecommunications. Although the cavity solitons have been widely studied for their special time-frequency characteristics over the past decade, the spectral flatness will be a large limitation in some applications without any extra filtering process. In this paper, we report on the generation of a distinct cavity soliton in a cyclic polarization permutation fiber resonator. With the simultaneous excitation of two orthogonal polarization modes with equally opposite dispersion, vectorial cavity solitons possessing broader and flatter spectra can be generated. In order to verify the concept, a numerical model of the polarization-maintaining fiber is proposed and the soliton with a flattened spectrum can be formed. The results enrich the soliton dynamics in the vectorial dissipation system.

With the developments of the tunable laser source (TLS), there are increasing demands for high-resolution dynamic wavelength calibration in recent years. Considering mutual constraints between wide measurement range and high calibration resolution, we propose a dynamic wavelength calibration method based on an auxiliary Mach-Zehnder interferometer (MZI) and the synchrosqueezed wavelet transform (SSWT). Our proposed method can achieve a calibration resolution of 5 fm and a tuning range of 10 nm. Moreover, the measurement range and spatial resolution of the optical frequency domain reflectometer (OFDR) system are improved to ∼80 m and ∼mm, respectively. Our proposed approach can substantially reduce the subtle spectrum distortion (tens of fm) in coherent optical spectrum analyzer (COSA) systems.

A novel fundamentally mode-locked, GHz-repetition-rate ring cavity Yb-doped femtosecond fiber laser is demonstrated, which utilizes polarization-maintaining gain fiber and is enable by SESAM mode-locking. Thanks to the isolator-free structure, the ring cavity laser is operated bidirectionally and the two polarization-multiplexed output pulse trains are demonstrated synchronous. As a result, tunable waveforms one of which is with reduced pedestal and shorter pulse width in comparison with each individual, are generated by combination of the two orthogonal-polarized output pulses. Furthermore, a similar ring cavity structure that generates GHz picosecond pulses is demonstrated. We believe such high-repetition-rate polarization-multiplexed mode-locked fiber lasers could find further uses in various applications in need of gigahertz repetition rate and tunable waveforms.

Femtosecond laser micromachining has been considered as a powerful tool for fabricating versatile fiber devices and received increasing attention in recent years. Here, we report on a compact sensor by integrating a bridge-like waveguide inside a single-mode fiber to construct an in-line Mach-Zehnder interferometer and then inscribing a second-order Bragg grating in the core of the same fiber. The interference dip shows good performance in torsion sensing – the maximum torsion sensitivity of 1.5573 nm/(rad/m), the ability to identify the torsion direction, and low perturbation of axial strain. In order to compensate the cross impact of temperature, the fiber Bragg grating dip is employed as the second indicator and combined with the interference dip for discriminating temperature and directional torsion simultaneously. The proposed device also has the merits such as compact size, high thermal stability, and so on.

Ultrahigh-order optical orbital angular momentum (OAM) states of the identification over ±270 orders are implemented by annular phase grating (APG) and Gaussian beams with different wavelengths. Particularly, the far-field diffraction intensity patterns feature the spiral stripes instead of Hermitian-Gaussian (HG)-like fringes. It’s worth noting that the spiral stripes present uniform distribution, thus the order of OAM states can be intuitively acquired. More specifically, the OAM states can be confirmed from the total amount and rotating direction of the spiral stripes. Compared with traditional methods, the propose scheme contributes to the perfect-distributed and sharper spiral stripes. Moreover, it also makes an easier observation of the patterns in the CCD camera with limited imaging targets. In our experimental setup, the optical filter is removed and the APG parameters are not strictly required. Therefore, the propose optical transmission system is equipped with the advantages of efficiency, robustness and low cost, which paves a promising way for the communication capacity enhancement.

The multimode waveguide is essential for future mode-division-multiplexing optical interconnection systems to further improve data capacity. However, the complex waveguide design process based on numerical methods is time consuming and requires a lot of computational effort. By using machine learning trained models, one can find the rules from the observed data (sample) and use the learned rules (model) to predict the unknown or unobservable data quickly. In this paper, to accelerate the multimode waveguide design, several regression models consisting of fully connected and long short-term memory layers are employed to predict the effective refractive indices from the fundamental mode to fourth-order TE mode with various waveguide geometries. For air cladding waveguides, the percentages of eligible data whose prediction errors are less than 10 − 3 of different modes are 89.95%, 88.10%, 82.29%, 75.83%, and 71.19%. For S i O 2 cladding ones, they are 95.40%, 92.81%, 90.90%, 81.99%, and 86.39%. Based on these models, two kinds of mode multiplexers are designed, and coupling efficiency of higher than 85% is achieved.

A fiber-optic displacement sensor with ~λ/400 spatial resolution, at an operating wavelength of ~1550 nm, using an off-axis fiber interferometer and achieving the best displacement resolution of 4-nm in the optical domain has been demonstrated.

A temperature insensitive glucose sensor has been presented by narrow band fiber ring laser with 45° tilted fiber Bragg grating. The sensitivity has attained 34.3μm/mM at 0-30 mM and R2 reached to 0.999.

In this paper, we have studied the performance of higher-order cladding modes of long-period fiber gratings (LPFG) in volume and surface-based applications. Distinct properties of higher-order cladding modes are being observed during the transition of modes.

An erbium-doped fiber ring laser proximity sensor with a high accuracy of ±16 nm, when the output power difference between the dual lasing wavelengths (~1560 nm) is smaller than 3 dB wavelength has been demonstrated.

Two frequency combs emitting from a single cavity are of great potential in the field of dual-comb spectroscopy because they are mutually coherent and therefore the common mode noise can be suppressed naturally. However, it is difficult to fully and flexibly control the repetition frequency difference in most of the all-optical schemes. In this paper, a birefringence-compensation Kerr resonator is proposed for the mutual dual-comb generation. It is shown that by offset aligning the fast and slow axis with appropriate fiber length, the total birefringence of the cavity can be equalized while the local one keeps at a high level. Theoretical investigations reveal that the polarization decoupled mutual dual-comb can be generated with nearly the same power level and arbitrary repetition frequency difference. Additionally, a numerical model of polarization-maintaining fiber (PMF) with near-zero dispersion is proposed for the proof of the concept. Based on this fiber, the coherent polarization-decoupled dual-comb with 10-dB bandwidth of 33 nm can be obtained. And the repetition frequency difference can be flexibly tuned compared to the cavity without offset alignment.

We report on the experimental observation of pulsating internal oscillation of soliton pairs featuring twofold relative phase evolution. These results could enrich the transient dynamics of soliton complexes and fuel the molecular analogy.

By tapering a weakly-coupled MCF into a strongly-coupled MCF, supermode interference effects can be generated with the best extinction ratio, 47.4dB and tuning efficiency, 392nm/RIU. Up-conversion nanoparticles are attracted and aligned by hexagonal evanescent fields.

We have designed a CNN model to detect Bragg wavelengths in overlapping spectra. The mean RMS error of 0.123pm and mean testing time of 12.4ms are achieved, which outperforms most of the existing techniques.

We numerically investigate the effect of mode-area dispersion on nonlinear pulse propagation dynamics in gas-filled anti-resonant hollow-core fibers. We find that this effect becomes nonnegligible when the system is pumped close to the zero dispersion.

The phase-shifted fiber Bragg grating (FBG) plays an important role in optical communication and sensing due to its ultra-narrow 3-dB bandwidth. Here, we demonstrate the fabrication and thermal property of a high-quality (Q)-factor phase-shifted helical fiber Bragg grating (PS-HFBG). A single-mode fiber is twisted and then inscribed point-by-point with a third-order uniform FBG by a single round of laser irradiation. The grating is curved slightly into a helical shape after the torsion is released, generating a phase shift in the grating. With annealing treatment, the PS-HFBG responds very stably to temperature with a linear sensitivity of 15.24 pm/°C within the range from 100 to 1100°C. Moreover, the PS-HFBG peak tends to narrower for higher temperature and the minimum 3-dB bandwidth is as low as 32 pm, indicating the highest Q-factor of 4.91 × 104. In addition, the PS-HFBG shows a low strain sensitivity (0.896 pm/ μ ε ). The proposed device is very promising to be applied as a high-precision and stable high-temperature sensor.

Curvature measurement plays an important role in structural health monitoring, robot-pose measuring, etc. High-resolution curvature measurement is highly demanded. In this paper, an optical curvature sensor with high resolution based on in-fiber Mach-Zehnder interferometer (MZI) and microwave photonic filter (MPF) is proposed and experimentally demonstrated. The in-fiber MZI is constructed with a ring-core fiber (RCF) which is fusion spliced between two coreless fibers (CLFs). The structure of CLF-RCF-CLF is then sandwiched between two segments of single-mode fiber (SMF), making the whole interferometer structure of SMF-CLF-RCF-CLF-SMF. The operating principle is that different curvatures will cause the variations of the interference spectrum of MZI due to elastic-optic effect, and then the variations are converted into the frequency-shift of the MPF. The factors affecting the visibility of the interference spectrum are researched. A preliminary exploration of the multiplexing demodulation for the in-fiber-MZIs is also investigated and discussed, which is for the first time to the best of our knowledge, holding great potential to pave the way for constructing the sensing network composed of interferometric sensors. The curvature measurement sensitivity is -147.634 MHz/m-1, and the resolution is 6.774 × 10−6 m-1 which is the highest value up to now.

We propose and demonstrate a compact and highly efficient single-pixel imaging system with data compression based on the traditional serial time-encoded amplified microscopy (STEAM) structure. Our proposal solves the big data issue and low-efficiency bottleneck in the conventional STEAM systems. Also, a cascaded Mach-Zehnder interferometric (MZI) structure as the optical random pattern generator is proposed and experimentally demonstrated as the low-cost approach to achieve data compression. A 45° tilted fiber grating (TFG) is applied to increase the efficiency of the imaging system. A state-of-art experiment is performed and data compression ratios from 5% to 20% at an imaging size of 1mm*1.8mm are conducted using our proposed system for the first time to the best of our knowledge. The proposed optical imaging system mitigates the big data issue, decreases the system cost, and increases energy efficiency, and lays a solid foundation for low-cost, compact, and highly efficient ultrafast optical imaging.

In this paper, a dual-mode interferometric sensor based on a subwavelength grating (SWG) waveguide is proposed. The design operates at the dispersion turning point (DTP) of the two dispersive TE modes in the periodic structure, exhibiting a group velocity difference of almost zero, to improve the performance of refractive index (RI) sensors. The necessary conditions for realizing DTP in SWG are analyzed. Moreover, the interference spectrum characteristics and the influence of structural parameters on gas RI sensing performance are further studied. The results show that the sensitivity value of the proposed sensor can reach ∼ 38 , 000 n m / R I U . The proposed structure can also achieve a gas pressure sensitivity of ∼ 105 n m / M P a . The proposed sensors provide a potential platform for highly sensitive RI sensors.

Long-range surface plasmon resonances (LRSPRs) are featured with longer propagation and deeper penetration, compared with conventional surface plasmon resonances (SPRs). Thus, LRSPR-based fiber sensors are considered to have great potential for highly sensitive detection in chemistry or biomedicine areas. Here, we propose and demonstrate a near-infrared LRSPR sensor based on a D-shaped honeycomb microstructured optical fiber (MOF) directly coated with gold film. Although there is no additional heterogeneous buffer layer, the optical field of the long-range surface plasmon polariton (LRSPP) mode penetrates strongly into the analyte region. Thus the effective refractive index of the LRSPP mode depends highly on the analyte’s material refractive index and an abnormal dispersion relationship between the LRSPP mode and MOF’s y-polarized core mode is observed. The mechanism of the LRSPR excitation in the coupling zone is attributed to an avoided crossing effect between these two modes. It also results in the generation of a narrow-bandwidth peak in the loss spectrum of the core mode. Further discussion shows that the resonance wavelength is mainly determined by the core size that is contributed by the MOF’s cladding pitch, silica-web thickness and planar-layer-silica thickness together. It indicates that the operation wavelength of the proposed LRSPR device can be flexibly tuned in a broadband wavelength range, even longer than 2 µm, through appropriately designing the MOF’s structural parameters. Finally, the proposed LRSPR sensor shows the highest wavelength sensitivity of 14700 nm/RIU and highest figure of merit of 475 RIU−1 for the analyte refractive index range from 1.33 to 1.39.

In this study, the laser sintering technique combined with the stack-capillary-draw method was used to manufacture Er3+/Yb3+ co-doped silica glass microstructure optical fiber for up-conversion (UC) luminescence for the first time. The obtained fiber, whose structure can be maintained well, shows no obvious element diffusion during the fiber-drawing process. The core glass is amorphous without any crystallization, and Er and Yb elements are uniformly distributed throughout the fiber core. Intense upconverted green and red emissions corresponding to (2H11/2, 4S3/2) → 4I15/2 and 4F9/2 → 4I15/2 transitions, respectively, together with an emission corresponding to 2H11/2 → 4I13/2 transition were observed upon the excitation of a 976-nm laser. The intensity of green and red UC luminescence shows a linear dependence on the excitation power which demonstrates that a two-photon absorption process is responsible for the UC process. The results demonstrated the laser sintering technique provides an alternative method to manufacture Er3+/Yb3+ co-doped fiber, which would promote the development of UC luminescent materials.

In quasi-distributed fiber Bragg grating (FBG) sensor networks, challenges are known to arise when signals are highly overlapped and thus hard to separate, giving rise to substantial error in signal demodulation. We propose a multi-peak detection deep learning model based on a dilated convolutional neural network (CNN) that overcomes this problem, achieving extremely low error in signal demodulation even for highly overlapped signals. We show that our FBG demodulation scheme enhances the network multiplexing capability, detection accuracy and detection time of the FBG sensor network, achieving a root-mean-square (RMS) error in peak wavelength determination of < 0.05 pm, with a demodulation time of 15 ms for two signals. Our demodulation scheme is also robust against noise, achieving an RMS error of < 0.47 pm even with a signal-to-noise ratio as low as 15 dB. A comparison on our high-performance computer with existing signal demodulation methods shows the superiority in RMS error of our dilated CNN implementation. Our findings pave the way to faster and more accurate signal demodulation methods, and testify to the substantial promise of neural network algorithms in signal demodulation problems.

A two-core photonic crystal fiber (TC-PCF) based highly-sensitive temperature sensor was proposed and demonstrated. By selectively infiltrating the central airhole with refractive index liquid (RIL), a three-parallel-waveguide structure was formed. A dual-component interference pattern, consisting of a large spectrum envelope and fine interference fringes, was observed in the transmission spectrum. The simulation results confirmed that the interference was arising from a few hybrid supermodes in the fiber coupler structure. They were in good agreement with the experimental observation on the discrete temperature windows with different temperature sensitivities due to couplings between different hybrid supermodes in respective temperature windows. By tracing the wavelength shifts of the large spectrum envelope, high sensitivities were achieved at 42.621 nm/°C in the temperature range from 54.2 °C to 55 °C and 32.159 nm/°C from 51.8 °C to 52.6 °C.

Recent progress in optical fiber Mach–Zehnder interferometers (MZIs) has gained many achievements in sensing application. However, the strain sensitivity of optical fiber MZIs is low due to the low elasto-optical coefficient of silica. In this Letter, we propose and demonstrate a method to modulate the guided modes in an MZI based on a special hollow core microstructured optical fiber (HCMOF) by fabricating periodical deformations. Specifically, periodical deformations reduce the extinction ratio of the transmission spectrum. Furthermore, the axial tension modulates these periodical deformations, leading to the enhanced strain sensitivity in comparison to the configuration without deformations. In our experiment, the strain response from 0 to 1000 µ ε is obtained with a sensitivity of 0.00359 d B / µ ε corresponding to an improvement of 13 times compared with a sensor based on same HCMOF without deformations.

The evolution of soliton molecules emphasizes the complex soliton dynamics akin to matter molecules. Beyond the simplest soliton molecule—a soliton pair constituted by two bound pulses—soliton molecules with more constituents have more degrees of freedom because of the temporal pulse separations and relative phases. Here we detailedly characterize the transient dynamics of soliton triplets in fiber lasers by using the dispersive Fourier transform measurement. A particular form of leading, central, and tailing pulses is constructed to shed new light on more intriguing scenarios and fuel the molecular analogy. Especially the vibrating dynamics of the central and tailing pulses are captured near the regime of equally spaced soliton triplets, which is reminiscent of the recurrent timing jitters within multi-pulse structures. Further insights enable access into a universal form of unequally spaced soliton triplets interpreted as 2 + 1 soliton molecules. Different binding strengths of intramolecular and intermolecular bonds are validated with respect to the diverse internal motions involved in this soliton triplet molecule. All these findings unveil the transient dynamics with more degrees of freedom as well as highlight the possible application for all-optical bit storage.

The performance of on-chip gas sensors based on light absorption is mainly determined by the light–gas interaction. In this paper, slow-light Bloch slot waveguides (BSW) are proposed to improve sensing performance. The sensing performance is enhanced in two mechanisms. On the one hand, light is confined in the slot to increase the overlap of the mode field and the gas; on the other hand, the slow-light effect is achieved by adjusting the subwavelength grating period to increase the group index. By joint engineering the evanescent fields and group index, for a low pump power of 10 mW and a propagation loss of 3 dB/cm, the detection limit of 0.034 ppm in the near-infrared and the detection limit of 0.29 ppm in the mid-infrared at the optimum propagation length of 1.45 cm are obtained, respectively. The proposed BSW provides a promising platform for high-performance gas sensing.