Tunable single frequency Hz-magnitude narrow linewidth Brillouin fiber laser based on parity-time symmetry.

An Hz-magnitude ultra-narrow linewidth single-frequency Brillouin fiber laser (BFL) is proposed and experimentally demonstrated. The single frequency of the laser is selected by parity-time (PT) symmetry, which consists of a stimulated Brillouin scatter (SBS) gain path excited by a 24 km single-mode fiber (SMF) and an approximately equal length loss path tuned with a variable optical attenuator (VOA). These paths are coupled through a fiber Bragg grating (FBG) into a wavelength space. Accomplishing single-frequency oscillation involves the precise adjustment of polarization control (PC) and VOA to attain the PT broken phase. In the experiment, the linewidth of the proposed BFL is 9.58 Hz. The optical signal-to-noise ratio (OSNR) reached 78.89 dB, with wavelength and power fluctuations of less than 1pm and 0.02 dB within one hour. Furthermore, the wavelength can be tuned from 1549.9321 nm to 1550.2575 nm, with a linewidth fluctuation of 1.81 Hz. The relative intensity noise (RIN) is below -74 dB/Hz. The proposed ultra-narrow single-frequency BFL offers advantages such as cost-effectiveness, ease of control, high stability and excellent output characteristics, making it highly promising for the applications in the coherent detection.

A Time-Consistency Curriculum for Learning From Instance-Dependent Noisy Labels.

Many machine learning algorithms are known to be fragile on simple instance-independent noisy labels. However, noisy labels in real-world data are more devastating since they are produced by more complicated mechanisms in an instance-dependent manner. In this paper, we target this practical challenge of Instance-Dependent Noisy Labels by jointly training (1) a model reversely engineering the noise generating mechanism, which produces an instance-dependent mapping between the clean label posterior and the observed noisy label and (2) a robust classifier that produces clean label posteriors. Compared to previous methods, the former model is novel and enables end-to-end learning of the latter directly from noisy labels. An extensive empirical study indicates that the time-consistency of data is critical to the success of training both models and motivates us to develop a curriculum selecting training data based on their dynamics on the two models' outputs over the course of training. We show that the curriculum-selected data provide both clean labels and high-quality input-output pairs for training the two models. Therefore, it leads to promising and robust classification performance even in notably challenging settings of instance-dependent noisy labels where many SoTA methods could easily fail. Extensive experimental comparisons and ablation studies further demonstrate the advantages and significance of the time-consistency curriculum in learning from instance-dependent noisy labels on multiple benchmark datasets.

Research on Simultaneous Measurement of Magnetic Field and Temperature Based on Petaloid Photonic Crystal Fiber Sensor.

In this paper, we propose and design a magnetic field and temperature sensor using a novel petaloid photonic crystal fiber filled with magnetic fluid. The PCF achieves a high birefringence of more than 1.43 × 10-2 at the wavelength of 1550 nm via the design of material parameters, air hole shape and the distribution of the photonic crystal fiber. Further, in order to significantly improve the sensitivity of the sensor, the magnetic-fluid-sensitive material is injected into the pores of the designed photonic crystal fiber. Finally, the sensor adopts a Mach-Zehnder interferometer structure combined with the ultra-high birefringence of the proposed petaloid photonic crystal fiber. Magnetic field and temperature can be simultaneously measured via observing the spectral response of the x-polarization state and y-polarization state. As indicated via simulation analysis, the sensor can realize sensitivities to magnetic fields and temperatures at -1.943 nm/mT and 0.0686 nm/°C in the x-polarization state and -1.421 nm/mT and 0.0914 nm/°C in the y-polarization state. The sensor can realize the measurement of multiple parameters including temperature and magnetic intensity and has the advantage of high sensitivity.

Scattering direction sampling methods for polarized Monte Carlo simulation of oceanic lidar.

Monte Carlo techniques have been widely applied in polarized light simulation. Based on different preconditions, there are two main types of sampling strategies for scattering direction: one is the scalar sampling method; the others are polarized sampling approaches, including the one- and two-point rejection methods. The polarized simulation of oceanic lidar involves a variety of mediums, and an efficient scattering sampling method is the basis for the coupling simulation of the atmosphere and ocean. To determine the optimal scattering sampling method for oceanic lidar simulation, we developed a polarized Monte Carlo model and simulated Mie scattering, Rayleigh scattering, and Petzold average-particle scattering experiments. This simulation model has been validated by comparison with Ramella-Roman's program [Opt. Express13, 4420 (2005)OPEXFF1094-408710.1364/OPEX.13.004420], with differences in reflectance and transmittance Stokes less than 1% in Mie scattering. The simulation results show these scattering sampling methods differ in runtime, scattering angle distributions, and reflectance and transmittance Stokes. Considering the current simulation accuracy of oceanic lidar, the differences in reflectance and transmittance Stokes are acceptable; thus, the runtime becomes the main evaluation factor. The one-point rejection method and scalar sampling method are preferable for the oceanic lidar polarized simulation. Under complex atmosphere-ocean coupling systems, scalar sampling methods may be a better choice since the calculation process of the sampling is independent of the incident Stokes vector.

A Parametrical Model for Instance-Dependent Label Noise.

In label-noise learning, estimating the transition matrix is a hot topic as the matrix plays an important role in building statistically consistent classifiers. Traditionally, the transition from clean labels to noisy labels (i.e., clean-label transition matrix (CLTM)) has been widely exploited on class-dependent label-noise (wherein all samples in a clean class share the same label transition matrix). However, the CLTM cannot handle the more common instance-dependent label-noise well (wherein the clean-to-noisy label transition matrix needs to be estimated at the instance level by considering the input quality). Motivated by the fact that classifiers mostly output Bayes optimal labels for prediction, in this paper, we study to directly model the transition from Bayes optimal labels to noisy labels (i.e., Bayes-Label Transition Matrix (BLTM)) and learn a classifier to predict Bayes optimal labels. Note that given only noisy data, it is ill-posed to estimate either the CLTM or the BLTM. But favorably, Bayes optimal labels have no uncertainty compared with the clean labels, i.e., the class posteriors of Bayes optimal labels are one-hot vectors while those of clean labels are not. This enables two advantages to estimate the BLTM, i.e., (a) a set of examples with theoretically guaranteed Bayes optimal labels can be collected out of noisy data; (b) the feasible solution space is much smaller. By exploiting the advantages, this work proposes a parametrical model for estimating the instance-dependent label-noise transition matrix by employing a deep neural network, leading to better generalization and superior classification performance.

Dual wavelength laser Doppler anemometer for simultaneous velocity and particulate size distribution measurements in submarine environments.

An in-situ laser Doppler current probe (LDCP) for the simultaneous measurements of the micro-scale subsurface current speed and the characterizations of micron particles is dedicated in this paper. The LDCP performs as an extension sensor for the state-of-the-art laser Doppler anemometry (LDA). The all-fiber LDCP utilized a compact dual wavelength (491 nm and 532 nm) diode pumped solid state laser as the light source to achieve the simultaneous measurements of the two components of the current speed. Besides its ability for the measurements of the current speed, the LDCP is also capable of obtaining the equivalent spherical size distribution of the suspended particles within small size range. The micro-scale measurement volume formed by two intersecting coherent laser beams makes it possible to accurately estimate the size distribution of the micron suspended particles with high temporal and spatial resolution. With its deployment during the field campaign at Yellow Sea, the LDCP has been experimentally demonstrated as an effective instrument to capture the micro-scale subsurface ocean current speed. The algorithm for retrieving the size distribution of the small suspended particles (2∼7.5µm) has been developed and validated. The combined LDCP system could be applied to the continuous long-term observations of plankton community structure, ocean water optical parameter over a wide range, and useful to elucidate the processes and interactions of the carbon cycles in the upper ocean.

Overcoming cohort heterogeneity for the prediction of subclinical cardiovascular disease risk.

Cardiovascular disease remains a leading cause of mortality with an estimated half a billion people affected in 2019. However, detecting signals between specific pathophysiology and coronary plaque phenotypes using complex multi-omic discovery datasets remains challenging due to the diversity of individuals and their risk factors. Given the complex cohort heterogeneity present in those with coronary artery disease (CAD), we illustrate several different methods, both knowledge-guided and data-driven approaches, for identifying subcohorts of individuals with subclinical CAD and distinct metabolomic signatures. We then demonstrate that utilizing these subcohorts can improve the prediction of subclinical CAD and can facilitate the discovery of novel biomarkers of subclinical disease. Analyses acknowledging cohort heterogeneity through identifying and utilizing these subcohorts may be able to advance our understanding of CVD and provide more effective preventative treatments to reduce the burden of this disease in individuals and in society as a whole.

Validation of the polarized Monte Carlo model of shipborne oceanic lidar returns.

The polarized Monte Carlo (PMC) model has been applied to study the backscattering measurement of oceanic lidar. This study proposes a PMC model for shipborne oceanic lidar simulation. This model is validated by the Rayleigh scattering experiment, lidar equation, and in-situ lidar LOOP (Lidar for Ocean Optics Profiler) returns [Opt. Express30, 8927 (2022)10.1364/OE.449554]. The relative errors of the simulated Rayleigh scattering results are less than 0.07%. The maximum mean relative error (MRE) of the simulated single scattering scalar signals and lidar equation results is 30.94%. The maximum MRE of simulated total scattering signals and LOOP returns in parallel and cross channels are 33.29% and 22.37%, respectively, and the maximal MRE of the depolarization ratio is 24.13%. The underwater light field of the laser beam is also simulated to illustrate the process of beam energy spreading. These results prove the validity of the model. Further analyses show that the measured signals of shipborne lidar LOOP are primarily from the particle single scatterings. This model is significant for analyzing the signal contributions from multiple scattering and single scattering.

Coherent high-spectral-resolution lidar for the measurement of the atmospheric Mie-Rayleigh-Brillouin backscatter spectrum.

In this study, a 1550 nm coherent high-spectral-resolution lidar (CHSRL) is developed to measure the optical properties of aerosols and atmospheric wind profiles in the atmospheric boundary layer. To determine the optical properties, a coherent frequency discriminator based on the fast Fourier transform is designed in the CHSRL to separate the Mie and the Rayleigh-Brillouin backscatter spectra to fulfill the needs of high-spectral measurements. The atmospheric wind velocity is retrieved using the simultaneously measured Doppler shift. This non-optical frequency discriminator is a feasible and low-cost solution compared to a narrow-bandwidth optical filter, such as a Fabry-Perot interferometer or an atomic filter. However, shot, amplifier spontaneous emission, and detector noise interfere with the Rayleigh-Brillouin spectrum. Therefore, a spectrum correction algorithm is proposed to recover the interfered Rayleigh-Brillouin spectrum, and the measurement results of the spectral line agree well with those modeled with Tenti S6 at different central frequencies. Finally, field observations for comparison are conducted with the co-located CHSRL, Raman lidar, and coherent Doppler wind lidar. The comparison results indicate that the correlation coefficient of the aerosol backscatter coefficient is 0.84. The correlation coefficient and standard deviation of wind velocity are 0.98 and 0.13 m · s-1, respectively.

Shipborne variable-FOV, dual-wavelength, polarized ocean lidar: design and measurements in the Western Pacific.

For the requirement of high-precision vertical profile of the polarization and optical properties of natural seawater, a ship-borne variable-FOV, dual-wavelength, polarized ocean lidar system is designed to obtain the volume linear depolarization ratio (VDR), color ratio and optical parameter profiles of seawater. With the high signal-to-noise ratio, which benefits from the high power (355 nm with 120 mJ, 532 nm with 200 mJ) solid-state laser and a photon counting recorder with a sampling rate of 1 GHz, the attenuated backscattered signal of seawater in the western Pacific campaign reaches to the depth of 50 m, where a plankton layer presents. The receiver of lidar is capable of switching to wide and narrow field of view (FOV), respectively, to obtain the lidar attenuation coefficient Klidar, which is in good agreement with the beam attenuation coefficient of seawater c with a narrow FOV and diffuse attenuation coefficient Kd with a wide FOV. Besides, the Klidar, and the VDR, at two wavelengths of 355 nm and 532 nm are compared to explore the possibility of multi-wavelength of laser application in the ocean lidar. The VDR and the color ratio profiles have a desirable correlation with the in-situ measurement of chlorophyll a (Chla) and chromophoric dissolved organic matter (CDOM) profiles, respectively. With the combination of the Klidar, the VDR and the color ratio profiles, measured in different regions and time periods during the campaign, the multi-wavelength and polarization lidar shows its potential to explore various ocean compositions, such as the ocean particles size shape, the species and vertical migration characteristics of planktons, and the profile distribution of the ocean compositions.

Bridging the Gap Between Few-Shot and Many-Shot Learning via Distribution Calibration.

A major gap between few-shot and many-shot learning is the data distribution empirically oserved by the model during training. In few-shot learning, the learned model can easily become over-fitted based on the biased distribution formed by only a few training examples, while the ground-truth data distribution is more accurately uncovered in many-shot learning to learn a well-generalized model. In this paper, we propose to calibrate the distribution of these few-sample classes to be more unbiased to alleviate such an over-fitting problem. The distribution calibration is achieved by transferring statistics from the classes with sufficient examples to those few-sample classes. After calibration, an adequate number of examples can be sampled from the calibrated distribution to expand the inputs to the classifier. Specifically, we assume every dimension in the feature representation from the same class follows a Gaussian distribution so that the mean and the variance of the distribution can borrow from that of similar classes whose statistics are better estimated with an adequate number of samples. Extensive experiments on three datasets, miniImageNet, tieredImageNet, and CUB, show that a simple linear classifier trained using the features sampled from our calibrated distribution can outperform the state-of-the-art accuracy by a large margin. Besides the favorable performance, the proposed method also exhibits high flexibility by showing consistent accuracy improvement when it is built on top of any off-the-shelf pretrained feature extractors and classification models without extra learnable parameters. The visualization of these generated features demonstrates that our calibrated distribution is an accurate estimation thus the generalization ability gain is convincing. We also establish a generalization error bound for the proposed distribution-calibration-based few-shot learning, which consists of the distribution assumption error, the distribution approximation error, and the estimation error. This generalization error bound theoretically justifies the effectiveness of the proposed method.

Temperature self-calibrated pH sensor based on GO/PVA-coated MZI cascading FBG.

A temperature self-calibrated potential of hydrogen (pH) sensor based on the single mode fiber-tapered dual core photonic crystal fiber-single mode fiber (SMF-TDCPCF-SMF) structure cascaded with a fiber Bragg grating (FBG) is proposed and demonstrated. The TDCPCF structure formed Mach-Zehnder interferometer (MZI) is modified with a coating of graphene oxide/polyvinyl alcohol (GO/PVA) hybrid hydrogel to realize the measurement of pH, and the uncoated FBG is used to calibrate temperature. In our experiment, the sensitivity coefficient of 0.69 nm/pH with R2=0.99 and the hysteresis loss of less than 0.007 are achieved within the pH range from pH 4.00 to pH 9.85. The measured response time from pH 7.00 to pH 4.00, 6.00 and 9.85 are no higher than 10s. Moreover, the resonant wavelengths of MZI and FBG also exhibit good linear relationship with the temperature sensitivity coefficient of 0.15 nm/°C (R2=0.99) and 0.09 nm/°C (R2=0.97) respectively. It is demonstrated successfully that the proposed sensor has broad application prospects in the field of environmental monitoring, biological sensing and chemical analysis, due to the good performance of the temperature self-calibrated pH monitoring, repeatability, linearity, response time and reversibility.

Active and passive optical remote sensing of the aquatic environment: introduction to the feature issue.

Through decades of efforts and practices, we have achieved great progress in understanding ocean biology and biogeochemistry through satellite measurements of ocean (water) color, or passive remote sensing. These include detailed global maps of the distribution of surface phytoplankton, the production of newly formed particulate organic matter through photosynthesis (i.e., primary production), as well as the change and feedback of phytoplankton in a changing climate, to name a few. However, these results are still far from a full account of ocean biology and biogeochemistry, where we want more detailed information of phytoplankton (e.g., types and sizes), as well as information in the vertical dimension. For such, we are happy to see new developments in ocean optics and ocean color remote sensing. These include, but certainly are not limited to, hyperspectral sensors, measurements via polarized setups, as well as ocean lidar systems. In particular, through pumping laser light into deeper ocean, lidar has demonstrated great potential to fill the gap of passive ocean color remote sensing. These developments in technology are providing exciting new findings where breakthroughs in ocean biogeochemistry are on the horizon. Thus, we organized this feature issue in Applied Optics to summarize a few recent developments and achievements, where readers and the community can easily capture progress on both fronts, as well as the potential and advantages of the fusion of passive and active optical sensing. Specifically, this issue contains 12 papers describing research in both active and passive optical remote sensing of aquatic environment. They are still limited in number and subject, but are expected to stimulate the ocean color community with findings relevant for satellite applications.

Self-assisted membrane-penetrating helical polypeptides mediate anti-inflammatory RNAi against myocardial ischemic reperfusion (IR) injury.

Anti-inflammatory RNA interference (RNAi) provides a promising paradigm for the treatment of myocardial ischemia reperfusion (IR) injury. To overcome the membrane barriers against intracardial siRNA delivery, various guanidinated helical polypeptides with potent and aromaticity-assisted membrane activities were herein developed and used for the delivery of siRNA against RAGE (siRAGE), a critical regulator of the pro-inflammatory cascade. Aromatic modification of the polypeptide led to notably enhanced trans-membrane siRNA delivery efficiencies, and more importantly, allowed more siRNA cargoes to get internalized via non-endocytosis, an effective pathway toward gene transfection. Subsequently, benzyl-modified polypeptide (P-Ben) was identified as the top-performing material with the highest RAGE silencing efficiency yet lowest cytotoxicity in H9C2 cells. Intracardial injection of the P-Ben/siRAGE polyplexes at 150 µg siRNA per kg led to remarkable RAGE knockdown by ∼85%, thereby attenuating the inflammatory cytokine release and reducing the cardiomyocyte apoptosis as well as myocardium fibrosis to recover the cardiac function after IR injury. This study therefore provides an effective strategy for the design of membrane-penetrating gene delivery materials, and may provide a promising addition to the anti-inflammatory treatment of myocardial IR injury.

Aircraft wake vortex and turbulence measurement under near-ground effect using coherent Doppler lidar.

This paper evaluates the wake vortex characteristics using pulsed coherent Doppler lidar (PCDL) under near-ground effect (NGE). A wake vortex visualization demonstrator (V2D) is developed in order to visualize wake vortex in real-time. The combination of radial velocity distribution and FFT spectrum characterization are used to identify the core position of wake vortex. The velocity envelope and Burnham-Hallock model correction are used to retrieve the circulation of wake vortex under NGE. The circulation error, which is caused by PCDL scanning mode, is simulated and corrected. To investigate the dissipation rate's effect on wake vortex in real atmosphere, the cross wind and atmospheric turbulence are concurrently retrieved from the same measurement of wake vortex by using structure function. The statistics of wake vortex parameters are analyzed, based on the measurement campaign at Beijing Capital International Airport (BCIA) in 2017.

High-sensitivity refractive index and temperature sensor based on cascaded dual-wavelength fiber laser and SNHNS interferometer.

A novel differential intensity-measurement high-sensitivity refractive index (RI) sensor based on cascaded dual-wavelength fiber laser and single-mode-no-core-hollow-core-no-core-single-mode (SNHNS) structure is proposed and demonstrated. The sensing unit consists of one uniform fiber Bragg grating (FBG) and an SNHNS structure as all-fiber interferometer filter. The dual-wavelength fiber laser has a ring cavity composed of two FBGs with central wavelengths of 1550.10nm and 1553.61nm. Through monitoring the wavelength shift and the output power difference of the dual-wavelength fiber laser, the simultaneous measurement for RI and temperature is realized. In our experiment, the proposed fiber laser sensor exhibits high RI sensitivities of -193.1dB/RIU and 174.8dB/RIU in the range of 1.334-1.384. The relative variation of output power at the two FBG wavelengths shows a higher RI sensitivity of -367.9dB/RIU with better stability, which is greater than the traditional modal interferometer structure. Meanwhile, the temperature sensitivity of the proposed sensor is 8.53 × 10-3nm/°C, and the changes of laser output power caused by temperature are -0.223dB/°C and 0.215dB/°C.

Fiber ring laser based on MMF-PMFBG-MMF filter for three parameters sensing.

A dual-wavelength fiber ring laser based on multimode fiber-polarization maintaining fiber Bragg grating-multimode fiber (MMF-PMFBG-MMF) filter for simultaneously axial strain, temperature and refractive index (RI) sensing is proposed and experimentally demonstrated. In the ring laser, stable dual-wavelength lasing is determined by the MMF-PMFBG-MMF filter with two polarization states. The fiber birefringence affected by axial strain is far less than the effect of the temperature. Through monitoring the variations of each wavelength shift and output power, the simultaneous measurement for the axial strain, temperature and RI is realized. In our experiment, the proposed fiber laser sensor exhibits an axial strain sensitivity of 1.16 × 10-3nm/µÎµ and an RI sensitivity of 81.2dB/RIU. Meanwhile, the temperature sensitivities of two wavelengths are experimentally measured to be 9.74 × 10-3nm/°C and 9.2 × 10-3nm /°C, respectively.

Doppler lidar investigation of wind turbine wake characteristics and atmospheric turbulence under different surface roughness.

Four field experiments based on Pulsed Coherent Doppler Lidar with different surface roughness have been carried out in 2013-2015 to study the turbulent wind field in the vicinity of operating wind turbine in the onshore and offshore wind parks. The turbulence characteristics in ambient atmosphere and wake area was analyzed using transverse structure function based on Plane Position Indicator scanning mode. An automatic wake processing procedure was developed to determine the wake velocity deficit by considering the effect of ambient velocity disturbance and wake meandering with the mean wind direction. It is found that the turbine wake obviously enhances the atmospheric turbulence mixing, and the difference in the correlation of turbulence parameters under different surface roughness is significant. The dependence of wake parameters including the wake velocity deficit and wake length on wind velocity and turbulence intensity are analyzed and compared with other studies, which validates the empirical model and simulation of a turbine wake for various atmosphere conditions.

High sensitivity axial strain and temperature sensor based on dual-frequency optoelectronic oscillator using PMFBG Fabry-Perot filter.

A dual-frequency optoelectronic oscillator (OEO) incorporating a polarization-maintaining fiber Bragg grating (PMFBG) Fabry-Perot filter for high-sensitivity and high-speed axial strain and temperature sensing is proposed and experimentally demonstrated. In the OEO loop, two oscillation frequencies are determined by a PMFBG Fabry-Perot filter with two ultra-narrow notches and two laser sources which operate as a dual-passband microwave photonic filter. The fiber birefringence affected by axial strain is far less than the temperature. Through monitoring the variations of two oscillating frequencies and beat frequency, the simultaneous measurement for the axial strain and temperature is realized. The sensitivities of the proposed OEO sensor for axial strain and temperature are experimentally measured to be as high as 100.6 or 100.5 MHz/µÎµ and -41 MHz/°C, respectively.

Wind turbine wake visualization and characteristics analysis by Doppler lidar.

Wind power generation is growing fast as one of the most promising renewable energy sources that can serve as an alternative to fossil fuel-generated electricity. When the wind turbine generator (WTG) extracts power from the wind, the wake evolves and leads to a considerable reduction in the efficiency of the actual power generation. Furthermore, the wake effect can lead to the increase of turbulence induced fatigue loads that reduce the life time of WTGs. In this work, a pulsed coherent Doppler lidar (PCDL) has been developed and deployed to visualize wind turbine wakes and to characterize the geometry and dynamics of wakes. As compared with the commercial off-the-shelf coherent lidars, the PCDL in this work has higher updating rate of 4 Hz and variable physical spatial resolution from 15 to 60 m, which improves its capability to observation the instantaneous turbulent wind field. The wind speed estimation method from the arc scan technique was evaluated in comparison with wind mast measurements. Field experiments were performed to study the turbulent wind field in the vicinity of operating WTGs in the onshore and offshore wind parks from 2013 to 2015. Techniques based on a single and a dual Doppler lidar were employed for elucidating main features of turbine wakes, including wind velocity deficit, wake dimension, velocity profile, 2D wind vector with resolution of 10 m, turbulence dissipation rate and turbulence intensity under different conditions of surface roughness. The paper shows that the PCDL is a practical tool for wind energy research and will provide a significant basis for wind farm site selection, design and optimization.