Machine learning for sensing with a multimode exposed core fiber specklegram sensor

Fiber Bragg grating sensors demodulated by a speckle silicon chip

Fiber Bragg gratings (FBGs) are widely used as sensors for temperature, strain, and vibration measurement. However, current FBG demodulation methods face issues with stability, size, and cost. In this study, we proposed a silicon-on-insulator (SOI) chip to demodulate FBGs based on random speckles. A 20-mm-long coiled multimode silicon waveguide was designed to generate the speckle pattern, which was then compressed into 8 single-mode outputs. The architecture similarity between the convolutional neural network (CNN), and the proposed SOI chip was discussed. A multilayer perceptron (MLP) network was applied to regress the speckle data for prediction. The demonstrated experiments indicated that a standard deviation of 0.0414°C was achieved in the single FBG demodulation. Furthermore, we also explored the capability of demodulating multiple FBGs. This speckle-based SOI chip provides a highly stable, compact, and lightweight solution in a FBG sensing system.

High-sensitivity ocean temperature sensor using a reflective optical microfiber coupler and machine learning methods

This paper proposes a novel seawater temperature sensor, to the best of our knowledge, that utilizes an optical microfiber coupler combined with a reflective silver mirror (OMCM). The sensor's sensitivity and durability are enhanced by encapsulating it in polydimethylsiloxane (PDMS). Additionally, a specially designed metal casing prevents the OMCM from responding to pressure, thus avoiding the challenge of multi-parameter demodulation and increasing its adaptability to harsh environments. The paper analyzes the advantages of the new sensor structure and evaluates its performance in terms of temperature sensitivity and compressive strength through experiments. Finally, the paper employs machine learning demodulation methods. Compared with traditional demodulation methods, the particle swarm optimization support vector regression (PSO-SVR) algorithm demonstrates a substantial reduction in the demodulation error. Specifically, the mean absolute percentage error (MAPE) relative to the full scale drops from 2.16% to 0.157%. This paper provides an effective solution for high-precision monitoring of the ocean environmental temperature.

Deep Learning-Based Simultaneous Temperature- and Curvature-Sensitive Scatterplot Recognition

Since light propagation in a multimode fiber (MMF) exhibits visually random and complex scattering patterns due to external interference, this study numerically models temperature and curvature through the finite element method in order to understand the complex interactions between the inputs and outputs of an optical fiber under conditions of temperature and curvature interference. The systematic analysis of the fiber's refractive index and bending loss characteristics determined its critical bending radius to be 15 mm. The temperature speckle atlas is plotted to reflect varying bending radii. An optimal end-to-end residual neural network model capable of automatically extracting highly similar scattering features is proposed and validated for the purpose of identifying temperature and curvature scattering maps of MMFs. The viability of the proposed scheme is tested through numerical simulations and experiments, the results of which demonstrate the effectiveness and robustness of the optimized network model.

Optical hyperdimensional soft sensing: speckle-based touch interface and tactile sensor

Hyperdimensional computing (HDC) is an emerging computing paradigm that exploits the distributed representation of input data in a hyperdimensional space, the dimensions of which are typically between 1,000-10,000. The hyperdimensional distributed representation enables energy-efficient, low-latency, and noise-robust computations with low-precision and basic arithmetic operations. In this study, we propose optical hyperdimensional distributed representations based on laser speckles for adaptive, efficient, and low-latency optical sensor processing. In the proposed approach, sensory information is optically mapped into a hyperdimensional space with >250,000 dimensions, enabling HDC-based cognitive processing. We use this approach for the processing of a soft-touch interface and a tactile sensor and demonstrate to achieve high accuracy of touch or tactile recognition while significantly reducing training data amount and computational burdens, compared with previous machine-learning-based sensing approaches. Furthermore, we show that this approach enables adaptive recalibration to keep high accuracy even under different conditions.

Deep Learning for Optical Sensor Applications: A Review

Over the past decade, deep learning (DL) has been applied in a large number of optical sensors applications. DL algorithms can improve the accuracy and reduce the noise level in optical sensors. Optical sensors are considered as a promising technology for modern intelligent sensing platforms. These sensors are widely used in process monitoring, quality prediction, pollution, defence, security, and many other applications. However, they suffer major challenges such as the large generated datasets and low processing speeds for these data, including the high cost of these sensors. These challenges can be mitigated by integrating DL systems with optical sensor technologies. This paper presents recent studies integrating DL algorithms with optical sensor applications. This paper also highlights several directions for DL algorithms that promise a considerable impact on use for optical sensor applications. Moreover, this study provides new directions for the future development of related research.

Machine learning-assisted high-accuracy and large dynamic range thermometer in high-Q microbubble resonators

Whispering gallery mode (WGM) resonators provide an important platform for fine measurement thanks to their small size, high sensitivity, and fast response time. Nevertheless, traditional methods focus on tracking single-mode changes for measurement, and a great deal of information from other resonances is ignored and wasted. Here, we demonstrate that the proposed multimode sensing contains more Fisher information than single mode tracking and has great potential to achieve better performance. Based on a microbubble resonator, a temperature detection system has been built to systematically investigate the proposed multimode sensing method. After the multimode spectral signals are collected by the automated experimental setup, a machine learning algorithm is used to predict the unknown temperature by taking full advantage of multiple resonances. The results show the average error of 3.8 × 10^-3°C within the range from 25.00°C to 40.00°C by employing a generalized regression neural network (GRNN). In addition, we have also discussed the influence of the consumed data resource on its predicted performance, such as the amount of training data and the case of different temperate ranges between the training and test data. With high accuracy and large dynamic range, this work paves the way for WGM resonator-based intelligent optical sensing.

Spatially-resolved bending recognition based on a learning-empowered fiber specklegram sensor

Fiber specklegram sensors do not rely on complex fabrication processes and expensive sensor interrogation schemes and provide an alternative to routinely used fiber sensing technologies. Most of the reported specklegram demodulation schemes focus on correlation calculation based on statistical properties or classification according to features, resulting in limited measurement range and resolution. In this work, we propose and demonstrate a learning-empowered spatially resolved method for fiber specklegram bending sensors. This method can learn the evolution process of speckle patterns through a hybrid framework constructed by a data dimension reduction algorithm and regression neural network, which can simultaneously identify the curvature and perturbed position according to the specklegram, even for the unlearned curvature configuration. Rigorous experiments are performed to verify the feasibility and robustness of the proposed scheme, and the results show that the prediction accuracy for the perturbed position is 100%, and the average prediction errors for the curvature of the learned and unlearned configurations are 7.79 × 10^-4m^-1 and 7.02 × 10^-2m^-1, respectively. The proposed method promotes the application of fiber specklegram sensors in the practical scene and provides insights for the interrogation of sensing signals by deep learning.

Demodulation of Fabry-Perot sensors using random speckles

Random speckles are proposed to demodulate Fabry-Perot (FP) sensors in this study. A piece of multimode fiber is used to interrogate the FP transmission spectrum, and tiny spectral changes lead to significant variations in the generated speckle patterns. In the demonstration experiments, the pressure resolution of 0.001 MPa can be obtained from an open cavity FP sensor based on the convolutional neural network (CNN) demodulation algorithm. It is worth noting that the spectral differences in neighboring orders can be precisely distinguished due to the high sensitivity of speckles. Thus, the fringe-order ambiguity problem is solved and the dynamic measurement range can be greatly improved. The speckle-based demodulation scheme provides a new way to balance resolution, dynamic range, speed, and cost of FP sensors.