For the development of minimized and high-rate photonic-integrated fiber Bragg grating interrogation (FBGI) systems, arrayed waveguide grating (AWG) has been widely used as one of the critical components. In this paper, we present an 8-channel SOI-based AWG for a photonic integrated FBG interrogation microsystem. The channel spacing of the AWG is designed to be 3 nm to meet a high-dynamic-range demodulation requirement. The core size of the fabricated AWG is about 335 × 335 μm². The simulation results and experimental results are in high agreement, showing that AWG has a fine transmission spectrum with crosstalk below −16 dB, nonuniformity below 0.4 dB, insertion loss below −6.35 dB, 3 dB bandwidth about 1.3 nm and 10 dB bandwidth of 2.3 nm. The proposed AWG can be applied perfectly to the SOI-based AWG demodulation microsystem, exhibiting a large dynamic range of 1.2 nm, the resolution for measurements is 1.27 pm and a high accuracy of 20.6 pm.

A fiber Bragg grating (FBG) interrogator is a scientific instrument that converts the wavelength change of FBG sensors into readable electrical signals. To achieve miniaturization and integration of FBG interrogator, we designed and fabricated a 36-channel array waveguide grating (AWG) on silica-based planar lightwave circuits (PLC) as a key device in a built FBG interrogation system. It is used to achieve continuous demodulation in C-band, while maintaining high resolution. This AWG has a 1.6 nm channel spacing, 3-dB bandwidth of 1.76 nm, non-adjacent channel crosstalk of −29.76 dB, and insertion loss of 3.46 dB. The dynamic range of the FBG interrogation system we built was tested to be 1522.4–1578.4 nm, with an interrogation resolution of 1 pm and accuracy of less than 1 pm in the dynamic range of 1523.16–1523.2 nm. The test results show that the FBG interrogation technology, based on AWG, can realize FBG wavelengths accurately demodulated, which has high application value in aerospace, deep sea exploration, and environmental monitoring, as well as other fields.

The convolution neural network (CNN) is a classical neural network with advantages in image processing. The use of multiport optical interferometric linear structures in neural networks has recently attracted a great deal of attention. Here, we use three 3 × 3 reconfigurable optical processors, based on Mach-Zehnder interferometers (MZIs), to implement a two-layer CNN. To circumvent the random phase errors originating from the fabrication process, MZIs are calibrated before the classification experiment. The MNIST datasets and Fashion-MNIST datasets are used to verify the classification accuracy. The optical processor achieves 86.9% accuracy on the MNIST datasets and 79.3% accuracy on the Fashion-MNIST datasets. Experiments show that we can improve the classification accuracy by reducing phase errors of MZIs and photodetector (PD) noises. In the future, our work provides a way to embed the optical processor in CNN to compute matrix multiplication.

Fiber Bragg grating (FBG) interrogator is a scientific instrument that can indirectly monitor physical quantities such as temperature and pressure by detecting the center wavelength of FBG. The arrayed waveguide grating (AWG), one of the fundamental parts of the FBG interrogation system, is essential for driving the downsizing of the system and high-speed rate. In this paper, a 12-channel silicon on insulator based AWG with an average 3 dB bandwidth of 2.23 nm , insertion loss of less than 5 dB, and crosstalk of −16.64 dB is designed and fabricated for a photonic integrated FBG interrogation system. The wavelength resolution of the FBG interrogation system of this AWG is 8 pm , and the wavelength accuracy measured in the dynamic range of 0.8 nm is 10.59 pm . This AWG can be well used in FBG interrogation systems.

The arrayed waveguide grating (AWG) is an optical device that can be used to interrogate optical signals reflected from fiber Bragg gratings (FBGs). However, achieving high-precision interrogation over a large dynamic range is hard for conventionally manufactured AWGs due to limitations in their spectral characteristics. In this paper, the structure of AWG is optimized to increase the bandwidth by incorporating a multimode interference (MMI) coupler to increase the dynamic range of interrogation. A cascaded neural network consisting of a Transformer, convolutional neural network (CNN), and backpropagation neural network (BPNN) is proposed to significantly improve interrogation precision. The experimental results demonstrate that our proposed interrogation system can continuously demodulate in the C-band, improving interrogation precision from an average of 37 pm to better than 1 pm. This research offers a solution for achieving high-performance wavelength interrogation.