Photonic chips have great potential for neural network computing due to their fast speed, low power consumption, and parallelism. We propose a quantized neural network modeling method based on microring resonators (MRR). We analyze the optical properties of the MRRs and utilize lasers with different wavelengths as inputs of the neural network. The quantization aware method is adopted to train the neural network, and the stochastic search method is utilized to determine hyperparameters of the network. We transform the network parameters and hyperparameters into MRR parameters to simulate neural network matrix multiplication operations. Finally, we used the Mixed National Institute of Standards and Technology database for testing the proposed model. For 4-, 5-, and 6-bit quantization of weight parameters, we obtain classification accuracies of 94.23%, 94.73%, and 96.11%, respectively. Thus our study demonstrates the feasibility of building a neural network inference system using a microring structure and provides a theoretical support for applying MRRs in neural networks.
Infrared detectors have the advantages of passive detection, strong anti-jamming ability and easy to carry. Narrow bandgap antimonide semiconductor is recognized as the preferred material system for the third generation of focal plane array infrared detector. Since 2000, InAs/GaSb type II superlattice(T2SL) have a great progress in preparation of molecular beam epitaxy and detector technology. The defect level associated with Ga near the center of the gap band increase the recombination rate, which will lead to the fatal problem of short minority carrier lifetime. The InAs/InAsSb (Ga-free) T2SL has emerged as a promising candidate to improve the minority carrier lifetime and inhibition the SRH recombination, and these photodetectors can offer a cut-off wavelength ranging from 4 to 15 μm. However, the superior performance of T2SLs detectors are not realized, and the dark current has been proved as an important limited factor. In order to suppress the generation-recombination current and other dark currents, kinds of structures have been applied to T2SL infrared detectors. In this brief review paper, we overview the suppression methods of the dark current which is the important factor that affect the performance of detectors.
To achieve low-power convolutional neural networks, we develop a photoelectric hybrid neural network (PHNN), which consists of the optical interference unit (OIU) and field-programmable gate array (FPGA). The OIU composed of Mach–Zehnder interferometers (MZI) arrays, used as convolution kernels, performs multiplication and accumulation operations. The convolution kernel is split and reorganized, forming a new unitary matrix, which reduces MZI quantity. FPGA realizes nonlinear calculation, data scheduling and storage, and phase encoding and modulation. Our PHNN has an accuracy rate of 88.79%, and the energy efficiency ratio is 1.73 times that of traditional electronic products.
A planar lightwave circuit-based multi-channel arrayed waveguide grating (AWG) is used as a critical part of the fiber Bragg grating (FBG) interrogation unit and designed to be integrated into the fiber grating sensing system. We designed and fabricated a high-performance, 40 channels AWG in the device, enabling whole C-band wavelength demodulation while maintaining high-measurement resolution. The AWG was designed to have 100-GHz channel spacing; the channel crosstalk was measured to be −45 dB. An AWG-based FBG interrogation system was built to analyze the characteristics of our unit. The testing result shows that our system has the capability of demodulating wavelength from 1526.438 to 1563.863 nm with wavelength accuracy within ±10 pm; the wavelength resolution is measured to be 1 pm. Our testing result confirms that an AWG-based FBG interrogator is an excellent choice for FBG sensing systems.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.