Shanghai Jiao Tong University

Shaofu Xu

Weiwen Zou

Jianping Chen

Jing Wang

Lei Yu

Bowen Ma

Yuting Chen

Shaoyang Zhang

Sicheng Yi

A high-precision analog optical computing method and system based on bit slicing comprises decomposing high-precision signals into low-precision signal segments, decomposing high-precision weights into multiple low-precision weight segments, performing low-precision signal and weight multiplication operations in the optical domain respectively, and finally combining these low-precision calculation results to restore high-precision output. By splitting high-precision data into multiple low-precision bit slices for processing, it takes advantage of high-speed parallel characteristics of photonic computing to simultaneously handle multiple input bit slices and weight bit slices, improves computational efficiency of large-scale computing tasks and reduces the hardware requirements for optical computing. By adjusting values of M and N, it flexibly adapts to computing tasks of different scales and precisions, demonstrating strong adaptability and flexibility. It achieves balance between computational accuracy and hardware complexity, and fully exploits parallel processing advantages of photons, providing solutions for large-scale data processing and efficient optical domain computing.

High precision analog optical computing method and system based on bit slice

A high-speed and low-voltage electro-optical modulator based on a lithium niobate-silicon wafer. A silicon wafer is located above a lithium niobate wafer; a lithium niobate-silicon hybrid waveguide is formed by etching a silicon waveguide; and the power of light waves is differently distributed in the lithium niobate-silicon hybrid waveguide by changing the structure of the silicon waveguide. When higher power is distributed in the silicon waveguide, the high-speed and low-voltage electro-optical modulator is suitable for realizing a compact wave splitting function, a wave combining function and a thermo-optical modulation function; and when higher power is distributed in the lithium niobate waveguide, the high-speed and low-voltage electro-optical modulator is suitable for realizing a high-speed and low-voltage electro-optical modulation function. The present invention takes advantage of the lithium niobate and silicon material platforms respectively, and is suitable for high-speed and low-voltage electro-optical modulation.

High-speed and low-voltage electro-optical modulator based on lithium niobate-silicon wafer

A monolithically integrated optical analog-to-digital conversion system based on a lithium niobate-silicon wafer, and a method for manufacturing the same, wherein a novel wafer (lithium niobate-silicon wafer) is used to implement the monolithically integrated optical analog-to-digital conversion system having multiple photonic devices, including an electro-optical modulator array, a tunable delay line array, an electronic circuit, and the like. As a result, multiple devices are manufactured on one chip, and the performance advantages and the stability of the system are guaranteed. Moreover, the present invention provides a CMOS-compatible method for manufacturing the system, so that the monolithically integrated optical analog-to-digital conversion system based on the lithium niobate-silicon wafer can be implemented on platforms of most chip manufacturers.

Monolithically integrated optical analog-to-digital conversion system based on lithium niobate-silicon wafer and method for manufacturing the same

Method and system for intelligent decision-making photonic signal processing, where the system comprises a multi-functional input unit, an electro-optical conversion module, a signal processing module, a photoelectric conversion module, a multi-functional output unit, and an artificial intelligence chip. The invention combines the advantages of photonic high-speed, wide-band, and electronic flexibility, combined with heterogeneous photoelectron hybrid integration, packaging and other processes, along with deep learning algorithm, is an intelligent electronic information system that may simultaneously realize digital and analog signal processing.

Method and system for intelligent decision-making photonic signal processing

Integration method of a large-scale silicon-based lithium niobate film electro-optic modulator array. By using the method, the difficulty of a fabrication process of a lithium niobate crystal layer is reduced, requirements on precision of bonding lithium niobate and silicon is reduced, and fabrication and bonding of the large-scale array lithium niobate crystal layer can be completed at one time, so that production efficiency of the silicon-based lithium niobate film electro-optic modulator array is greatly improved; through design and optimization of the structure of the silicon crystal layers, light can be naturally alternated and mutually transmitted in silicon waveguides and lithium niobate waveguides, and a high-performance electro-optic modulation effect of the lithium niobate film is achieved.

Silicon-based lithium niobate film electro-optic modulator array and integration method thereof

A device for noise suppression and distortion correction of analog-to-digital converters based on deep learning that realizes effect of correcting noise and distortion of analog to digital converters. The method is applied to electronic ADCs or photonic ADCs. It utilizes the learning ability of the deep network to perform system response learning on ADCs which need noise suppression and distortion correction, establishes a computational model in the deep network that can suppress the reconstruction of noises and distorted signals, performs noise suppression and distortion correction on the signals obtained by ADCs, and thereby improves performance of the learned ADCs. The device improves the performance of the microwave photon system with high sampling precision of microwave photon radar and optical communication system.

Deep learning based method and device for noise suppression and distortion correction of analog-to-digital converters

A high-speed and high-precision photonic analog-to-digital conversion device capable of realizing intelligent signal processing. Learning ability of deep learning technology is utilized to learn the nonlinear response and channel mismatch effect of the system and configure optimal parameters of the deep network. Deterioration of photonic analog-to-digital conversion system performance caused by nonlinear distortion and channel mismatch distortion is eliminated in real time, and performance indicators thereof are improved. By using the induction and deduction ability of deep learning technology, intelligent signal processing of the input signal is realized, and users are provided with digital signals that meet the requirements. It's important for improving the performance of microwave photonic systems that require high sampling rate, high time precision, and high sampling accuracy, such as microwave photonic radar and optical communication systems, and also critical to improve the signal processing ability of such systems under complex conditions.

Growing community of inventors

Shanghai, China

Shaofu Xu