On-chip integrated all-optical neural network optical computing chip

Abstract

The invention discloses an on-chip integrated all-optical neural network optical computing chip, and the chip comprises a multi-amplitude pulse light source module which comprises at least two lasers and is used for generating a multi-amplitude optical computing input pulse signal; the cross gain modulation module comprises a Mach-Zehnder interferometer SOA-MZI module based on a semiconductor optical amplifier and a semiconductor optical amplifier SOA module, and is used for carrying out wavelength conversion on a multi-amplitude optical calculation input pulse signal according to a nonlinear activation Sigmoid function based on cross gain modulation of the SOA-MZI module and the SOA module; a pulse optical signal with a balanced amplitude is output; the threshold value adjusting module is used for adjusting the saturation state of the SOA in the SOA-MZI module and the SOA module so as to adjust the threshold value of the Sigmoid function; and the photonic integrated chip substrate is used as a chip substrate of the all-optical neural network optical computing chip. The all-optical monolithic integration technology is adopted, and high-speed, low-power-consumption, low-cost and high-integration-level neural network optical calculation of a chip-level all-optical structure is achieved.

Description

technical field [0001] The invention relates to the technical field of semiconductor optoelectronics, in particular to an on-chip integrated all-optical neural network optical computing chip. Background technique [0002] With the rapid development of today's information science, from image processing, speech recognition to artificial intelligence technology and deep learning applications, many important computing tasks are inseparable from the software implementation of brain-like neural network computing. Neural network computing is a computing method analogous to human brain neurons. It is composed of many neurons with adjustable connection weights. It has large-scale parallel processing, distributed information storage, and good self-organizing and self-learning capabilities. features. [0003] Traditional neural network computing is an electronic neural network based on an integrated circuit (IC) chip. It is an "electronic + logic" information processing model based on...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present invention provides an on-chip integrated all-optical neural network optical computing chip, which solves the technical problems of low computing speed, high energy consumption, high cost and difficult high-density integration of traditional electronic neural networks.

Images