Fundamental Research of High Computation-Capability Devices for Optical Tensor Convolution Operation

ZHANG Wen-jia

doi:10.12263/DZXB.20240294

您当前的位置：

首页 >

文章列表页 >

Fundamental Research of High Computation-Capability Devices for Optical Tensor Convolution Operation

PERSPECTIVE | 更新时间：2025-07-24

- Fundamental Research of High Computation-Capability Devices for Optical Tensor Convolution Operation
- ACTA ELECTRONICA SINICA Vol. 53, Issue 4, Pages: 1361-1364(2025)
- 作者机构：
  
  上海交通大学光子传输与通信全国重点实验室，上海 200240
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China(62235011)
- DOI：10.12263/DZXB.20240294
  CLC： TP381;TP183
- Received：02 April 2024，
  
  Revised：2025-05-08，
  
  Published：25 April 2025
- 稿件说明：
移动端阅览
张文甲. 高算力光学张量卷积运算芯片基础研究[J]. 电子学报, 2025, 53(04): 1361-1364.

ZHANG Wen-jia. Fundamental Research of High Computation-Capability Devices for Optical Tensor Convolution Operation[J]. Acta Electronica Sinica, 2025, 53(04): 1361-1364.
张文甲. 高算力光学张量卷积运算芯片基础研究[J]. 电子学报, 2025, 53(04): 1361-1364. DOI：10.12263/DZXB.20240294

ZHANG Wen-jia. Fundamental Research of High Computation-Capability Devices for Optical Tensor Convolution Operation[J]. Acta Electronica Sinica, 2025, 53(04): 1361-1364. DOI：10.12263/DZXB.20240294

摘要

卷积神经网络是计算机视觉和目标检测等领域应用最成功的算法之一.随着高清图像和视频等数据爆发式增长，智能处理芯片需要更强的算力和更小的功耗.光子技术的多维特征和波动物理模型为高算力张量卷积运算提供了物理基础，有望从根本上突破电芯片在提升算力和降低功耗上不可逾越的物理限制.本文介绍高算力光学张量卷积运算芯片基础研究的研究动机、主要研究挑战与解决思路及未来展望，探讨限制光学张量卷积运算应用的主要因素，推动光学张量卷积计算从基础研究走向大规模应用.

Abstract

Convolutional Neural Network is one of the most successful algorithms in the fields of computer vision and object detection. With the explosive bandwidth growth of high-definition images and videos

intelligent computing processors require higher computational capability with less power consumption. Photonic technology has inherent capability of coherent combination and multidimensional manipulation

and will become an inevitable approach to realize tensor convolution operations. This paper introduces the research motivation

primary research challenges

solution approaches

and future prospects of high computation-capability optical tensor convolutional devices. It also explores the main limiting factors restraining the application of optical tensor convolution operations

aiming to drive this promising technology from basic research to large-scale applications.

关键词

Keywords

references

MCMAHON P L . The physics of optical computing [J ] . Nature Reviews Physics , 2023 , 5 : 717 - 734 .

SHEKHAR S , BOGAERTS W , CHROSTOWSKI L , et al . Roadmapping the next generation of silicon photonics [J ] . Nature Communications , 2024 , 15 ( 1 ): 751 .

HUANG Y Y , ZHANG W J , YANG F , et al . Programmable matrix operation with reconfigurable time-wavelength plane manipulation and dispersed time delay [J ] . Optics Express , 2019 , 27 ( 15 ): 20456 - 20467 .

JIANG Y , ZHANG W J , YANG F , et al . Photonic convolution neural network based on interleaved time-wavelength modulation [J ] . Journal of Lightwave Technology , 2021 , 39 ( 14 ): 4592 - 4600 .

JIANG Y , ZHANG W J , LIU X Y , et al . Physical layer-aware digital-analog co-design for photonic convolution neural network [J ] . IEEE Journal of Selected Topics in Quantum Electronics , 2023 , 29 ( 6 ): 7400509 .

张文甲 , 姜越 , 何祖源 . 光子卷积神经网络的研究思考 [J ] . 中国计算机学会通讯 , 2022 , 18 ( 7 ): 62 - 68 .

CHEN Y T , NAZHAMAITI M , XU H , et al . All-analog photoelectronic chip for high-speed vision tasks [J ] . Nature , 2023 , 623 ( 7985 ): 48 - 57 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

No data

Related Author

No data

Related Institution

No data

⁰