Compact and voltage-tunable surface plasmon polariton-based optical neural networks

作     者：Yang, Chengwang Wu, Yongli Zhong, Chengyan Wang, Xiang Li, Lingfei Guo, Junxiong Huang, Wen Liu, Yu 

作者机构：Nanjing Univ Posts & Telecommun Coll Integrated Circuit Sci & Engn Nanjing 210023 Peoples R China Nanjing Univ Posts & Telecommun Natl & Local Joint Engn Lab RF Integrat & Microass Nanjing 210023 Peoples R China Univ Elect Sci & Technol China Sch Integrated Circuit Sci & Engn State Key Lab Elect Thin Films & Integrated Device Chengdu 610054 Peoples R China Zhejiang Univ Coll Integrated Circuits Hangzhou 310027 Peoples R China Zhejiang Univ Hangzhou Global Sci & Technol Innovat Ctr Hangzhou 310027 Peoples R China Chengdu Univ Inst Adv Study Sch Elect Informat & Elect Engn Chengdu 610106 Peoples R China UESTC Chengdu Res Inst Chengdu 610207 Peoples R China Tsinghua Univ Sch Integrated Circuits Beijing 100084 Peoples R China 

出 版 物：《OPTICS LETTERS》 (Opt. Lett.)

年 卷 期：2025年第50卷第4期

页      面：1109-1112页

核心收录：

学科分类：070207[理学-光学] 07[理学] 08[工学] 0803[工学-光学工程] 0702[理学-物理学] 

基　　金：National Natural Science Foundation of China [62371095, 62201096, 62401276] China Postdoctoral Science Foundation [2024 T170097, 2024M760343] Natural Science Research Start-up Foundation of Recruiting Talents of Nanjing University of Posts and Telecommuni-cations [NY223161] Jiangsu Provincial Key Research and Development Program [BE2022126] Key Research and Development Program of Sichuan Province [2022ZHCG0041] National Key Research and Development Pro-gram of China [2022YFB3206100] Natural Science Foundation of Sichuan Province [2024NSFSC0509] 

主　　题：Neural networks Numerical simulation Optical computing Optical neural systems Optical properties Photonic integrated circuits 

摘      要：Optical neural networks (ONNs) offer advantages in parallel processing, low power consumption, and high-speed operation. However, existing ONN designs face challenges in miniaturization, stability, tunability, and integration. This study proposes a graphene surface plasmon polariton (GSPP) waveguide switch array for all-optical neural networks. The design features a compact structure with a lateral area of only 0.045 mu m2. Numerical simulations show that within the 30.2 to 49.4 THz range, the transmission rate is tunable from 0 to 0.875, accurately simulating synaptic weights in ONNs. The compact switch array achieves a recognition accuracy of 93.83% on the CIFAR-10 dataset, demonstrating its potential for high-speed, low-power, and highly integrated neural network computing platforms. (c) 2025 Optica Publishing Group. All rights, including for text and data mining (TDM), Artificial Intelligence (AI) training, and similar technologies, are reserved.