Abstract

We present a global optical power allocation architecture, which can enhance the calculation accuracy of the integrated photonic tensor flow processor (PTFP). By adjusting the optical power splitting ratio according to the weight value and loss of each calculating unit, this architecture can efficiently use optical power so that the signal-to-noise ratio of the PTFP is enhanced. In the case of considering the on-chip optical delay line and spectral loss, the calculation accuracy measured in the experiment is enhanced by more than 1 bit compared with the fixed optical power allocation architecture.

© 2022 Optica Publishing Group

Optical processor for a binarized neural network

Long Huang and Jianping Yao
Opt. Lett. 47(15) 3892-3895 (2022)

Nonlinear optical response of inverse-designed integrated photonic devices

Hao Chen, Ziming Zhang, Guoqing Wang, Zhenyuan Shang, Jiatong Li, Zexing Zhao, Mengyu Zhang, Kai Guo, Junbo Yang, and Peiguang Yan
Opt. Lett. 47(5) 1254-1257 (2022)

Architecture for integrated RF photonic downconversion of electronic signals

Nathan P. O’Malley, Keith A. McKinzie, Mohammed S. Alshaykh, Junqiu Liu, Daniel E. Leaird, Tobias J. Kippenberg, Jason D. McKinney, and Andrew M. Weiner
Opt. Lett. 48(1) 159-162 (2023)

Synaptic delay plasticity based on frequency-switched VCSELs for optical delay-weight spiking neural networks

Yao Lu, Wenjia Zhang, Bangqi Fu, Jiangbing Du, and Zuyuan He
Opt. Lett. 47(21) 5587-5590 (2022)

Experimental demonstration of a photonic convolutional accelerator based on a monolithically integrated multi-wavelength distributed feedback laser

Zhenzhen Xu, Kaifei Tang, Xiang Ji, Zhenxing Sun, Yaguang Wang, Ziming Hong, Pan Dai, Rulei Xiao, Yuechun Shi, and Xiangfei Chen
Opt. Lett. 47(22) 5977-5980 (2022)