Abstract:Optical convolution computing is gaining traction owing to its inherent parallelism, multi-dimensional processing, and energy efficiency. To handle input dimensions of N, conventional implementations necessitate N2 optical elements, such as Mach-Zehnder interferometers or micro-ring resonators, to process multiply-accumulate (MAC) operations, limiting scalability and resulting in elevated power consumption. Here, a direct convolution computing method based on wavelength routing, utilizing the unique sliding property of an arrayed waveguide grating router (AWGR) to perform the sliding window operation of the convolution in the wavelength-space domains is proposed. With two input vectors directly loaded onto two modulator arrays, the convolution result is instantaneously produced at a photodetector array. The entire convolution computation is executed within a single clock cycle without the need for preprocessing or decomposition into elementary MAC operations. The number of active elements is minimal, only needed for input/output. The proposed optical convolution unit has striking advantages of high scalability, high speed, and processing simplicity compared to those based on optical matrix-vector multipliers. In the first experimental demonstration, a remarkable classification accuracy of up to 98.2% in handwritten digit recognition tasks using a LeNet-5 neural network is achieved. A direct convolution computing method based on wavelength routing, utilizing the unique sliding property of an arrayed waveguide grating router (AWGR) to perform the sliding window operation of the convolution in the wavelength-space domains is proposed and demonstrated. With two inputs vectors directly loaded onto modulator arrays, the convolution result is instantaneously produced at a photodetector array. image

Microbinary Element for Optical Wavelet Transform

Texture Information Processing System With Binary Optical Wavelet Element

Design and Application of Binary Hybrid Optical System Used in the Visible Spectra

Micro-optical Multiwavelet Element for Hybrid Texture Segmentation Processor

Spherical Lens with Binary Optical Element for Visible and UV Dual-Band Optical System

Binary photon sieve

Design and Fabrication of Infrared Hybrid Refractive/diffractive Optical System for 3-5Μm Band

A Bio-Inspired Optical System with a Polymer Membrane and Integrated Structure

Application and Research of Harmonic Diffractive/Refractive Optics in Visible Spectrum

Binary optical elements used for agile beam transformation and their application

Using Optical Separable Wavelet Transform to Construct Oriented Optical Filter

Ultra-High Dynamic Extinction Ratio Electro-Optical Modulator with Coupled Microring on SOI Platform

Microoptic Element for Optical Disk Memory Read-Write Heads

Direct Optical Convolution Computing Based on Arrayed Waveguide Grating Router

Tunable, Ultra-Narrow-Band Optical Filter Based on A Whispering Gallery Mode Hybrid-Microsphere

A Binary Beam-Splitter For Lidar Beam-Shaping In Laser Imaging Application - Art. No. 66241j

Developed, Binary, Image Processing in a Dual-Channel, Optical, Real-Time Morphological Processor

Binary Optical Elements Generating Special Focused Pattern

Some Studies on Application of Binary Optical Elements

Binary Optics Used in High Power Laser Shaping

Dual Multichannel Optical Wavelet Transform Processor