Design and experimental verification for optical module of optical vector–matrix multiplier

Abstract

Optical computing is a new method to implement signal processing functions. The multiplication between a vector and a matrix is an important arithmetic algorithm in the signal processing domain. The optical vector-matrix multiplier (OVMM) is an optoelectronic system to carry out this operation, which consists of an electronic module and an optical module. In this paper, we propose an optical module for OVMM. To eliminate the cross talk and make full use of the optical elements, an elaborately designed structure that involves spherical lenses and cylindrical lenses is utilized in this optical system. The optical design software package ZEMAX is used to optimize the parameters and simulate the whole system. Finally, experimental data is obtained through experiments to evaluate the overall performance of the system. The results of both simulation and experiment indicate that the system constructed can implement the multiplication between a matrix with dimensions of 16 by 16 and a vector with a dimension of 16 successfully.

Photonic matrix multiplication lights up photonic accelerator and beyond

“…The first kind of optical MVM (PLC-MVM) is implemented by the diffraction of light in free space. Figure 1a shows a typical MVM configuration 34 , 35 . First, the incident vector of X distributed along the x direction can be expanded and replicated along the y direction through a cylindrical lens or other optical elements.…”

The challenges of modern computing and new opportunities for optics

“…Moreover, benefiting from the quickly developed liquid-crystal-on-silicon (LCoS) technology and driving from the display industry, the resolution of SLM or DMD becomes fairly large (4 K resolution is commercially available). But the crosstalk error is the main obstacle to demonstrate the utmost performance of VMM employing high resolution SLM or DMD [36]. Though the crosstalk issue could be circumvented by enlarging the pixel size of SLM or DMD, the functional area of SLM or DMD restricts Here, V, Σ and U represent a unitary matrix, a diagonal matrix and a unitary matrix, respectively.…”

Computation at the speed of light: metamaterials for all-optical calculations and neural networks

“…Metasurfaces in particular have been proven to perform all-optical matrix calculations through the ability to manipulate the properties of light at defined spatial locations through the correct design of subwavelength-sized meta-atoms 101 – 103 This functionality is extremely important in the field of machine learning, where models are trained through a combination of matrix multiplications and backpropagation.…”