Design and Analysis of systolic Array for Convolution Neural Networks

Main Article Content

Nayana D K, et. al.

Abstract

Convolutional neural networks (CNN) are used in many functions mainly Digital signal processing (DSPs), Networking, Image and video classification for its accuracy. CNNs need huge estimation and support. To fix hardware, Convolutional neural network accelerators required to be enhanced to bring down size of the memory, resource utilization and power usage. In this paper, the systolic array with Multiply-Accumulate unit (MAC) is designed. The input image of 128x128 size is considered and processed through main memory to systolic array. The systolic array performs the MAC unit operations rows and columns wise. The work is carried out using Xilinx ISE 14.7 environment and synthesised on Artix-7 FPGA.

Downloads

Download data is not yet available.

Metrics

Metrics Loading ...

Article Details

How to Cite
et. al., N. D. K. . (2021). Design and Analysis of systolic Array for Convolution Neural Networks . Turkish Journal of Computer and Mathematics Education (TURCOMAT), 12(12), 2315–2320. https://doi.org/10.17762/turcomat.v12i12.7818
Section
Research Articles