An Efficient VHDL Implementation of two Artificial Neural Networks on Zynq-7000 FPGA

被引：0

作者：

Sadeghikhah, Kaveh ^{[1
]}

Zhang, Lei ^{[1
]}

Paranjape, Raman ^{[1
]}

机构：

[1] Univ Regina, Engn & Appl Sci, Regina, SK, Canada

来源：

2023 IEEE CANADIAN CONFERENCE ON ELECTRICAL AND COMPUTER ENGINEERING, CCECE | 2023年

关键词：

artificial neural network; FPGA; VHDL; training; activation function; MATLAB; weights; and biases;

D O I：

10.1109/CCECE58730.2023.10288905

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

This paper presents the FPGA implementation of two different topologies of an Artificial Neural Network (ANN) on the Xilinx Zynq-7000 evaluation board. The engine dataset available in MATLAB is used to train the neural network. The resulting parameters of a neural network are taken from MATLAB and are used to implement it on FPGA. Two structures are implemented with different accuracy of sfix_24_8 and sfix_32_16 and different clock frequencies and resource utilization is measured. The maximum achievable frequency measured is 83.33 MHz and the minimum power is 0.203 W.

引用

页数：6

共 9 条

[1]

Abdu-Aljabar R., 2018, Journal of Engineering and Development

[2]

Demuth Howard B., 2017, MATLAB Neural Network Toolbox User's Guide

[3]

Dondon P, 2014, 2014 12TH SYMPOSIUM ON NEURAL NETWORK APPLICATIONS IN ELECTRICAL ENGINEERING (NEUREL), P37, DOI 10.1109/NEUREL.2014.7011454

[4] Reconfigurable Hardware Design Approach for Economic Neural Network [J].

Khalil, Kasem ;

Kumar, Ashok ;

Bayoumi, Magdy .

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS, 2022, 69 (12) :5094-5098

[5]

Rosado-Munoz Alfredo, 2019, IEEE Access

[6]

Schmitz Jesse, 2018, 9 INT S HIGHL EFF AC

[7]

Xilinx Vivado Design Suite Tutorial, 2020, Model-Based Design Using Model Composer

[8]

Xilinx Vivado Design Suite Tutorial, 2021, Zynq-7000 All Programmable SoC: Embedded Design Tutorial

[9] Implementation of Fixed-point Neuron Models with Threshold, Ramp and Sigmoid Activation Functions [J].

Zhang, Lei .

4TH INTERNATIONAL CONFERENCE ON MECHANICS AND MECHATRONICS RESEARCH (ICMMR 2017), 2017, 224

← 1 →