Design and Implementation of Intelligent Hardware of Neural Network Based on FPGA

2018 2nd annual International Conference on Cloud Technology and Communication Engineering

计算机科学;无线电电子学

Jing, Haiming^1 ; Zhao, Ning^1

Department of Information Science and Technology, Department of Economic and Management, Shijiazhuang Tiedao University, 17 Northeast, Second Inner Ring, Shijiazhuang, China^1

关键词: Design and implementations;    Embedded hardware;    Hardware circuits;    Hardware implementations;    Hardware realization;    Memory resources;    Neural network systems;    Simulation tests;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/466/1/012094/pdf
DOI  :  10.1088/1757-899X/466/1/012094

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

Neural networks, with its inherent parallel nature, is not suitable to realize in microprocessor, however, FPGA is widely applied in the hardware realization of neural network, for its parallel nature, abundant embedded hardware core unit with multiplying and adding function, and memory resources. According to the hardware circuit design, the neural network system is divided into several modules, and then the hardware implementation of each module is designed. The hardware circuit is realized by Verilog language, and the simulation test is carried out in Quartus II. According to the simulation results, whether the expected results are achieved is analyzed.