Deep-Sea: A Reconfigurable Accelerator for Classic CNN

Author:

Xiong Hao12ORCID,Sun Kelin1ORCID,Zhang Bing1,Yang Jingchuan1,Xu Huiping1

Affiliation:

1. Institute of Deep-Sea Science and Engineering, Chinese Academy of Sciences, Hainan 572000, China

2. School of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing 100049, China

Abstract

To meet the changing real-time edge engineering application requirements of CNN, aiming at the lack of universality and flexibility of CNN hardware acceleration architecture based on ARM+FPGA, a general low-power all pipelined CNN hardware acceleration architecture is proposed to cope with the continuously updated CNN algorithm and accelerate in hardware platforms with different resource constraints. In the framework of the general hardware architecture, a basic instruction set belonging to the architecture is proposed, which can be used to calculate and configure different versions of CNN algorithms. Based on the instruction set, the configurable computing subsystem, memory management subsystem, on-chip cache subsystem, and instruction execution subsystem are designed and implemented. In addition, in the processing of convolution results, the on-chip storage unit is used to preprocess the convolution results, to speed up the activation and pooling calculation process in parallel. Finally, the accelerator is modeled at the RTL level and deployed on the XC7Z100 heterogeneous device. The lightweight networks YOLOv2-tiny and YOLOv3-tiny commonly used in engineering applications are verified on the accelerator. The results show that the peak performance of the accelerator reaches 198.37 GOP/s, the clock frequency reaches 210 MHz, and the power consumption is 4.52 w under 16-bit width.

Funder

National Key Research and Development Plan of China

Publisher

Hindawi Limited

Subject

Electrical and Electronic Engineering,Computer Networks and Communications,Information Systems

Cited by 4 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Fast and Scalable Multicore YOLOv3-Tiny Accelerator Using Input Stationary Systolic Architecture;IEEE Transactions on Very Large Scale Integration (VLSI) Systems;2023-11

2. Shuffle-octave-yolo: a tradeoff object detection method for embedded devices;Journal of Real-Time Image Processing;2023-03-02

3. Distributed Network of Adaptive and Self-Reconfigurable Active Vision Systems;Symmetry;2022-10-31

4. Adaptive Hardware Architecture for Neural-Network-on-Chip;2022 IEEE 65th International Midwest Symposium on Circuits and Systems (MWSCAS);2022-08-07

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3