High-Speed CNN Accelerator SoC Design Based on a Flexible Diagonal Cyclic Array
-
Published:2024-04-19
Issue:8
Volume:13
Page:1564
-
ISSN:2079-9292
-
Container-title:Electronics
-
language:en
-
Short-container-title:Electronics
Author:
Lee Dong-Yeong1, Aliev Hayotjon1, Junaid Muhammad1ORCID, Park Sang-Bo1, Kim Hyung-Won1, Lee Keon-Myung2ORCID, Sim Sang-Hoon1
Affiliation:
1. Department of Electronics Engineering, Chungbuk National University, Cheongju 28644, Republic of Korea 2. Department of Computer Science, Chungbuk National University, Cheongju 28644, Republic of Korea
Abstract
The latest convolutional neural network (CNN) models for object detection include complex layered connections to process inference data. Each layer utilizes different types of kernel modes, so the hardware needs to support all kernel modes at an optimized speed. In this paper, we propose a high-speed and optimized CNN accelerator with flexible diagonal cyclic arrays (FDCA) that supports the acceleration of CNN networks with various kernel sizes and significantly reduces the time required for inference processing. The accelerator uses four FDCAs to simultaneously calculate 16 input channels and 8 output channels. Each FDCA features a 4 × 8 systolic array that contains a 3 × 3 processing element (PE) array and is designed to handle the most commonly used kernel sizes. To evaluate the proposed CNN accelerator, we mapped the widely used YOLOv5 CNN model and evaluated the performance of its implementation on the Zynq UltraScale+ MPSoC ZCU102 FPGA. The design consumes 249,357 logic cells, 2304 DSP blocks, and only 567 KB BRAM. In our evaluation, the YOLOv5n model achieves an accuracy of 43.1% (mAP@0.5). A prototype accelerator has been implemented using Samsung’s 14 nm CMOS technology. It achieves 1.075 TOPS, a peak performance with a 400 MHz clock frequency.
Reference34 articles.
1. Akkad, G., Mansour, A., and Inaty, E. (IEEE Trans. Artif. Intell., 2023). Embedded Deep Learning Accelerators: A Survey on Recent Advances, IEEE Trans. Artif. Intell., early access. 2. Jocher, G., Stoken, A., Chaurasia, A., Borovec, J., Xie, T., Kwon, Y., Michael, K., Changyu, L., and Fang, J. (2021, October 12). Yolov5. NanoCode012. v6.0—Models. Available online: https://github.com/ultralytics/yolov5. 3. FPGA-Based High-Throughput CNN Hardware Accelerator with High Computing Resource Utilization Ratio;Huang;IEEE Trans. Neural Netw. Learn. Syst.,2022 4. S2 Engine: A Novel Systolic Architecture for Sparse Convolutional Neural Networks;Yang;IEEE Trans. Comput.,2022 5. Wei, X., Yu, C.H., Zhang, P., Chen, Y., Wang, Y., Hu, H., Liang, Y., and Cong, J. (2017, January 18–22). Automated systolic array architecture synthesis for high throughput CNN inference on FPGAs. Proceedings of the 2017 54th ACM/EDAC/IEEE Design Automation Conference (DAC), Austin, TX, USA.
Cited by
1 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
|
|