Booth Encoded Bit-Serial Multiply-Accumulate Units with Improved Area and Energy Efficiencies

Bit-serial multiply-accumulate units (MACs) play a crucial role in various hardware accelerator applications, including deep learning, image processing, and signal processing. Despite the advantages of bit-serial MACs, such as a small footprint, full hardware utilization, and high frequency, their s...

Full description

Bibliographic Details
Main Authors: Xiaoshu Cheng, Yiwen Wang, Jiazhi Liu, Weiran Ding, Hongfei Lou, Ping Li
Format: Article
Language:English
Published: MDPI AG 2023-05-01
Series:Electronics
Subjects:
Online Access:https://www.mdpi.com/2079-9292/12/10/2177
_version_ 1797600365217054720
author Xiaoshu Cheng
Yiwen Wang
Jiazhi Liu
Weiran Ding
Hongfei Lou
Ping Li
author_facet Xiaoshu Cheng
Yiwen Wang
Jiazhi Liu
Weiran Ding
Hongfei Lou
Ping Li
author_sort Xiaoshu Cheng
collection DOAJ
description Bit-serial multiply-accumulate units (MACs) play a crucial role in various hardware accelerator applications, including deep learning, image processing, and signal processing. Despite the advantages of bit-serial MACs, such as a small footprint, full hardware utilization, and high frequency, their serial nature can lead to high latency and potentially compromised performance. This study investigates the potential of bit-serial solutions by applying Booth encoding to bit-serial multipliers within MACs to enhance area and power efficiencies. We present two types of bit-serial MACs based on radix-2 and radix-4 Booth encoding multipliers, respectively. Their performance is assessed through simulations and synthesis results, demonstrating the benefits of the proposed approach. The radix-4 Booth bit-serial MAC improves power and area efficiencies compared to the original bit-serial MAC. Operating at TSMC 90 nm and 150 MHz, our design exhibits a remarkable 96.39% reduction in area-power-product (APP). Moreover, the prototype verification on a Xilinx Kintex-7 FPGA proved successful. The proposed solution offers significant advantages in energy efficiency, area reduction, and APP, making it a promising candidate for next-generation hardware accelerators in offline inference, low-power devices, and other applications.
first_indexed 2024-03-11T03:48:05Z
format Article
id doaj.art-c58d8e19efea4930876b6c3bc5aca2d5
institution Directory Open Access Journal
issn 2079-9292
language English
last_indexed 2024-03-11T03:48:05Z
publishDate 2023-05-01
publisher MDPI AG
record_format Article
series Electronics
spelling doaj.art-c58d8e19efea4930876b6c3bc5aca2d52023-11-18T01:08:44ZengMDPI AGElectronics2079-92922023-05-011210217710.3390/electronics12102177Booth Encoded Bit-Serial Multiply-Accumulate Units with Improved Area and Energy EfficienciesXiaoshu Cheng0Yiwen Wang1Jiazhi Liu2Weiran Ding3Hongfei Lou4Ping Li5School of Integrated Circuit Science and Engineering, University of Electronic Science and Technology of China, Chengdu 610054, ChinaSchool of Integrated Circuit Science and Engineering, University of Electronic Science and Technology of China, Chengdu 610054, ChinaSchool of Integrated Circuit Science and Engineering, University of Electronic Science and Technology of China, Chengdu 610054, ChinaSchool of Integrated Circuit Science and Engineering, University of Electronic Science and Technology of China, Chengdu 610054, ChinaSchool of Integrated Circuit Science and Engineering, University of Electronic Science and Technology of China, Chengdu 610054, ChinaSchool of Integrated Circuit Science and Engineering, University of Electronic Science and Technology of China, Chengdu 610054, ChinaBit-serial multiply-accumulate units (MACs) play a crucial role in various hardware accelerator applications, including deep learning, image processing, and signal processing. Despite the advantages of bit-serial MACs, such as a small footprint, full hardware utilization, and high frequency, their serial nature can lead to high latency and potentially compromised performance. This study investigates the potential of bit-serial solutions by applying Booth encoding to bit-serial multipliers within MACs to enhance area and power efficiencies. We present two types of bit-serial MACs based on radix-2 and radix-4 Booth encoding multipliers, respectively. Their performance is assessed through simulations and synthesis results, demonstrating the benefits of the proposed approach. The radix-4 Booth bit-serial MAC improves power and area efficiencies compared to the original bit-serial MAC. Operating at TSMC 90 nm and 150 MHz, our design exhibits a remarkable 96.39% reduction in area-power-product (APP). Moreover, the prototype verification on a Xilinx Kintex-7 FPGA proved successful. The proposed solution offers significant advantages in energy efficiency, area reduction, and APP, making it a promising candidate for next-generation hardware accelerators in offline inference, low-power devices, and other applications.https://www.mdpi.com/2079-9292/12/10/2177hardware acceleratorBooth encodingbit-serialmultipliermultiply-accumulate unit (MAC)
spellingShingle Xiaoshu Cheng
Yiwen Wang
Jiazhi Liu
Weiran Ding
Hongfei Lou
Ping Li
Booth Encoded Bit-Serial Multiply-Accumulate Units with Improved Area and Energy Efficiencies
Electronics
hardware accelerator
Booth encoding
bit-serial
multiplier
multiply-accumulate unit (MAC)
title Booth Encoded Bit-Serial Multiply-Accumulate Units with Improved Area and Energy Efficiencies
title_full Booth Encoded Bit-Serial Multiply-Accumulate Units with Improved Area and Energy Efficiencies
title_fullStr Booth Encoded Bit-Serial Multiply-Accumulate Units with Improved Area and Energy Efficiencies
title_full_unstemmed Booth Encoded Bit-Serial Multiply-Accumulate Units with Improved Area and Energy Efficiencies
title_short Booth Encoded Bit-Serial Multiply-Accumulate Units with Improved Area and Energy Efficiencies
title_sort booth encoded bit serial multiply accumulate units with improved area and energy efficiencies
topic hardware accelerator
Booth encoding
bit-serial
multiplier
multiply-accumulate unit (MAC)
url https://www.mdpi.com/2079-9292/12/10/2177
work_keys_str_mv AT xiaoshucheng boothencodedbitserialmultiplyaccumulateunitswithimprovedareaandenergyefficiencies
AT yiwenwang boothencodedbitserialmultiplyaccumulateunitswithimprovedareaandenergyefficiencies
AT jiazhiliu boothencodedbitserialmultiplyaccumulateunitswithimprovedareaandenergyefficiencies
AT weiranding boothencodedbitserialmultiplyaccumulateunitswithimprovedareaandenergyefficiencies
AT hongfeilou boothencodedbitserialmultiplyaccumulateunitswithimprovedareaandenergyefficiencies
AT pingli boothencodedbitserialmultiplyaccumulateunitswithimprovedareaandenergyefficiencies