An area‐efficient memory‐based multiplier powering eight parallel multiplications for convolutional neural network processors

Autor:	Suhwan Cho, Byeong-Gyu Nam, Seongrim Choi
Rok vydání:	2021
Předmět:	Computer science Multiplier (economics) Electrical engineering. Electronics. Nuclear engineering Electrical and Electronic Engineering Arithmetic Convolutional neural network TK1-9971
Zdroj:	Electronics Letters, Vol 57, Iss 15, Pp 573-575 (2021)
ISSN:	1350-911X 0013-5194
Popis:	Convolutional neural network (CNN) is widely used for various deep learning applications because of its best‐in‐class classification performance. However, CNN needs several multiply‐accumulate (MAC) operations to realize human‐level cognition capabilities. In this regard, an area‐efficient multiplier is essential to integrate a large number of MAC units in a CNN processor. In this letter, we present an area‐efficient memory‐based multiplier targeting CNN processing. The proposed architecture adopts a 32‐port memory shared across eight multiplications. Simulation results show that area is reduced by 18.4% compared with the state‐of‐the‐art memory‐based multiplier.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_dedup___::bd522efa06215922d09adc42c59578f2 https://doi.org/10.1049/ell2.12206 Zobrazit plný text záznamu Plný text