본문 바로가기 주메뉴 바로가기
국회도서관 홈으로 정보검색 소장정보 검색

초록보기

Convolutional Neural Networks (CNNs) are getting fame due to their simpler design and higher performance. However, CNNs suffer from a large area and power consumption constraints. The multiply-and-accumulate (MAC) unit, which performs the convolution operation inside a CNN, consumes a significant amount of power consumption. In this study, we propose a mixed-signal approach for implementing analog MAC unit that can replace the digital MAC unit in CNNs. The Analog MAC unit architecture is constituted from binary weighted current steering digital-to-analog (DAC) circuit and capacitors. A digital parallel interface is designed to provide input image and filter values to the MAC unit. To realize a complete CNN model a low-power analog-to-digital (ADC) is then employed at the output to convert the final value back to a digital value. When a 3×3 convolution is performed, the analog MAC unit offers a 10.7% reduction in area and a 59.2% reduction in power consumption compared to its fully digital counterparts.

권호기사

권호기사 목록 테이블로 기사명, 저자명, 페이지, 원문, 기사목차 순으로 되어있습니다.
기사명 저자명 페이지 원문 목차
Design of power management integrate circuit implemented at the active EMI filter Sangyeong Jeong, Jingook Kim p. [1-6]

(A) low power mixed signal convolutional neural network for deep learning SoC Malik Summair Asghar, Syed Asmat Ali Shah, Hyung Won Kim p. [1-6]

Blood glucose monitoring and communicative implantable chip using a wireless power transfer Hyeonkeon Lee, Honghyeon Park, Sanghoek Kim p. [1-7]

D-band high gain and wide bandwidth power amplifier design in 65nm CMOS adopting dual-peak Gmax technique Hyo-Ryeong Jeon, Sang-Gug Lee p. [1-5]

Ultra-wideband pulse generator with simultaneous optimization of sidelobe suppression and essential bandwidth Hafiz Usman Mahmood, Jusung Kim, Sang-Gug Lee p. [1-6]

(A) low-ripple charge pump based on a parallelized ring oscillator with latches Woo Jin Jang, Yeon Jae Shin, Jung Hyup Lee p. [1-6]

Design of a continuous-time delta-sigma modulator for bio-signal acquisition Ye-Dam Kim, Seung-Tak Ryu p. [1-6]

9-bit 500-MS/s pipelined SAR ADC using dynamic amplifier with background calibration Soohoon Lee, Hyeonsik Kim, Jintae Kim p. [1-7]