Analog/Mixed-Signal Standard Cell Based Approach for Automated Circuit Generation of Neural Network Accelerators

Roland Muller, Loreto Mateu, Ralf Brederlow

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Analog and mixed-signal neural network accelerators are a promising solution to apply deep learning methods to edge applications where high energy and area efficiency are required. Such in-memory computing implementations use regular and repetitive circuit structures that take great advantage of design automation. An analog/mixed-signal standard cell design approach in combination with an automation framework has been developed to ease the design of such systems. The framework discussed here provides the basic functionality such as schematic and layout creation. It is based on manually designed standard cells and technology and topology parameters to steer the automation. The presented methodology drastically reduces the (re-)design time and engineering effort leading to a reduced time-to-market whilst errors occurring in manual executed circuit design can be avoided.

Original languageEnglish
Title of host publication2023 38th Conference on Design of Circuits and Integrated Systems, DCIS 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350303858
DOIs
StatePublished - 2023
Event38th Conference on Design of Circuits and Integrated Systems, DCIS 2023 - Malaga, Spain
Duration: 15 Nov 202317 Nov 2023

Publication series

Name2023 38th Conference on Design of Circuits and Integrated Systems, DCIS 2023

Conference

Conference38th Conference on Design of Circuits and Integrated Systems, DCIS 2023
Country/TerritorySpain
CityMalaga
Period15/11/2317/11/23

Keywords

  • AI accelerators
  • Electronic design automation
  • analog computing
  • analog/mixed-signal circuits
  • integrated circuits
  • neuromorphic computing
  • neuromorphic hardware

Fingerprint

Dive into the research topics of 'Analog/Mixed-Signal Standard Cell Based Approach for Automated Circuit Generation of Neural Network Accelerators'. Together they form a unique fingerprint.

Cite this