Reprint

Circuits and Systems Advances in Near Threshold Computing

Edited by
May 2021
120 pages
  • ISBN978-3-0365-0720-0 (Hardback)
  • ISBN978-3-0365-0721-7 (PDF)

This book is a reprint of the Special Issue Circuits and Systems Advances in Near Threshold Computing that was published in

Chemistry & Materials Science
Engineering
Physical Sciences
Summary
Modern society is witnessing a sea change in ubiquitous computing, in which people have embraced computing systems as an indispensable part of day-to-day existence. Computation, storage, and communication abilities of smartphones, for example, have undergone monumental changes over the past decade. However, global emphasis on creating and sustaining green environments is leading to a rapid and ongoing proliferation of edge computing systems and applications. As a broad spectrum of healthcare, home, and transport applications shift to the edge of the network, near-threshold computing (NTC) is emerging as one of the promising low-power computing platforms. An NTC device sets its supply voltage close to its threshold voltage, dramatically reducing the energy consumption. Despite showing substantial promise in terms of energy efficiency, NTC is yet to see widescale commercial adoption. This is because circuits and systems operating with NTC suffer from several problems, including increased sensitivity to process variation, reliability problems, performance degradation, and security vulnerabilities, to name a few. To realize its potential, we need designs, techniques, and solutions to overcome these challenges associated with NTC circuits and systems. The readers of this book will be able to familiarize themselves with recent advances in electronics systems, focusing on near-threshold computing.
Format
  • Hardback
License
© 2022 by the authors; CC BY-NC-ND license
Keywords
machine learning; neural networks; gait analysis; embedded system; NTV; NTC; low-power; low-voltage memory and clocking circuits; minimum-energy design; power-performance; resilient adaptive computing; edge devices; power management; energy efficiency; near-threshold computing (NTC); deep neural network (DNN); accelerators; timing error; AI; tensor processing unit (TPU); multiply and accumulate (MAC); energy efficiency; reliability; Near-Threshold Computing; functional unit; energy efficiency; performance optimization; cross-layer optimization