Timing error detection in ultra dynamic voltage scaling systems

L. Koskinen, M. Turnquist, M. Hiienkari, Jani Mäkipää

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

Abstract

Energy per operation minimum can be reached, depending on the process node, at near- or subthreshold voltages. However, throughput of a system is severely limited at ultra-low operation voltages. Therefore, it would be ideal to have the system operate across a wide voltage range but achieving minimum energy operation simultaneously with operation over the entire voltage range is challenging. Further, process, supply voltage, temperature, and aging (PVTA) induced variation becomes a design challenge. In order to conform to varying throughput requirements and to compensate for PVTA variations, an adaptive design is required. In this paper, an extension to a previously presented subthreshold Timing-Error Detection (TED) microprocessor is proposed. The extension allows the system to operate efficiently at both low and high operation frequencies from 300 mV to 1.2 V .
Original languageEnglish
Title of host publicationProceedings
Subtitle of host publicationIEEE 27th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2012
PublisherIEEE Institute of Electrical and Electronic Engineers
ISBN (Electronic)978-1-4673-4681-8
ISBN (Print)978-1-4673-4682-5
DOIs
Publication statusPublished - 2012
MoE publication typeNot Eligible
EventIEEE 27th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2012 - Eilat, Israel
Duration: 14 Nov 201217 Nov 2012

Conference

ConferenceIEEE 27th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2012
Abbreviated titleIEEEI 2012
Country/TerritoryIsrael
CityEilat
Period14/11/1217/11/12

Keywords

  • Timing-Error Detection
  • Ultra-Dynamic Voltage Scaling
  • Ultra-Low Power

Fingerprint

Dive into the research topics of 'Timing error detection in ultra dynamic voltage scaling systems'. Together they form a unique fingerprint.

Cite this