Generalized ILP scheduling and allocation for high-level DSP synthesis

Lori E. Lucke, Keshab K. Parhi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Scopus citations

Abstract

Recently integer linear programming techniques have been proposed to solve the single iteration scheduling problem. In this paper, we generalize the integer linear programming formulation to support unfolded schedules. We present a new time assignment formulation which automatically pipelines and re-times the DFG. We present a new processor allocation formulation which automatically unfolds the DFG. Although unfolding was proposed before for reducing the iteration period, this paper, for the first time, proposes systematic unfolding for improving processor utilization in digital signal processing architectures. These models fully optimize the processor utilization for a specific iteration period.

Original languageEnglish (US)
Title of host publicationProceedings of the Custom Integrated Circuits Conference
PublisherPubl by IEEE
Pages5.4.1-5.4.4
ISBN (Print)0780308263
StatePublished - Jan 1 1993
EventProceedings of the IEEE 1993 Custom Integrated Circuits Conference - San Diego, CA, USA
Duration: May 9 1993May 12 1993

Publication series

NameProceedings of the Custom Integrated Circuits Conference
ISSN (Print)0886-5930

Other

OtherProceedings of the IEEE 1993 Custom Integrated Circuits Conference
CitySan Diego, CA, USA
Period5/9/935/12/93

Fingerprint Dive into the research topics of 'Generalized ILP scheduling and allocation for high-level DSP synthesis'. Together they form a unique fingerprint.

Cite this