SYSTOLIC DESIGN FOR STATE SPACE MODELS: KALMAN FILTERING AND COMPUTATIONAL NEURAL NETWORKS.

S. Y. Kung, J. N. Hwang

Research output: Contribution to journalConference articlepeer-review

3 Scopus citations

Abstract

A systematic mapping methodology is introduced for deriving systolic and wavefront arrays from regular computational algorithms. It consists of three stages of mapping design: (data) dependence graph (DG) design, signal flow graph (SFG) design, and array processor design. This methodology allows systolic design with many desirable properties, such as local communication and fastest pipeline rates, etc. Based on this methodology, systolic array designs are developed for two applications of adaptive state-space models. One is for the Kalman filtering algorithm which is popular in many digital signal processing applications, and the other is the Hopfield model for artificial neural networks (ANN), which has recently received increasing attention from the artificial-intelligence (AI) and parallel-processing research community.

Original languageEnglish (US)
Pages (from-to)1461-1467
Number of pages7
JournalProceedings of the IEEE Conference on Decision and Control
StatePublished - 1987

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Modeling and Simulation
  • Control and Optimization

Fingerprint

Dive into the research topics of 'SYSTOLIC DESIGN FOR STATE SPACE MODELS: KALMAN FILTERING AND COMPUTATIONAL NEURAL NETWORKS.'. Together they form a unique fingerprint.

Cite this