Efficient Correlation Matrix Estimators for FPGA Implementation

Abstract

Effective missile defense requires computer systems with extraordinary real time computing capacity. Parallel architectures are necessary to provide these levels of performance. Alternative architectures can be developed by integrating the design of the numerical algorithm with the computing hardware. One such emerging technology is reconfigurable computing based on Field Programmable Gate Arrays (FPGAs). ISL has developed nonlinear operators that are easily implemented on FPGAs and can be used to implement correlators, matched filters, adaptive filters, and neural networks. In Phase 1 of this SBIR, ISL has implemented the nonlinear correlation estimators in FPGA processors and compared their performance in a standard application: DOA estimation. We have also considered the interconnection on multiple FPGA's and potential applications of such a reconfigurable computer. Based on this investigation, we propose recommendations for future development in a Phase 2 continuation of the present work.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Feb 02, 1998
Accession Number
ADA337434

Entities

People

  • Amir Sarajedini
  • J. Doss Halsey
  • Paul M. Chan

Tags

Communities of Interest

  • Advanced Electronics
  • Energy and Power Technologies
  • Materials and Manufacturing Processes
  • Space

DTIC Thesaurus Topics

  • Algorithms
  • Computational Complexity
  • Computer Architecture
  • Computer Programming
  • Computer Programs
  • Computers
  • Data Processing
  • Detectors
  • Digital Signal Processing
  • Estimators
  • Field Programmable Gate Arrays
  • Information Systems
  • Integrated Circuits
  • Parallel Computing
  • Situational Awareness
  • Target Recognition
  • Web Browsers

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Integrated Circuit Design and Technology.
  • Parallel and Distributed Computing.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference