A Fast Maximum Likelihood Multidimensional Search Algorithm

Abstract

This memorandum examines an application of the least-squares approximation of certain structured matrixes to a new algorithm for the maximum likelihood (ML) estimation of directions of arrival (DOAs) and frequencies of sinusoids. The method involves reordering the steps in the standard ML search algorithm so that the majority of the calculations are made prior to execution of the multidimensional search. The computational efficiency is due to the fact that only scalar operations are performed during the search. This new fast maximum likelihood estimation (FMLE) algorithm is more than one order of magnitude faster than the standard direct-search ML method.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 15, 1998
Accession Number
ADA637097

Entities

People

  • Craig S. Macinnes

Organizations

  • Naval Undersea Warfare Center

Tags

Communities of Interest

  • Ground and Sea Platforms

DTIC Thesaurus Topics

  • Algorithms
  • Arrays
  • Computational Complexity
  • Computational Science
  • Covariance
  • Direction Finding
  • Efficiency
  • Equations
  • Frequency
  • Mathematical Analysis
  • Maximum Likelihood Estimation
  • Noise
  • Numbers
  • Simulations
  • Standards
  • Undersea Warfare
  • Very Low Frequency

Fields of Study

  • Engineering

Readers

  • Operations Research
  • Statistical inference.