A FREQUENCY-DOMAIN SYNTHESIS PROCEDURE FOR MULTIDIMENSIONAL MAXIMUM- LIKELIHOOD PROCESSING OF SEISMIC ARRAYS

Abstract

A frequency-domain synthesis method for multidimensional maximum- likelihood filtering of sampled data obtained from seismic arrays is presented. This procedure is shown to possess several advantages relative to the time- domain synthesis technique. The primary advantage is that the frequency-domain method requires approximately ten times less computer time to synthesize the filter than does the time-domain technique. The details of a direct segment method for the spectral matrix estimation required in the frequency-domain approach are presented. In addition, the bias, variance, mean square error, limiting distribution, and other properties of the spectral estimates are discussed. The details of a Fortran IV computer program implementation of the frequency-domain method are given. The experimental results obtained by processing two events recorded at the Large aperture Seismic Array are presented as well as a comparison of the performance of the frequency-domain method relative to the time-domain synthesis technique. It is found that the processed noise power reduction for the frequency-domain method is typically about two out of a total of 20 db worse than that of the time-domain technique.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 06, 1966
Accession Number
AD0634233

Entities

People

  • Jack Capon
  • Robert J. Kolker
  • Roy J. Greenfield

Organizations

  • Massachusetts Institute of Technology

Tags

Communities of Interest

  • Energy and Power Technologies
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Air Force
  • Algorithms
  • Coefficients
  • Computations
  • Computer Programs
  • Computers
  • Data Science
  • Frequency Domain
  • Gaussian Processes
  • Hard Copy
  • Information Science
  • Noise Reduction
  • Phase Velocity
  • Sampling
  • Seismic Arrays
  • Time Domain
  • Weighting Functions

Fields of Study

  • Engineering

Readers

  • Electromagnetic Wave Scattering and Antenna Radiation Engineering
  • Image Processing and Computer Vision.
  • Regression Analysis.