Signal Detection and Estimation of Directional Parameters for Multiple Arrays

Abstract

We develop an integrated approach to estimating velocities and azimuths from a collection of local arrays and then fusing the data into Bayesian locations and their associated uncertainty ellipses. A small-array theory is developed that characterizes the performance of local optimal detectors under signal correlation and decorrelation scenarios. We compare the performance of maximum likelihood estimators such as the beam power and the generalized beam power as a function of array geometry and signal to noise ratio. Optimal local-array geometries are suggested that are relevant to the problem of designing an optimal infrasound array. Wave-number estimators along with estimated variance covariance matrices are used as input to study the size and orientation of 90% posterior probability ellipses for various likely subsets of detecting stations within the global infrasound array proposed for the Prototype International Data Center (PIDC) Adding detecting stations decreased the size of the 90% ellipse by about 10-20% per added station, whereas increasing the signal to noise ratio from 2 to 3 decreased the size of the ellipse by 30-40%. Adding an inner triangle to the conventional 1 km triangular way gave more modest reductions of 7-10%.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 2001
Accession Number
ADA400949

Entities

People

  • Robert H. Shumway
  • Sung-eun Kim

Organizations

  • University of California, Davis

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Computational Science
  • Covariance
  • Data Science
  • Detection
  • Detectors
  • Earth Sciences
  • Estimators
  • Geography
  • Geometry
  • Information Science
  • Models
  • Probability
  • Statistical Algorithms
  • Statistical Analysis
  • Statistics
  • Surveys

Readers

  • Radar Systems Engineering.
  • Seismology
  • Statistical inference.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference