A stationary approximation in matched-field processing for a moving underwater acoustic source

Auteurs-es

  • C.A. Zala Barrodale Comput. Services Ltd., Victoria, BC, Canada
  • J.M. Ozard Barrodale Comput. Services Ltd., Victoria, BC, Canada

Mots-clés :

acoustic signal processing, underwater sound, normal-mode model, matched-field processing, moving source, array covariance matrices, snapshot vectors, replica matrices, stationary approximation procedure

Résumé

A normal-mode model for the acoustic field due to a moving source was recently used in the development of procedures for applying matched-field processing (MFP) techniques to a moving source. This approach involved matching `measured' array covariance matrices at multiple frequencies using replica matrices computed with a phase-expanded form of the Hawker model (1979). Excellent results were obtained, but the technique was compute-intensive. A procedure is described for using snapshot vectors at a single frequency to generate the replica matrices. By comparing the performance of the MFP techniques applied to simulated data, it is shown that this new stationary approximation procedure allows the position of a moving source to be estimated as accurately as, but much more efficiently than, the full multiple-frequency procedure

Fichiers supplémentaires

Publié-e

1992-09-01

Comment citer

1.
Zala C, Ozard J. A stationary approximation in matched-field processing for a moving underwater acoustic source. Canadian Acoustics [Internet]. 1 sept. 1992 [cité 19 févr. 2025];20(3):75-6. Disponible à: https://jcaa.caa-aca.ca/index.php/jcaa/article/view/743

Numéro

Rubrique

Actes du congrès de la Semaine canadienne d'acoustique