MATCHED FIELD PROCESSING FOR A MOVING SOURCE

Authors
  1. Zala, C.
Corporate Authors
Defence Research Establishment Pacific, Victoria BC (CAN);Barrodale Computing Services Ltd, Victoria BC (CAN)
Abstract
Methods are presented for applying matched field processing (MFP) to the estimation of the position and velocity of a moving source. These MFP methods for nonstationary sources match the full set of data obtained for an observation time during which the motion of the source is linear. Two classes of processor are presented. The first matches the observed vectors of hydrophone field values for each frequency bin and time segment with corresponding replica vectors. The second processes the observed data to form a time-averaged covariance matrix for each frequency bin and then matches these matrices using the largest eigenvectors of corresponding replica matrices. Versions of the Bartlett, minimum energy, and MUSIC beamformers are proposed for the new MFP procedures, and their performance is illustrated in tests with synthetic data. TRUNCATED
Report Number
DREP-CRS-91-05 — Contractors Report Series
Date of publication
15 Mar 1991
Number of Pages
52
DSTKIM No
91-02776
CANDIS No
69972
Format(s):
Hardcopy;Originator's fiche received by DSIS

Permanent link

Document 1 of 1

Date modified: