ASA 126th Meeting Denver 1993 October 4-8

4aUW12. Towed array processing as a spatial Kalman filter problem.

E. J. Sullivan

Naval Undersea Warfare Ctr., Code 103, Newport, RI 02841

J. V. Candy

Lawrence Livermore Natl. Lab., Livermore, CA 94550

A moving towed array essentially constitutes a system intended to provide a series of measurements that are sequential in space. This type of process fits quite naturally into an iterative measurement process that can be realized as a spatial predictor-corrector form of a Kalman filter. It is shown how this concept is formulated and constitutes a generalization of the overlap-correlator synthetic aperture algorithm [E. J. Sullivan and S. Stergiopoulos, J. Acoust. Soc. Am. 86, 158--171 (1989)]. Examples will be given that demonstrate how this scheme constitutes a self-consistent synthetic aperture algorithm capable of determining estimates of the bearings of multiple sources, both correlated or uncorrelated. It will also be shown how the method provides coherent synthetic aperture processing over integration times that are significantly longer than the coherence time of the source signals.