Uncertainties of differential phase estimation associated with interferometric sonars

The differential phase technique has been widely used in various sonar systems; however, uncertainties associated with the estimation of scatterer depths are not completely understood. Numerical simulations for multiple bottom scatterers are performed, and they show that the uncertainties of depth measurements, in the absence of noise interferences, are much greater than the amount explainable by the uncertainty associated with the signal-arrival-angle within an instantaneous insonified area. The cause of the excess deviation is analyzed, particularly for the two-scatterer case. This kind of error is referred to as “baseline decorrelation” which is related to the speckle phenomena and can be considered as an equivalent noise source. Experimental data obtained by a particular high-frequency (40 kHz) interferometric system, the Benthic Acoustic Measurement System (BAMS) developed by the Applied Physics Laboratory, University of Washington, at a flat sandy bottom off the coast of Panama City, FL, were analyzed. Both analytical formulas and a numerical model are given to estimate the measurement uncertainty caused by the baseline decorrelation, as well as noise interferences based on the parameters of the BAMS, in order to understand uncertainties of the differential phase estimation. It is found that baseline decorrelation is the main source of error for the BAMS for grazing angles greater than 12°. The measurement uncertainties at this grazing angle interval are in agreement with the theoretical predictions