Array element localization accuracy and survey design

Abstract

Accurate localization of the individual elements of an underwater acoustic receiver array is an important prerequisite to advanced array processing applications. Array element localization (AEL) methods are typically based on inverting acoustic arrival-time measurements from controlled sources at (approximately) known positions to the receivers to be localized. This paper presents and illustrates a general approach to AEL inversion and to AEL survey design based on quantifying the posterior receiverlocation uncertainty, taking into account uncertainties in the data, source locations, sound speed, and water depth. The inversion is based on a fast ray-tracing algorithm that employs Newton's method and the method of images to determine eigenrays for direct and reflected arrivals. The efficiency of this approach allows computationally intensive analysis such as Monte-Carlo appraisal and nonlinear optimization for designing optimal source configurations. These algorithms provide a rigorous approach that can be applied to examine all aspects of AEL accuracy and survey design, illustrated here by several examples. It is shown that synchronized AEL surveys (in which source transmission times are known) provide only a minor improvement over non-synchronized surveys (often much simpler logistically), and the difference can be made up by using more sources in an optimal configuration or by including additional arrivals. Including multiple-reflected arrivals improves receiver depth estimates (provided water depth is well known), but provides little improvement in horizontal localization.