Circular synthetic aperture sonar design

Abstract

Medical imaging enjoys a reputation of applying multi-aspect sensing to construct internal maps of patients. The CAT scan evokes a concept of an X-ray probe moving around the patient; the multi-aspect data is processed to construct images of the areas of interest. In contrast, advanced acoustic imaging of objects on the ocean floor has to date relied only on high resolution sensing from limited aspect angles using data gathered on straight-line trajectories. With the increased application of UUVs, it is fitting to consider the factors that govern the design of sonar that can reconstruct images taking advantage of data with aspect diversity. One future paradigm will be circular synthetic aperture sonar (CSAS) using tomographic methods to reconstruct scenes of interest from data obtained over a full circular aperture. This presentation will address the parameters and procedures that play a role in CSAS and the relations between them that govern the performance of such a system. The design of a CSAS for high resolution surveillance of objects on the ocean bottom depends on parameters such as the diameter of the circular trajectory, the speed of the platform, the size of the imaged circle, and the ping repetition rate. The spatial and range resolutions are functions of the sonar signal wavelength and its bandwidth. These parameters define in turn the maximum allowable uncorrectable platform position errors, which lead to defining the requirements of the hypernavigation system that is needed to correct for the errors between the actual ping and receiver positions and the ideal sonar trajectory used in the main imaging computation. A possible realization of such a navigation system is described in the presentation. The ping repetition rate requirements are determined by the largest dimension and shape of the reflecting areas to be imaged. Analytical expressions relating all of the above mentioned design parameters are derived. These relationships define various performance metrics as functions of the deployment parameters, allowing optimization of search strategies to maximize area coverage rate for a choice of multiple circle patterns. Images of several objects using data from a turntable testbed at Lake Travis Test Station of the Applied Research Laboratories of the University of Texas at Austin will be shown and discussed in relation to the design and test parameters.