A desirable feature of bathymetric sonar systems is the production of statistically independent soundings allowing a system to achieve its full capability in resolution and object detection. Moreover gridding algorithms such as the Combined Uncertainty Bathymetric Estimator (CUBE) rely on the statistical independence of soundings to properly estimate depth and discriminate outliers. Common methods of filtering to mitigate uncertainty in the signal processing of both multibeam and phase-differencing sidescan systems (curve fitting in zero-crossing detections and differential phase filtering respectively) can produce correlated soundings. Here we propose an alternative method for the generation of soundings from differential phase measurements made by either sonar type to produce statistically independent soundings. The method extracts individual, non-overlapping and unfiltered, phase-difference measurements (from either sonar type) converting these to sonar-relative receive angle, estimates their uncertainty, fixes the desired depth uncertainty level and combines these individual measurements into an uncertainty-weighted mean to achieve the desired depth uncertainty, and no more. When the signal to noise ratio is sufficiently high such that the desired depth uncertainty is achieved with an individual measurement, bathymetric estimates are produced at the sonar’s full resolution capability. When multiple measurements are required, the filtering automatically adjusts to maintain the desired uncertainty level, degrading the resolution only as necessary. Because no two measurements contribute to a single reported sounding, the resulting estimated soundings are statistically independent and therefore better resolve adjacent objects, increase object detectability and are more suitable for statistical gridding methodologies.

Publication Date


Journal or Conference Title

Canadian Hydrographic Conference 2014

Conference Date

April 14-17, 2014

Publisher Place

St John's, NL, Canada

Document Type

Conference Proceeding