|
Abstract |
|---|
There are many fine numerical methods for predicting underwater sound fields when the geometrical and mechanical properties of the environment are known. However, environmental knowledge at the requisite level of detail to fully exploit code capabilities is seldom available, and the resulting uncertainty in the predicted sound fields is problematic for a variety of sonar applications and environmental assessment techniques. In addition, the usual means of uncertainty assessment, direct-simulation and/or Monte-Carlo methods, involve repeated sound-field calculations and may be too computationally expensive for real-time applications.
Thus, an efficient way to predict acoustic uncertainty and environmental sensitivity is needed. This presentation covers four techniques for quantifying acoustic uncertainty in shallow ocean environments: field-distribution transport, direct simulation, polynomial chaos, and field shifting. Here the various techniques are used to estimate the probability density function (PDF) for the acoustic field produced by a harmonic point source at (0,zs) at a distant range-depth location (r,z). Comparisons between the various techniques are made in two-dimensional range-depth environments with one or more uncertain environmental parameters. The results presented are drawn from range-independent modal-sum field calculations at frequencies from 100 to 1,000 Hz and source-receiver ranges from 1 to 10 km in sound channels with depths of 50 and 100 m. The comparisons show that the first technique is analytically incomplete (at least at the present time), the next two are effective but computationally burdensome, while the final one is approximate but computationally efficient.
|
