An inverse problem in underwater acoustic source localization: robust matched-field processing

The model-based technique known as matched-field processing (MFP) has been shown to be effective for underwater source localization. It is well known that MFP can be sensitive to model mismatch and, in particular, environmental parameter mismatch. Therefore, robust MFP methods are required for practical application. The maximum a posteriori (MAP) estimator has been described as such a technique; however, it is computationally intensive and impractical to implement. In this paper, the L-infinity-norm estimator is derived as an efficient approximation to MAP. We demonstrate with simulated data that the performance of the L-infinity-norm estimator is superior to that of MAP as implemented in a manner required for practical application.