Robust broadband matched-field processing: performance in shallow water by simulation

Abstract

An issue of concern for matched-field processing is the strong dependence between performance and

precise knowledge about the environmental parameters. Recently a robust matched-field processor

based on minimax robust filtering methods was developed. In this report simulation methods are

employed to evaluate the performance of the minimax robust method as well as other robust methods

for a range-independent shallow-water environment. The performance of the robust methods is

compared with that of the nominal processor, that is, the processor based on a single set of environmental parameters thought to be closest to the actual. The matched-field processing performance is evaluated in terms of the peak-to-sidelobe ratio, the simulation results indicate that the robust methods provide significant performance improvements over the nominal processor in the presence of uncertainty in water-column sound-speed, channel depth and sound speed in the bottom.