James C. Preisig
Dept. of Appl. Ocean Phys. and Eng., Woods Hole Oceanogr. Inst., Woods Hole, MA 02543
Uncertainty in the characteristics of the propagation environment and the resulting processor mismatch can lead to a significant degradation in the performance of adaptive matched-field processors. Most previously proposed techniques for dealing with environmental uncertainty model the environment as being underspread (i.e., the environment can be assumed to be constant over the observation interval of the processor). However, in some scenarios there are both underspread and overspread components to the environmental fluctuations. An adaptive matched-field processor is proposed that separates the environmental fluctuations into overspread and underspread components and handles the two types of variability in fundamentally different ways. The processor simultaneously optimizes its worse-case performance evaluated over the range of underspread environmental conditions and its average performance evaluated over the range of overspread environmental conditions.