Recent array processing methods for ocean acoustics have utilized the
physics of wave propagation as an integral part of their design. The p
hysics of the propagation leads to both improved performance and to al
gorithms where the complexity of the ocean environment can be exploite
d in ways not possible with traditional plane wave based methods. Matc
hed field processing (MFP) is a generalized beamforming method which u
ses the spatial complexities of acoustic fields in an ocean waveguide
to localize sources in range, depth and azimuth or to infer parameters
of the waveguide itself. It has experimentally localized sources with
accuracies exceeding the Rayleigh limit for depth and the Fresnel lim
it for range by two orders of magnitude. MFP exploits the coherence of
the mode/multipath structure and it is especially effective at low fr
equencies where the ocean supports coherent propagation over very long
ranges. This contrasts with plane wave based models which are degrade
d by modal and multipath phenomena and are generally ineffective when
waveguide phenomena are important. MFP can have either conventional or
adaptive formulations and it has been implemented with an assortment
of both narrowband and wideband signal models. All involve some form o
f correlation between the replicas derived from the wave equation and
the data measured at an array of sensors. One can view MFP as an inver
se problem where one attempts to ''invert'' the wave equation for thes
e dependencies over the parameter space of the source and the environm
ent. There is currently a large literature discussing many theoretical
aspects of MFP including numerous simulations; several experiments ac
quiring data for MFP now have been conducted in several ocean environm
ents and these have demonstrated both its capabilities and some of its
limitations. Consequently, there is a modest understanding of both th
e theory and the experimental capabilities of MFP. This article provid
es an overview of both.