A model-based acoustic time-reversal mirror for robust variable focusing

A natural extension of the acoustic time-reversal mirror (TRM) is investigated where predicted receive array signals are backpropagated to an arbitrary virtual source location. First, large time-bandwidth product (TB), broadband signals propagating from a real probe source to the array are processed iteratively by a model-based matched filter (MBMF) receiver in searching for the environmental parameters that best correspond to the measured acoustic-impulse responses. The process ends when most of the time-spread energy across the array is recombined coherently (processing gain). Second, the resulting acoustic-channel model is used to synthesize the transmit signals that will focus their energy at the desired range and depth (the virtual source). The procedure overcomes the noise and reverberation limitations inherent to the TRM operation. The at-sea, focusing performances of a model-based TRM were predicted from related MBMF experiments: range-depth localization of a distant towed source in a duct (west sardinia 89\N90) and bottom geoacoustic characterization in shallow water (yellow shark 94\N95) [Hermand et al., IEEE J. Oceanic Eng. 18, 447–465 (1993); 24, 41–66 (1999)]. Simulation results using yellow shark 94, large TB, broadband vertical array data will be compared with trm 96\N97 experimental results [Hodgkiss et al., J. Acoust. Soc. Am. 105, 1597–1604 (1999)] obtained along the same transect