UC San Diego

A typical geoacoustic inversion procedure involves powerful source transmissions received on a large-aperture receiver array. A more practical approach is to use a moving single source/receiver, broadband, frequency- coherent matched-field inversion strategy that exploits coherently repeated transmissions to improve estimation of the geoacoustic parameters. The long observation time creates a synthetic aperture due to relative source- receiver motion. To correlate well with the measured field, waveguide Doppler and normal mode theory is applied. However, this method uses a waveguide Doppler model that constrains the source/receiver radial velocity to be constant. As a result, the inversion performance degrades when source/receiver acceleration exists. Furthermore, processing a train of pulses all-at-once does not take advantage of the natural incremental acquisition of new pulses along with the ability to assess the temporal evolution of parameter uncertainty. Therefore, a recursive Bayesian estimation approach is developed that coherently processes the data pulse-by-pulse and incrementally updates estimates of parameter uncertainty. It also approximates source/receiver acceleration by assuming piecewise constant but linearly changing source/receiver velocities. When the source/receiver acceleration exists, it is shown that modeling acceleration can reduce further the parameter estimation biases and uncertainties. Finally, the above methods depended on the assumption of constant underlying geophysical model parameters. A change-point detection method is proposed to detect the change in the model parameters using the importance samples and corresponding weights that already are available from the recursive Bayesian inversion. If the model parameters change abruptly, a change-point will be detected and the inversion will restart with the pulse measurement after the change-point. If the model parameters change gradually, the inversion (based on constant model parameters) may proceed to estimate an averaged value of the parameters until the accumulated model parameter mismatch is significant and triggers the detection of a change-point. These form the heuristics for controlling the coherent integration time in recursive Bayesian inversion. Examples are based either on synthetically generated acoustic fields using the waveguide Doppler model or a set of low SNR, 100-900 Hz LFM pulse data from a moving source- receiver pair in the Shallow Water 2006 experiment