Bayesian approaches to geoacoustic inversion

  • Anna-Liesa S. Lapinski DRDC Atlantic, Dartmouth, NS, Canada
Keywords: Inverse problems, Parameter estimation, Probability density function, Sediments, Simulated annealing, Adaptive simplex simulated annealing (ASSA), Fast Gibbs samplers (FGS)


The appropriate number of sediment layers and parameter estimates were determined by using the fast Gibbs sampler (FGS) approach and adaptive simplex simulated annealing (ASSA). An under-parameterized approach was applied to determine the appropriate number of sediment layers needed to represent the seabed. The results showed that the FGS while not as effective as ASSA, provided uncertainty bounds which were crucial for assessing the quality of the final estimate.
How to Cite
LapinskiA-LS. Bayesian approaches to geoacoustic inversion. Canadian Acoustics [Internet]. 2002Sep.1 [cited 2020May29];30(3):96-7. Available from: //
Proceedings of the Acoustics Week in Canada