Bayesian source tracking and environmental inversion


  • Stan E. Dosso School of Earth and Ocean Sciences, University of Victoria, Victoria BC V8W 3P6, Canada


Acoustics, Bayesian networks, Differential equations, Inverse problems, Random errors, Sedimentation, Signal to noise ratio, Speed, Ultrasonic velocity measurement, Underwater audition, Acoustic datum, Acoustic sources, Attenuation, , Bayesian, Bayesian approaches, Environmental, Gaussian errors, Radial velocities, Seabed geoacoustic, Sediment layers, Sound speeds, Source tracking


A Bayesian approach to two important and related inverse problems in underwater acoustics is presented. Seabed geoacoustic parameters include the thickness of an upper sediment layer with sound speed, density, and attenuation, overlaying a semi-infinite basement with sound speed, density and attenuation. The track consists of an acoustic source at 30-m depth moving away from the array at a constant radial velocity of 5 m/s. Acoustic data are collected at the array once per 9 minutes, corresponding to to source-receiver ranges of 4.0, 4.3, and 6.4 km. Random complex-Gaussian errors are added to the data to achieve a signal-to-noise (SNR) that varies from -2 to -8 dB.



How to Cite

Dosso SE. Bayesian source tracking and environmental inversion. Canadian Acoustics [Internet]. 2008Sep.1 [cited 2021Apr.17];36(3):174-5. Available from:



Proceedings of the Acoustics Week in Canada