Geoacoustic inversion in a range-dependent environment under the assumption of range-independence

Michael G. Morley, Stan E. Dosso, N. Ross Chapman

Abstract


The estimation of the geoacoustic properties of the seabed by matched-field inversion (MFI) under the assumption of range-independence is discussed. The assumption of range independence reduces the number of parameters in the problem and significantly improves the computational efficiency of MFI. The degree of random, range dependent variability in water depth and seabed sound speed that can be tolerated are examined by inverting synthetic datasets generated from many unique and random realizations of the environment. The effect on the accuracy of the inversion due to the ignorance of the random variability of individual model parameters was also investigated. The single frequency inversion results are broadly distributed with large biases, which indicate that significant fluctuations of sound speed in the seabed can be effectively averaged over by including multiple frequencies in the inversion.

Keywords


Acoustic signal processing; Computation theory; Database systems; Parameter estimation; Speech processing; Computational efficiency; Multiple frequencies; Synthetic datasets

Full Text:

PDF

Refbacks

  • There are currently no refbacks.