Broadband synthetic aperture geoacoustic inversion

Range-dependent geoacoustic inversion using equivalent environmental model in the presence of doppler effect

Geoacoustic inversion using moving sensors attracts lots of interest due to the ease of deployment and low cost. However, the well-established techniques, such as matched-field inversion (MFI), may run into difficulties when the sensors are in a range-dependent environment for mismatch issues and increasing unknown parameters. Given a range-dependent environment, the paper focuses on the inversion using a synthetic aperture created by moving sensors in the presence of the Doppler effect. The derivation is given to obtain an equivalent range-independent environmental model for fast inversion, instead of a range-dependent one. The received fields are modified using the Doppler-shifted wavenumbers. The simulations and results of the SWellEx-96 experimental data verify the effectiveness of the proposed inversion method.

Range-coherent matched field processing for low signal-to-noise ratio localization

Range-coherent matched field processing (MFP) coherently combines snapshots to localize a moving, narrowband source. This approach differs from existing MFP approaches that treat each snapshot as having a random phase due to both unknown motion through the medium and imprecise knowledge of the source frequency. Range-coherent MFP requires determination of the source phase acquired between snapshots. With that information, MFP can be applied to the cross-spectrum of snapshots acquired at different times, since relative phase between snapshots is determined by the medium properties, source location, and source velocity. Viewed another way, range-coherent MFP is simply MFP applied to a passive synthetic aperture formed from a moving source. The synthetic aperture geometry depends on source velocity, which is included in the MFP search space. Range-coherent MFP produces robust velocity estimates at low signal-to-noise ratio (SNR), which permits the use of a longer fast Fourier transform in pre-processing. The synthetic aperture array gain plus the increased input SNR afforded by the enhanced pre-processing significantly lowers the required signal level for successful localization. In data from the SWellEx-96 experiment, range-coherent MFP successfully localizes a source that is too quiet for conventional methods to localize.

Matched-field geoacoustic inversion based on radial basis function neural network

Multi-layer neural networks (NNs) are combined with objective functions of matched-field inversion (MFI) to estimate geoacoustic parameters. By adding hidden layers, a radial basis function neural network (RBFNN) is extended to adopt MFI objective functions. Specifically, shallow layers extract frequency features from the hydrophone data, and deep layers perform inverse function approximation and parameter estimation. A hybrid scheme of backpropagation and pseudo-inverse is utilized to update the RBFNN weights using batch processing for fast convergence. The NNs are trained using a large sample set covering the parameter interval. Numerical simulations and the SWellEx-96 experimental data results demonstrate that the proposed NN method achieves inversion performance comparable to the conventional MFI due to utilizing big data and integrating MFI objective functions.

Acoustic signal characterization based on hidden Markov models with applications to geoacoustic inversions

A probabilistic characterization scheme for acoustic signals with applications in acoustical oceanography is presented. This scheme aims at the definition of a set of stochastic observables that could characterize the signal. To this end, the signal is decomposed into several levels using the stationary wavelet packet transform. The extracted wavelet coefficients are then modeled by a hidden Markov model (HMM) with Gaussian emission distributions. The association of a signal with a representative HMM is performed utilizing the expectation-maximization algorithm. Eventually, the signal is characterized by the set of parameters that describe the HMM. The Kullback-Leibler divergence is employed as the similarity measure of two signals, comparing their corresponding HMMs. To validate the performance of the proposed characterization scheme, which is denoted as the probabilistic signal characterization scheme (PSCS), a simulated and a real experiment have been considered. The measured signal is characterized by the proposed PSCS method, and the model parameters of the seabed are estimated by means of an inversion procedure employing a genetic algorithm. The inversion results confirmed the reliability and efficiency of the proposed method when applied with typical signals used in applications of acoustical oceanography.

Multiple-array passive acoustic source localization in shallow water

This paper considers concurrent matched-field processing of data from multiple, spatially-separated acoustic arrays with application to towed-source data received on two bottom-moored horizontal line arrays from the SWellEx-96 shallow water experiment. Matched-field processors are derived for multiple arrays and multiple-snapshot data using maximum-likelihood estimates for unknown complex-valued source strengths and unknown error variances. Starting from a coherent processor where phase and amplitude is known between all arrays, likelihood expressions are derived for various assumptions on relative source spectral information (amplitude and phase at different frequencies) between arrays and from snapshot to snapshot. Processing the two arrays with a coherent-array processor (with inter-array amplitude and phase known) or with an incoherent-array processor (no inter-array spectral information) both yield improvements in localization over processing the arrays individually. The best results with this data set were obtained with a processor that exploits relative amplitude information but not relative phase between arrays. The localization performance improvement is retained when the multiple-array processors are applied to short arrays that individually yield poor performance.

Dictionary learning of sound speed profiles

To provide constraints on the inversion of ocean sound speed profiles (SSPs), SSPs are often modeled using empirical orthogonal functions (EOFs). However, this regularization, which uses the leading order EOFs with a minimum-energy constraint on the coefficients, often yields low resolution SSP estimates. In this paper, it is shown that dictionary learning, a form of unsupervised machine learning, can improve SSP resolution by generating a dictionary of shape functions for sparse processing (e.g., compressive sensing) that optimally compress SSPs; both minimizing the reconstruction error and the number of coefficients. These learned dictionaries (LDs) are not constrained to be orthogonal and thus, fit the given signals such that each signal example is approximated using few LD entries. Here, LDs describing SSP observations from the HF-97 experiment and the South China Sea are generated using the K-SVD algorithm. These LDs better explain SSP variability and require fewer coefficients than EOFs, describing much of the variability with one coefficient. Thus, LDs improve the resolution of SSP estimates with negligible computational burden.

Compressive acoustic sound speed profile estimation

Ocean acoustic sound speed profile (SSP) estimation requires the inversion of acoustic fields using limited observations. Compressive sensing (CS) asserts that certain underdetermined problems can be solved in high resolution, provided their solutions are sparse. Here, CS is used to estimate SSPs in a range-independent shallow ocean by inverting a non-linear acoustic propagation model. It is shown that SSPs can be estimated using CS to resolve fine-scale structure.

Change-point detection for recursive Bayesian geoacoustic inversions

In order to carry out geoacoustic inversion in low signal-to-noise ratio (SNR) conditions, extended duration observations coupled with source and/or receiver motion may be necessary. As a result, change in the underlying model parameters due to time or space is anticipated. In this paper, an inversion method is proposed for cases when the model parameters change abruptly or slowly. A model parameter change-point detection method is developed to detect the change in the model parameters using the importance samples and corresponding weights that are already 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 until the accumulated model parameter mismatch is significant and triggers the detection of a change-point. These change-point detections form the heuristics for controlling the coherent integration time in recursive Bayesian inversion. The method is demonstrated in simulation with parameters corresponding to the low SNR, 100-900 Hz linear frequency modulation pulses observed in the Shallow Water 2006 experiment [Tan, Gerstoft, Yardim, and Hodgkiss, J. Acoust. Soc. Am. 136, 1187-1198 (2014)].

Recursive Bayesian synthetic aperture geoacoustic inversion in the presence of motion dynamics

A low signal to noise ratio (SNR), single source/receiver, broadband, frequency-coherent matched-field inversion procedure recently has been proposed. It exploits coherently repeated transmissions to improve estimation of the geoacoustic parameters. The long observation time improves the SNR and creates a synthetic aperture due to relative source-receiver motion. To model constant velocity source/receiver horizontal motion, waveguide Doppler theory for normal modes is necessary. However, 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 data along with the ability to assess the temporal evolution of parameter uncertainty. Here 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. The method is demonstrated in simulation and in the analysis of low SNR, 100-900 Hz linear frequency modulated (LFM) pulses from the Shallow Water 2006 experiment.