Terahertz Imaging of Thin Film Layers with Matched Field Processing.

Terahertz (THz) time of flight (TOF) tomography systems offer a new measurement modality for non-destructive evaluation (NDE) of the subsurface layers of protective coatings and/or laminated composite materials for industrial, security and biomedical applications. However, for thin film samples, the time-of-flight within a layer is less than the duration of the THz pulse and consequently there is insufficient range resolution for NDE of the sample under test. In this paper, matched field processing (MFP) techniques are applied to thickness estimation in THz TOF tomography applications, and these methods are demonstrated by using measured THz spectra to estimate the the thicknesses of a thin air gap and its depth below the surface. MFP methods have been developed over several decades in the underwater acoustics community for model-based inversion of geo-acoustic parameters. It is expected that this research will provide an important link for THz researchers to access and apply the robust methods available in the MFP literature.

Beamforming using subspace estimation from a diagonally averaged sample covariance.

The potential benefit of a large-aperture sonar array for high resolution target localization is often challenged by the lack of sufficient data required for adaptive beamforming. This paper introduces a Toeplitz-constrained estimator of the clairvoyant signal covariance matrix corresponding to multiple far-field targets embedded in background isotropic noise. The estimator is obtained by averaging along subdiagonals of the sample covariance matrix, followed by covariance extrapolation using the method of maximum entropy. The sample covariance is computed from limited data snapshots, a situation commonly encountered with large-aperture arrays in environments characterized by short periods of local stationarity. Eigenvectors computed from the Toeplitz-constrained covariance are used to construct signal-subspace projector matrices, which are shown to reduce background noise and improve detection of closely spaced targets when applied to subspace beamforming. Monte Carlo simulations corresponding to increasing array aperture suggest convergence of the proposed projector to the clairvoyant signal projector, thereby outperforming the classic projector obtained from the sample eigenvectors. Beamforming performance of the proposed method is analyzed using simulated data, as well as experimental data from the Shallow Water Array Performance experiment.

Performance metrics for depth-based signal separation using deep vertical line arrays.

A recent publication by McCargar and Zurk [(2013). J. Acoust. Soc. Am. 133(4), EL320-EL325] introduced a modified Fourier transform-based method for passive source depth estimation using vertical line arrays deployed below the critical depth in the deep ocean. This method utilizes the depth-dependent modulation caused by the interference between the direct and surface-reflected acoustic arrivals, the observation of which is enhanced by propagation through the reliable acoustic path. However, neither the performance of this method nor its limits of applicability have yet been thoroughly investigated. This paper addresses both of these issues; the first by identifying and analyzing the factors that influence the resolution and ambiguity in the transform-based depth estimate; the second by introducing another, much simpler depth estimation method, which is used to determine the target trajectories required for observation of the interference pattern and the array requirements for accurate depth estimation.

Eigenvector pruning method for high resolution beamforming.

This paper introduces an eigenvector pruning algorithm for the estimation of the signal-plus-interference eigenspace, required as a preliminary step to subspace beamforming. The proposed method considers large-aperture passive array configurations operating in environments with multiple maneuvering targets in background noise, in which the available data for estimation of sample covariances and eigenvectors are limited. Based on statistical properties of scalar products between deterministic and complex random vectors, this work defines a statistically justified threshold to identify target-related features embedded in the sample eigenvectors, leading to an estimator for the signal-bearing eigenspace. It is shown that data projection into this signal subspace results in sharpening of beamforming outputs corresponding to closely spaced targets and provides better target separation compared to current subspace beamformers. In addition, the proposed threshold gives the user control over the worst-case scenario for the number of false detections by the beamformer. Simulated data are used to quantify the performance of the subspace estimator according to the distance between estimated and true signal subspaces. Beamforming resolution using the proposed method is analyzed with simulated data corresponding to a horizontal line array, as well as experimental data from the Shallow Water Array Performance experiment.

A computational method to predict and study underwater noise due to pile driving.

A hybrid modeling approach that uses the parabolic equation (PE) with an empirical source model is presented to study and predict the underwater noise due to pile driving in shallow, inhomogeneous environments over long propagation ranges. The empirical source model uses a phased point source array to simulate the time-dependent pile source. The pile source is coupled with a broadband application of a PE wave propagation model that includes range dependent geoacoustic properties and bathymetry. Simulation results are shown to be in good agreement with several acoustic observations of pile driving in the Columbia River between Portland, OR and Vancouver, WA. The model is further applied to predict sound levels in the Columbia River and study the effects of variable bathymetry and sediment configurations on underwater sound levels.

An eigenvector-based test for local stationarity applied to array processing.

In sonar array processing, a challenging problem is the estimation of the data covariance matrix in the presence of moving targets in the water column, since the time interval of data local stationarity is limited. This work describes an eigenvector-based method for proper data segmentation into intervals that exhibit local stationarity, providing data-driven higher bounds for the number of snapshots available for computation of time-varying sample covariance matrices. Application of the test is illustrated with simulated data in a horizontal array for the detection of a quiet source in the presence of a loud interferer.

Depth-based signal separation with vertical line arrays in the deep ocean.

Deep vertical line arrays can exploit the reliable acoustic path (RAP), which provides low transmission loss (TL) for targets at moderate ranges, and increased TL for distant interferers. However, nearby surface interference also has favorable RAP propagation and cannot be separated from a submerged target without horizontal aperture. In this work, a physics-based Fourier transform variant is introduced, which achieves depth-based signal separation by exploiting the spatial structure resulting from the coherent addition of the direct and surface-reflected propagation paths present for submerged sources. Simulation results demonstrate depth-based signal separation without requiring knowledge of the ocean environment.

Automated extraction and classification of time-frequency contours in humpback vocalizations.

A time-frequency contour extraction and classification algorithm was created to analyze humpback whale vocalizations. The algorithm automatically extracted contours of whale vocalization units by searching for gray-level discontinuities in the spectrogram images. The unit-to-unit similarity was quantified by cross-correlating the contour lines. A library of distinctive humpback units was then generated by applying an unsupervised, cluster-based learning algorithm. The purpose of this study was to provide a fast and automated feature selection tool to describe the vocal signatures of animal groups. This approach could benefit a variety of applications such as species description, identification, and evolution of song structures. The algorithm was tested on humpback whale song data recorded at various locations in Hawaii from 2002 to 2003. Results presented in this paper showed low probability of false alarm (0%-4%) under noisy environments with small boat vessels and snapping shrimp. The classification algorithm was tested on a controlled set of 30 units forming six unit types, and all the units were correctly classified. In a case study on humpback data collected in the Auau Chanel, Hawaii, in 2002, the algorithm extracted 951 units, which were classified into 12 distinctive types.

Trans-dimensional geoacoustic inversion of wind-driven ambient noise.

This letter applies trans-dimensional Bayesian geoacoustic inversion to quantify the uncertainty due to model selection when inverting bottom-loss data derived from wind-driven ambient-noise measurements. A partition model is used to represent the seabed, in which the number of layers, their thicknesses, and acoustic parameters are unknowns to be determined from the data. Exploration of the parameter space is implemented using the Metropolis-Hastings algorithm with parallel tempering, whereas jumps between parameterizations are controlled by a reversible-jump Markov chain Monte Carlo algorithm. Sediment uncertainty profiles from inversion of simulated and experimental data are presented.

Striation-based beamforming for active sonar with a horizontal line array.

A physics-based method for beamforming signals measured on a horizontal array is developed with an application to underwater active sonar systems. The proposed striation-based beamformer coherently combines the pressure from each element in the array at different frequencies, and these frequencies are selected based on a striation hypothesis. The linear frequency shift and corresponding phase term introduced in the array weight vector accounts for multipath-induced fading, producing beam output with increased signal gain. The method is demonstrated using data collected on an array towed in the North Atlantic. The combination of the striation-based beamformer with the waveguide invariant concept to improve tracker performance is discussed.

Bayesian geoacoustic inversion using wind-driven ambient noise.

This paper applies Bayesian inversion to bottom-loss data derived from wind-driven ambient noise measurements from a vertical line array to quantify the information content constraining seabed geoacoustic parameters. The inversion utilizes a previously proposed ray-based representation of the ambient noise field as a forward model for fast computations of bottom loss data for a layered seabed. This model considers the effect of the array's finite aperture in the estimation of bottom loss and is extended to include the wind speed as the driving mechanism for the ambient noise field. The strength of this field relative to other unwanted noise mechanisms defines a signal-to-noise ratio, which is included in the inversion as a frequency-dependent parameter. The wind speed is found to have a strong impact on the resolution of seabed geoacoustic parameters as quantified by marginal probability distributions from Bayesian inversion of simulated data. The inversion method is also applied to experimental data collected at a moored vertical array during the MAPEX 2000 experiment, and the results are compared to those from previous active-source inversions and to core measurements at a nearby site.

Striation-based beamforming for estimating the waveguide invariant with passive sonar.

The waveguide invariant summarizes the pattern of constructive and destructive interference between acoustic modes propagating in the ocean waveguide. For many sonar signal-processing schemes, it is essential to know the correct numerical value for the waveguide invariant. While conventional beamforming can estimate the ratio between the waveguide invariant and the range to the source, it cannot unambiguously separate the two terms. In the present work, striation-based beamforming is developed. It is shown that the striation-based beamformer can be used to produce an estimate for the waveguide invariant that is independent of the range. Simulation results are presented.

Extraction of small boat harmonic signatures from passive sonar.

This paper investigates the extraction of acoustic signatures from small boats using a passive sonar system. Noise radiated from a small boats consists of broadband noise and harmonically related tones that correspond to engine and propeller specifications. A signal processing method to automatically extract the harmonic structure of noise radiated from small boats is developed. The Harmonic Extraction and Analysis Tool (HEAT) estimates the instantaneous fundamental frequency of the harmonic tones, refines the fundamental frequency estimate using a Kalman filter, and automatically extracts the amplitudes of the harmonic tonals to generate a harmonic signature for the boat. Results are presented that show the HEAT algorithms ability to extract these signatures.

Experimental observations of active invariance striations in a tank environment.

The waveguide invariant in shallow water environments has been widely studied in the context of passive sonar. The invariant provides a relationship between the frequency content of a moving broadband source and the distance to the receiver, and this relationship is not strongly affected by small perturbations in environment parameters such as sound speed or bottom features. Recent experiments in shallow water suggest that a similar range-frequency structure manifested as striations in the spectrogram exists for active sonar, and this property has the potential to enhance the performance of target tracking algorithms. Nevertheless, field experiments with active sonar have not been conclusive on how the invariant is affected by the scattering kernel of the target and the sonar configuration (monostatic vs bistatic). The experimental work presented in this paper addresses those issues by showing the active invariance for known scatterers under controlled conditions of bathymetry, sound speed profile and high SNR. Quantification of the results is achieved by introducing an automatic image processing approach inspired on the Hough transform for extraction of the invariant from spectrograms. Normal mode simulations are shown to be in agreement with the experimental results.

Radiative transfer theory applied to ocean bottom modeling.

Research on the propagation of acoustic waves in the ocean bottom sediment is of interest for active sonar applications such as target detection and remote sensing. The interaction of acoustic energy with the sea floor sublayers is usually modeled with techniques based on the full solution of the wave equation, which sometimes leads to mathematically intractable problems. An alternative way to model wave propagation in layered media containing random scatterers is the radiative transfer (RT) formulation, which is a well established technique in the electromagnetics community and is based on the principle of conservation of energy. In this paper, the RT equation is used to model the backscattering of acoustic energy from a layered elastic bottom sediment containing distributions of independent scatterers due to a constant single frequency excitation in the water column. It is shown that the RT formulation provides insight into the physical phenomena of scattering and conversion of energy between waves of different polarizations.

Demonstration of the invariance principle for active sonar.

Active sonar systems can provide good target detection potential but are limited in shallow water environments by the high level of reverberation produced by the interaction between the acoustic signal and the ocean bottom. The nature of the reverberation is highly variable and depends critically on the ocean and seabed properties, which are typically poorly known. This has motivated interest in techniques that are invariant to the environment. In passive sonar, a scalar parameter termed the waveguide invariant, has been introduced to describe the slope of striations observed in lofargrams. In this work, an invariant for active sonar is introduced. This active invariant is shown to be present in the time-frequency structure observed in sonar data from the Malta Plateau, and the structure agrees with results produced from normal mode simulations. The application of this feature in active tracking algorithms is discussed.

Source motion mitigation for adaptive matched field processing.

Application of adaptive matched field processing to the problem of detecting quiet targets in shallow water is complicated by source motion, both the motion of the target and the motion of discrete interferers. Target motion causes spreading of the target peak, thereby reducing output signal power. Interferer motion increases the dimensionality of the interference subspace, reducing adaptive interference suppression. This paper presents three techniques that mitigate source motion problems in adaptive matched field processing. The first involves rank reduction, which enables adaptive weight computation over short observation intervals where motion effects are less pronounced. The other two techniques specifically compensate for source motion. Explicit target motion compensation reduces target motion mismatch by focusing snapshots according to a target velocity hypothesis. And time-varying interference filtering places time-varying nulls on moving interferers not otherwise suppressed by adaptive weights. The three techniques are applied to volumetric array data from the Santa Barbara Channel Experiment and are shown to improve output signal-to-background-plus-noise ratio by more than 3 dB over the standard minimum-variance, distortionless response adaptive beam-former. Application of the techniques in some cases proves to be the difference between detecting and not detecting the target.