Transverse Acoustic Variability Experiment Research Articles

Evaluation of Relative Entropy for Distributed Passive Detection of Weak Acoustic Signals

Passive detection of weak signals in noise is formulated without assuming a parametric distribution of the received signal statistics. A binary likelihood-ratio test as to whether the probability densities of two candidate noise populations are the same, or significantly different, leads to the comparison of relative entropy with a Neyman-Pearson detection threshold. The method is demonstrated using the acoustical intensity of narrowband signals at the hydrophones of a bottom-mounted horizontal line array, which was deployed during the Transverse Acoustic Variability Experiment. Receiver operating characteristics of the relative entropy detector are compared with those for the mean signal-to-noise ratio on a set of hydrophones. It is found that at large false-alarm rates the relative entropy detector has comparatively inferior performance while remaining competitive in the region of small false-alarm probability. Relative entropy detection is insensitive to statistical dependence among the sensors, but is sensitive to data having nonstationary intensity statistics characteristic of the ocean environment in which it was tested.

The effect of large- and small-scale sound velocity field structure on shallow water sound propagation in the East China Sea

During the Transverse Acoustic Variability Experiment (TAVEX) in 2008, towed CTD measurements were used to record the sound speed field over an area of the East China Sea surrounding 34- and 20-km acoustic paths. The data revealed several propagating internal wave packets along with sound speed variability due to density-compensated thermohaline variations (spice). Concurrent acoustic data were also recorded from 300- and 500-Hz sources. Parabolic equation modeling is used to examine the effects of the measured internal waves and spice on acoustic propagation and to make comparisons with the recorded acoustic data.

A relative-entropy approach to distributed passive detection

There is currently much interest within the ocean acoustics community on using distributed sensor networks to monitor ocean properties. One such task is the application of distributed sensors to passive detection of weak acoustic sources. The joint likelihood detection ratio for a set of distributed sensors leads naturally to the comparison of relative entropy with a detection threshold. As a nondimensional additive measure of information, relative entropy enables data fusion among disparate kinds of sensors, e.g., acoustic and electromagnetic, and mitigates calibration issues. Furthermore, because relative entropy is an integral over probability densities of sensor outputs, it is insensitive to false alarms from transients. In this talk, the theory of relative-entropy detection will be presented, and the method illustrated using acoustic intensity data from hydrophones deployed by the Transverse Acoustic Variability Experiment (TAVEX). For the method to work, accurate estimates of the probability densities for noise and signal intensities must be obtained. For the TAVEX data, superior receiver operating characteristic curves are obtained when the noise and signal distributions are represented by log-normal distributions in comparison with gamma and nonparametric distributions. [Work supported by the Office of Naval Research.]

Impact of fine-scale sound-speed fluctuations on acoustic autocorrelation times in the East China Sea

Autocorrelation times of acoustic signals propagating through shallow oceans are largely driven by sound-speed fluctuations. In August 2008, the Transverse Acoustic Variability Experiment obtained measurements in the East China Sea (65–80 m water depth) of fluctuating signals propagating 33 km from a moored source to a bottomed line array. Supporting environmental measurements were obtained by a towed conductivity-temperature-depth chain. For time periods without large nonlinear internal waves, the measured internal-wave power spectrum shows excess energy at high wavenumbers in comparison with the shallow-water internal-wave model of Levine. Likewise the associated sound-speed fluctuation spectrum exhibits high-wavenumber components in excess of a simple power law. Autocorrelation times of measured 300 Hz acoustic signals were compared with simulated times obtained using a parabolic-equation model to propagate acoustic fields through sound-speed fluctuations driven by linear internal-wave displacements. Results of the comparison are that median measured autocorrelation times (115 s) are shorter than simulations (300 s) driven by spectra with depleted high-wavenumber components. Simple frozen ocean simulations that translate the environment at 0.6 m/s produce autocorrelation times close to the data. [Work supported by the Office of Naval Research.]

Geoacoustic model of the transverse acoustic variability experiment area in the northern East China Sea

Transverse Acoustic Variability Experiment (TAVEX) is a joint ROK-US acoustic experiment for testing transverse acoustic variability in the northern limit of the East China Sea. TAVEX was performed in a water depth of 70–90 m around 32°30 to 40N, 125°20 to 30E, about 150 km southwest of Cheju Island, in August 2008. Purpose of this paper is to provide geoacoustic models of the TAVEX area, integrated and detailed as available. Two geoacoustic models of three layers were reconstructed, located at 32°40N/125°30E and 32°30N/125°30E of the northern East China Sea, respectively. The models were based on the data of the high-resolution subbottom and air-gun seismic profiles with cores. Geoacoustic data of the cores were extrapolated down to a depth of 100 m of the geoacoustic models. For actual underwater experiments, the P-wave speed of the models was compensated to in situ depth below the sea floor using the Hamilton method.

Properties of the acoustic intensity vector field in a shallow water waveguide

Acoustic intensity is a vector quantity described by collocated measurements of acoustic pressure and particle velocity. In an ocean waveguide, the interaction among multipath arrivals of propagating wavefronts manifests unique behavior in the acoustic intensity. The instantaneous intensity, or energy flux, contains two components: a propagating and non-propagating energy flux. The instantaneous intensity is described by the time-dependent complex intensity, where the propagating and non-propagating energy fluxes are modulated by the active and reactive intensity envelopes, respectively. Properties of complex intensity are observed in data collected on a vertical line array during the transverse acoustic variability experiment (TAVEX) that took place in August of 2008, 17 km northeast of the Ieodo ocean research station in the East China Sea, 63 m depth. Parabolic equation (PE) simulations of the TAVEX waveguide supplement the experimental data set and provide a detailed analysis of the spatial structure of the complex intensity. A normalized intensity quantity, the pressure-intensity index, is used to describe features of the complex intensity which have a functional relationship between range and frequency, related to the waveguide invariant. The waveguide invariant is used to describe the spatial structure of intensity in the TAVEX waveguide using data taken at discrete ranges.

Spectral effects of nonlinear internal waves on narrowband shallow-water signal propagation

Water column, bathymetry, and acoustical sediment properties collected during the transverse acoustic variability eXperiment (TAVEX) of 2008 are used to create a 30 km computational environment for propagation models. These models are used to display internal-wave-induced acoustic variability which is characteristic of the shallow-water northern East China Sea environment where the experimental work took place. Analyses of computational models are then compared to narrowband acoustic recordings made during the experiment using a spectral and time-of-arrival analysis approach. The expected acoustic effects and computational impact of specific internal wave characteristics on acoustic arrivals will be discussed along with recorded data comparison. [Work supported by the Office of Naval Research.]

Vertical intensity structure in a shallow water waveguide.

Acoustic intensity in an ocean waveguide is described by the local pressure and particle velocity, both of which can be described as a sum of modes. Analysis of the interaction between these modal components gives insight into the formation of characteristic intensity structures, such as interference patterns. Observations of the modal structure of the pressure and vertical velocity in a shallow water waveguide are presented using experimental data from an experiment off Korean coastal waters, the transverse acoustic variability experiment (TAVEX) that took place in August of 2008 17 km northeast of the Ieodo weather station, in waters 62 m deep. Mode filtering is performed on a 16 element vertical array that spans the water column (3 m spacing) for broadband (imploding light bulb) sources detonated at ranges from 200 to 1000 m at 40 m depth. The vertical velocity field, determined through the finite-difference approximation, and the pressure field at 230 Hz are represented by six propagating modes and their corresponding modal amplitudes. The interaction of these modal components is analyzed and PE simulated data are presented for comparison. Nondimensional indices are formulated relating the modal components of vector intensity and their utility as field indicators will be discussed. [Research supported by the U.S. Office of Naval Research and Agency for Defense Development, Korea.]

Interaction of acoustic modes with internal‐wave oscillations.

Acoustic modes propagating in shallow oceans interact with internal‐wave‐induced oscillations in the thermocline to produce fluctuations in the acoustic field that govern spatial and temporal coherence. Such interactions are described by both adiabatic‐mode path‐integral theory and coupled‐mode scattering‐matrix theory. In both theories there occurs an integral over the water depth that sums the interaction between the buoyancy, acoustic modes, and internal‐wave modes. This paper explores the strength of that interaction for the Transverse Acoustic Variability Experiment, which took place in the northern limit of the East China Sea in 65–80 m of water. A towed CTD chain provided measurements of the density and sound‐speed profiles required to compute the necessary acoustic and internal‐wave modes. Within this environment, two main conclusions pertain. First, with respect to internal‐wave mode contributions, the first mode contributes most significantly with a lesser contribution from the second mode and negligible contributions from higher modes. Second, with respect to acoustic mode contributions, the adiabatic mode contribution is most significant with lesser contributions from adjacent modes, and negligible contributions from widely separated mode pairs. [Work supported by the Office of Naval Research.]

Analysis of horizontal coherence during the transverse acoustic variability experiment.

The transverse acoustic variability experiment took place in the northern limit of the East China Sea in 65–80 m of water. 300-Hz and 500-Hz cw acoustic signals were transmitted over distances of 33 and 20 km to a broadside HLA where they were received at 22–27 dB SNR. Fourteen estimates of transverse mutual-coherence functions show high correlation with scale lengths on the order of 1000–1700 m. A towed CTD chain provided simultaneous measurements of the sound-speed fluctuation spectrum due to internal waves in an effort to determine whether the observed shallow-water coherence is understandable within the framework of a normal-mode formulation of path-integral theory. An important property of the mutual-coherence function is its behavior at small hydrophone separations. Here the phase-structure function is expected to follow a power law. Measured logarithmic phase-structure function have slopes that vary between 0.6 and 1.1, which disagree with path-integral solutions that predict slope 2. The discrepancy between observation and data is not currently understood. [Work supported by the Office of Naval Research.]

Adjoint study of water column variability and propagation in the transverse acoustic variability experiment (TAVEX 2008).

Internal wave and other water column activity in the conductivity, temperature, and depth (CTD) data collected in the East China Sea as a part of TAVEX 2008 are used as a basis for analyzing acoustic propagation at 500 Hz between a fixed source and fixed horizontal line array placed on the seafloor. Because the CTD data were not taken coincident with the acoustic propagation path (in general), adjoint modeling methods will be used to study acoustic data and prediction misfit. Usefulness of the adjoint method for these experimental conditions will be discussed. [Work supported by ONR.]

Performance of the transverse acoustic variability experiment horizontal line array.

The transverse acoustic variability experiment was designed to study the impact of shallow-water environmental fluctuations on acoustic coherence in the horizontal plane, transverse to the direction of propagation. The experiment was conducted during August 2008 in the East China Sea. A critical component to the success of the experiment was a two-segment 96-channel horizontal line array, moored on the sea floor at an approximate depth of 80 m. An analysis of the performance of this array is presented, including pulse arrival time statistics, beam patterns, gains, and signal-to-noise ratio. Methods used to determine the relative positions of array hydrophones are also discussed. [Work supported by the Office of Naval Research. We thank the staff at KORDI and the crews aboard the R/V Eardo and Sunjin for their assistance and bountiful hospitality.]