Aperture Vertical Array Research Articles

Azimuthal, range, and depth tracking of marine mammal sounds using a drifting acoustic vector sensor and hydrophone array platform

Deep-water acoustic tracking of marine mammals typically requires correlating hydrophone signals across both short and large-aperture hydrophone arrays. Here, we demonstrate how a single drifting, depth-controlled platform can obtain two and three-dimensional positional fixes of marine mammal sounds using an acoustic vector sensor and short-aperture arrays, both mounted on an autonomous drifting platform. In February and June 2023 two autonomous opto-acoustic drifters were deployed 50km off San Diego, CA at 250 m depth in 1030 m deep water. Both drifters were equipped with a CTD and acoustic recording system comprised of a 1.75 m aperture vertical hydrophone array, a tetrahedral hydrophone array, and either a 2-D Geospectrum M-35 or 3-D Wilcoxon VS-209 acoustic vector sensor. Throughout the two to three-day deployments, all acoustic sensors detected numerous marine mammal calls, including humpback whales and a pod of common dolphins. Estimates of azimuth and elevation were obtained from the simultaneously sampled acoustic vector sensor and hydrophone arrays, while multi-path and cross-platform processing were employed to extract range information. The results suggest that sparse deployments of drifting vector sensor platforms may be able to map biological distributions over spatial scales that would otherwise require larger numbers of hydrophone-only platforms. [Work supported by ONR TFO.]

Extracting Green's functions between ships of opportunity using a vertical array.

A theoretical method for estimating the Green's function between two points in an acoustic waveguide was proposed using a vertical source array that spans sufficient waveguide depth [Roux and Fink, J. Acoust. Soc. Am. 113, 1406-1416 (2003)]. This paper shows that by reversing the role of sources and receivers, the Green's function between two ships (sources) can be extracted using a vertical receiver array with a limited aperture. First, the Green's functions from each ship are estimated along the array via blind deconvolution. Then the Green's function between two ships is obtained by either correlation or convolution of the individual Green's functions summed over the array, depending on the array position with respect to the ships. The feasibility of extracting Green's functions between ships of opportunity radiating random broadband (100-500 Hz) noise is demonstrated using a 56.25-m aperture vertical array in approximately 100-m shallow water.

Inference regarding the state of a mobile underwater object from narrow band observations on a small aperture vertical array

A narrowband source ensonifies an area of interest in a refractive undersea environment. A small aperture vertical array is employed to infer the depth, range and speed of the acoustic scatterer by resolving the scattered direct and surface interacting wave fronts. Tracking the target by means of a continuous wave transmission is challenging due to the difficulty of inferring the frequencies and angles of the two returned closely spaced wave vectors. The propagation of sound in a refractive medium presents additional challenges for inversion of the wave vectors to range, depth, and speed. Joint posterior inference is made possible with a computational Bayesian Gibbs sampling scheme over all track parameters taking full advantage of the analytic tractability of the conditional densities of the received amplitudes and phases and of the ambient acoustic noise power. The conditional densities of the ordered wave vectors however are constructed numerically by 2-dimensional inverse quantile sampling. The inferr...

Inference regarding the state of mobile underwater object from a small aperture vertical array

A narrowband source ensonifies an area of interest with an oncoming submerged target in a refractive undersea environment. A vertical linear hydrophone array is employed to infer the depth, range, and speed of the target by resolving a direct path and one interacting surface path. Tracking the target by means of a continuous wave transmission is challenging due to the difficulty of inferring the frequencies and angles of the two returned closely spaced wave vectors. The complex propagation of sound in a refractive medium presents additional challenges for inversion of the wave vectors to range and depth. A Gibbs sampler is employed to construct the posterior joint density of all parameters taking full advantage of the analytic tractability of the conditional densities of the received amplitudes and phases and of the ambient acoustic noise power. The conditional densities of the ordered wave vectors however are constructed numerically by 2-dimensional inverse quantile sampling. The inferred joint posterior density of the target state is obtained by constructing an inverse transformation of the acoustic propagation model that accounts for the depth dependent sound speed. Simulation results demonstrate the approach at received signal to noise ratios below −3dB.

Extension of the array invariant to deep-water environments

The array invariant utilizes multiple arrivals separated in beam angle and travel time for broadband signals, achieving robust source-range estimation without detailed knowledge of the environments in shallow water. This approach includes the waveguide invariant parameter β, which is close to one for surface/bottom-interacting shallow-water environments. In deep water, however, β significantly varies from large negative to positive values depending on the group of modes, while being sensitive to the sound speed variation within the water column. In this paper, two different approaches are investigated: 1) an extension of the array invariant by integrating multiple arrivals with varying β and 2) the use of dispersion relationship between several pairs of arrivals. These two approaches will be applied to a deep-water environment (4-km deep) using a 13-m aperture vertical array deployed to the sound channel axis (e.g., 330-m) from the Medium Frequency Noise experiment (MFN), where a high-frequency source (above 1 kHz) also was positioned around the sound channel axis.

Applying double-difference methods for fine-scale acoustic tracking using a short aperture vertical array

Ray tracing can estimate an acoustic source’s depth and range in a waveguide by exploiting multipath arrival information on a vertical array. However, environmental mismatch in the model or array tilt can yield highly scattered trajectories when ray tracing multiple events. “Double-difference” methods have been used to localize earthquakes (Waldhauser and Ellsworth, 2000) and fin whales (Wilcock, 2012) by determining the relative locations of multiple events, rather than their absolute positions. This approach, which exploits changes in relative travel times between events, has recently been reformulated to recover the dive trajectory of a source using a single multi-hydrophone vertical array, whenever three acoustic rays are available for each event. Here, the method is expanded such that changes in ray elevation angles between events can be used to reduce the number of rays required. This technique is tested on data recorded on a short aperture vertical array off the coast of Southern California in 4 km deep water. Trajectories from both a controlled towed source and sperm whale dives are examined. [Work supported by Office of Naval Research—Marine Mammals and Biology and Ocean Acoustics Program.]

Applying double-difference methods for fine-scale acoustic tracking of controlled sources and sperm whales using a small aperture vertical array

Ray propagation modeling can estimate a source’s depth and range in a waveguide by exploiting multipath arrival information on a vertical array. However, environmental mismatch of the model, array tilt, and limited angular resolution of an array can yield highly scattered dive trajectories when ray tracing individual events. “Double-difference” methods have been used to localize earthquakes (Waldhauser and Ellsworth, 2000) and fin whales (Wilcock, 2012) by determining the location of multiple events relative to each other, rather than their absolute position. These same concepts can be reformulated into a “triple-difference” approach to track successive acoustic events on a single multi-hydrophone array. This method examines relative changes in the multipath arrival times and elevation angles over the course of a dive in order to establish a more robust track in terms of relative positions along the trajectory. Presented here are results of applying this new technique on both a towed source and sperm whales, using acoustic data recorded on a short aperture vertical array off the coast of Southern California in 4 km deep water. [Work supported by Office of Naval Research—Marine Mammals and Biology and Ocean Acoustics Program.]

The effects of internal tides on phase and amplitude statistics in the Philippine Sea

Moored oceanographic sensors and satellite altimetry has revealed energetic diurnal and semi-diurnal internal tides in the Western Philippine Sea. Because the internal tides have a complex spatio-temporal pattern and large vertical displacements, these waves have the potential for causing strong acoustic variability. This talk will present a tidal analysis of signal fluctuations from the PhilSea09 experiment in which broadband signals with a center frequency of 275 Hz and a bandwidth of 50 Hz were transmitted at the sound channel axis to a large aperture vertical array 180-km distant. Signal phase and amplitude statistics along distinct branches of the observed wavefronts will be analyzed and compared to ray-based model predictions using internal tide information obtained from moored oceanographic instruments at the source and receiver. Key issues are the acoustic effects of the internal tide nonlinearity, temporal stability, high mode structure, and complex horizontal interference patterns.

Extending passive acoustic capabilities of autonomous recorders by using multiple hydrophones.

Low-power data acquisition systems have attained sampling rates large enough to enable multielement hydrophone arrays to be deployed autonomously, at the cost of reducing the individual sampling rate per phone. This paper discusses what theoretical advantages bandlimited multielement recordings can provide over single-hydrophone data, including array gain for increased detection range, interfering noise source rejection, environmental inversion, and biological source tracking. These points are illustrated with data collected from an autonomous four-element vertical array off Queensland, Australia in 2003, and data collected from an autonomous eight-element, 21 m aperture vertical array deployed at 35 m depth in the Beaufort Sea in 2008. These examples also provide insight into the technical and logistical challenges required by such deployments, including deployment and recovery systems that do not endanger or entangle array cables. [Work supported by NPRB, logistics provided by Greeneridge Sciences and Shell Co.]

Evolution of a downslope propagating acoustic pulse

During the North Pacific Acoustics Laboratory (NPAL‐04) experiment broadband transmissions (50 Hz bandwidth, 75 Hz center frequency, 27.28 s period length) from a bottom‐mounted source off the island of Kauai were received on a towed array (350 m depth, 200 m aperture) near the island and a large aperture vertical line array (VLA) at a range of 2469.5 km. Previous work [Heaney, J. Acoust. Soc. Am. 117(3), 1635–1642 (2005)] has demonstrated that bottom interaction near the source is a fundamental process in the evolution of the wave‐train. In this paper results from eight receptions at ranges from 2 to 300 km and the reception on the distant VLA is used to examine the wavefront evolution as it propagates into deep water. Results from a local geo‐acoustic inversion will be presented as well as comparison of data with broadband parabolic equation simulations. The initial conclusion is that the first‐order bottom bounce, off the seamount near the source, acts as an image source, contributing to a complex arrival pattern even at great distances.

Estimating mutual information for the underwater acoustic channel

Estimates of the upper and lower bounds on the information rates attainable for acoustic communication systems with white Gaussian sources under uncertain channel conditions are provided that account for finite duration signaling and finite coherence time. For the case where the coherence time of the channel exceeds that of the packet duration the estimators are computationally efficient with the Levinson recursion and provide a means of tuning the accuracy of the estimates to fit a given computational budget. The reduction in the information rate relative to the known channel case is quantified in terms associated with the duration of signaling, the posterior covariance of the channel, and the prior covariance of the channel. For scenarios where the prior channel covariance is known exact bounds on the information rates are computable else the estimators provide confidence intervals for these rates. The estimators are tested on 18 kHz at sea data collected in shallow water north of Elba Italy on a 1.8-m aperture vertical array and on very shallow water sites at Keyport Washington. The effects of temporal coherence degradation as well as multi-path spread are quantified. [Work supported by the Office of Naval Research.]

Performance comparisons between passive-phase conjugation and decision-feedback equalizer for underwater acoustic communications

Passive-phase conjugation (PPC) uses passive time reversal to remove intersymbol interferences (ISI) for acoustic communications in a multipath environment. It is based on the theory of signal propagation in a waveguide, which says that the Green function (or the impulse response function) convolved with its time-reversed conjugate, summed over a large aperture vertical array of receivers is a delta function in space and time. A decision feedback equalizer (DFE) uses a nonlinear filter to remove ISI based on the minimum mean square errors (MSE) between the estimated and the true (or decision) symbols. These two approaches are motivated by different principles. In this paper, we analyze both using a common framework, illustrating the commonality and differences, pro and con between the two methods, and compare their performances in realistic ocean environments. The performance measures are MSE, output signal-to-noise ratio and bit error rate (BER) as a function of the number of receivers. For a small number of receivers, DFE outperforms PPC in all measures. As the number of receivers increases the BER for both processors approaches zero, but at a different rate. The results are supported by both simulated and real data. [Work supported by ONR.]

BROADBAND MATCHED-FIELD LOCALIZATION PERFORMANCE IN UNCERTAIN ENVIRONMENTS USING A SHORT ARRAY

Typical applications of matched-field localization use low frequency signals received on large aperture vertical-arrays. However, small aperture arrays are much more practical for real-world systems and must be considered. Additionally, any practical localization algorithm must also be robust to environmental mismatch. In this paper, we present the broadband L∞-norm estimator for robust matched-field localization of mid-frequency signals (e.g., 800–4000 Hz) received on very short aperture vertical-arrays. Realistic simulation results are presented using broadband signals in the band of 1000–3000 Hz received on a 3 m vertical array which demonstrate the substantial performance gains in using the L∞-norm estimator over the asymptotically-optimal maximum a posteriori estimator and the conventional Bartlett processor. Experimental data results in an uncertain shallow-water environment using a 2.13 m vertical array in the band of 3000–4000 Hz are also presented.

Broadband Matched-Field Localization Performance in Uncertain Environments Using a Short Array

Typical applications of matched-field localization use low frequency signals received on large aperture vertical-arrays. However, small aperture arrays are much more practical for real-world systems and must be considered. Additionally, any practical localization algorithm must also be robust to environmental mismatch. In this paper, we present the broadband L∞-norm estimator for robust matched-field localization of mid-frequency signals (e.g., 800–4000 Hz) received on very short aperture vertical-arrays. Realistic simulation results are presented using broadband signals in the band of 1000–3000 Hz received on a 3 m vertical array which demonstrate the substantial performance gains in using the L∞-norm estimator over the asymptotically-optimal maximum a posteriori estimator and the conventional Bartlett processor. Experimental data results in an uncertain shallow-water environment using a 2.13 m vertical array in the band of 3000–4000 Hz are also presented.

Range-dependent multitone matched-field processing in SWellEX-3

SWellEX-3 was an experimental study of shallow-water propagation and matched-field processing conducted 10 km off San Diego in July, 1994. Data were recorded on a 64-element, 118-m aperture vertical line array anchored in 200 m of water. A cw source emitting tonals from 53–197 Hz was towed over tracks where the water depth varied from 200–75 m. Examples of range-dependent matched-field processing of this multitone signal using the finite-element parabolic equation program for propagation modeling will be presented.

Direct measurement and matched-field inversion approaches to array shape estimation

Accurate knowledge of array shape is essential for carrying out full wavefield (matched-field) processing. Direct approaches to array element localization (AEL) include both nonacoustic (tilt-heading sensors) and acoustic (high-frequency, transponder-based navigation) methods. The low-frequency signature emitted from a distant source also can be used in an inversion approach to determine array shape. The focus of this paper is on a comparison of the array shape results from these three different methods using data from a 120-m aperture vertical array deployed during SWellEx-3 (Shallow Water evaluation cell Experiment 3). Located 2 m above the shallowest array element was a self-recording package equipped with depth, tilt, and direction-of-tilt sensors, thereby permitting AEL to be performed non-acoustically. Direct AEL also was performed acoustically by making use of transponder pings (in the vicinity of 12 kHz) received by high-frequency hydrophones spaced every 7.5 m along the vertical array. In addition to these direct approaches, AEL was carried out using an inversion technique where matched-field processing was performed on a multitone (50-200 Hz), acoustic source at various ranges and azimuths from the array. As shown, the time-evolving array shape estimates generated by all three AEL methods provide a consistent picture of array motion throughout the 6-h period analyzed.

Experimental spatiotemporal signal correlation functions in a shallow water environment (SWellEX-1)

During 2 weeks in August 1993, a high-fidelity acoustic experiment, SWellEX-1, was conducted in shallow waters off San Diego, CA. The SWellEX-1 site is a 50-to 200-m bottom-limited shallow water environment characterized by thin sediments and strong range dependence. During the course of the experiment extensive synoptic oceanographic measurements were collected. The site’s bathymetry and geoacoustic bottom properties are characterized on a 2-arcsecond lat/lon grid. Broadband and tonal signals (50–750 Hz) from calibrated projectors were received on a 48-channel, 90-m aperture vertical line array (VLA) spanning the lower half of the water column. Source tows were conducted along the isobath and up-slope relative to the VLA. Experimental results of signal propagation and spatiotemporal vertical correlation functions will be presented along with high-fidelity full-wave simulations. The impact of a strongly bottom-limited shallow water environment on signal correlation and the resulting effect on plane-wave signal processing schemes will be investigated. Relative signal phase stability results will be presented and its impact on inverse synthetic aperture processing discussed. [Work supported by ONR under NRaD block CS3A.]

Matched-field processing and source depth localization in a shallow-water environment (SWellEX-1)

During 2 weeks in August 1993, a high-fidelity acoustic experiment, SWellEX-1, was conducted in shallow waters off San Diego, CA. The SWellEX-1 site is a 50 to 200-m bottom-limited shallow-water environment characterized by strong range dependence in both bottom and water column properties. Calibrated broadband and tonal signals from a projector in the band 70–750 Hz were received on a 48-channel, 90-m aperture vertical line array (VLA) spanning the lower half of the water column. Source tows are depths from 15–100 m were conducted along isobaths and up-slope relative to the VLA. Results of source depth localization using plane-wave processing (omega-k plots) and matched-field tracking will be presented. Depth localization via matched-field processing with high-fidelity replica vectors generated using both range-dependent and range-independent models will be compared. The impact of bottom geoacoustic properties ‘‘mismatch’’ on matched-field processing will be discussed. [Work supported by ONR under NRaD block CS3A.]

Experimental demonstration of sound-speed inversion with matched-field processing

Matched-field processing is shown to be effective for estimating sound speed in a deep, range-independent ocean environment. The amplitude and phase of signals from a distant source measured on a large aperture vertical array are sensitive to changes in sound speed. This sensitivity is exploited to infer the environment’s sound-speed profile by matching predicted and measured amplitude and phase. Simulations and an initial experimental demonstration of the power of the technique are based on a modified version of Munk’s canonical sound field model. This model is used in a simulated environment with a 15-Hz source where source and receiver locations are known. The simulation demonstrates that changes in the depth and strength of the modeled sound channel axis directly result in trackable errors in range and depth as well as in reduction of the received power level. The same model is used to determine sound-speed profile using a 15-Hz signal from a 244-m explosive source detected on a 675-m vertical array at a range of 53 km in a deep water Pacific environment, characterized by a classical range-independent sound channel at 700 m. The search for the best estimated profile is conducted by varying the sound channel axis strength and depth in the modified Munk equation, while maintaining a constant sound-speed profile at great depths. For the single test case, resultant differences between the matched-field estimated and measured profiles are less than ±2 m/s; the sound channel axis depth is determined within 20 m of the measured axis depth.

A normal-mode based approach for localization of transient signals

Normal-mode modeling is a well-accepted representation for acoustic signals propagating in a number of environmental conditions. Detection of the spatial normal-mode structure makes possible signal localization enhancement against the noise that has no spatial structure. A simple algebraic argument allows one to separate the vector subspace that contains the signal from the vector subspace that contains the noise and to obtain a narrow-band estimate of the source location [S. M. Jesus, Signal Process. 28, 117–122 (1992)]. This subspace splitting algorithm has been extended for localizing broadband transient signals assuming that the signal only has a normal-mode structure. It is shown with synthetic data that the proposed broadband algorithm outperforms both the generalized minimum variance and the conventional processors. As an example, this processor has been used to localize short transient pulses collected in a 120-m depth shallow water area with a 62-m aperture vertical array. The experimental results show that stable and accurate localizations could be obtained during long time intervals. This shows that the sound field, received over a given frequency band, is relatively stable over time and is in agreement with the predictions given by a standard normal-mode propagation model.