Studies on mode feature extraction and source range and depth estimation with a single hydrophone based on the dispersion characteristic

Range estimation based on the array invariant concept exploits the inherent properties of shallow-water waveguide propagation. Multipath arrivals observed on a vertical line array, indexed on vertical arrival angle and time difference of arrival, form what is known as a beam migration pattern in the shape of an ellipse that is a function of source range and the acoustic invariant beta. Fitting the beam migration pattern to an ellipse yields the source range. A property of this beam migration pattern is that it is independent of source depth. Source depth determines how the multipath arrivals are distributed along that ellipse. We will extend previous work on estimating range and present how to estimate depth. This decouples the two estimates and reduces a 2D search process to two 1D processes. We will explore the CRLB structure of the range and depth estimates and demonstrate the results of processing experiment data from the RADAR 2007 experiment in shallow water off the coast of Portugal.

The polarization of the acoustic field in the ocean waveguide environment is a unique property that can be measured by using a particle velocity sensor in the water column. It can provide new ideas for locating and detecting the underwater target, so it is interesting to study the polarization. The polarization of a monochromatic signal has been described by the Stokes parameters, a set of four real-valued quantities in previous work. In this work, the Stokes parameters are extended to the broadband form, and the expression is simplified by using the nonstationary phase approximation, which reduces the complexity of the theoretical derivation and reveals the physical mechanism behind the significant variations in polarization with source depth and symmetrical depth. Theoretical analysis shows that the polarization characteristics in the ideal waveguide vary significantly in the sea surface, the sea bottom, the depth of the sound source and symmetrical depth. In this work the numerical simulation is used to verify the theoretical analysis and study the relationship between range and integral bandwidth when nonstationary phase approximation method is effective. The numerical results demonstrate that the simplified expression using the nonstationary phase approximation is effective and can better characterize the depth distribution characteristics of the polarization. Additionally, by normalizing the broadband Stokes parameters, the effect of range on the depth distribution characteristics of polarization can be removed. It means that the normalized broadband Stokes parameters are in theory free of the range and depend on the environment, the receiver depth and the source depth, which have the potential to be used for source depth estimation. Subsequently, focusing on normalized broadband Stokes parameters, we analyzes the effects of parameters such as source frequency, source depth, sound speed profile and water depth on the depth distribution characteristics of polarization. The analysis results show that environmental factors have great influence on the depth distribution characteristics of polarization. In the end, the validity of the nonstationary phase approximation and the range-independent property of the normalized broadband Stokes parameters are verified by the results of the RHUM-RUM experimental data processing. The findings provide a theoretical basis for passive target depth estimation based on polarization.

Experimental data from the SACLANTCEN 1993 Mediterranean Experiment are reviewed to assess the reduction of the search space for the localization and tracking of an acoustic source in a three-dimensional environment. Key to this goal is the availability of an initial estimate of source range and depth (called the 2D initial guess); an ambiguous estimate of source bearing can be obtained from the 2D initial guess through Environmental Signal Processing, and the ambiguity can be removed by searching for the source only in the range/bearing regions where bearing estimates are higher. This search provides a new estimate of source range and a single bearing, which together with the estimate for source depth constitute the center of the reduced search space for source localization and tracking. The suggested approach is tested on experimental data from the SACLANTCEN experiment considering different frequencies, as well as a stationary and a moving source.

Existing methods for source localization in an ocean waveguide generally require simultaneous range and depth estimation as they are intrinsically coupled in the acoustic field. Using beam intensities averaged over ranges longer than the mode cycle distances, it is shown that the range-averaged vertical beam distribution is insensitive to the source range and sensitive to the source depth. An incoherent matched beam processing method is presented and applied to the SWellEx96 moving source data to estimate source depth based on the 127 and 130 Hz signals with good results.

We estimate yields and source depths for the six North Korean underground nuclear explosions (UNEs) in 2006, 2009, 2013, 2016 (January and September), and 2017, based on regional seismic observations in South Korea. Spectral ratios of event pairs are calculated using seismograms from the six UNEs observed along the same propagation paths and at the same receivers. These relative seismic source scaling spectra for Pn provide a basis for a grid search source solution that estimates source yields and depths for each event pair based on assumed explosion source models by Mueller and Murphy (1971; MM71), Denny and Johnson (1991; DJ91), and Walter and Ford (2018; WF18). The grid search is used to identify the best fit to the empirical spectral ratios subject to the source models by minimizing the root mean square misfit in the frequency range of 0.2–15 Hz. To address the trade-off between depth and yield, a modified grid search was implemented that includes elastic propagation effects for different source depths using reflectivity Green’s functions, thus modeling slight differences in propagation path based on source depth. This addition reduces trade-offs between depth and yield, and results in better model fits to frequencies as high as 15 Hz compared with cases in which depth effects were not included. The modified grid search results indicate that both MM71 and WF18 models provide comparable source depth and yield estimates with good agreement between theoretical and observed spectral ratios matching both the long-period levels and the corner frequencies, whereas the DJ91 model estimates produce lower yields due to a difference in corner frequency scaling. The best grid search solutions produce yields from ∼0.9 kt for the first UNE and up to ∼290 kt for the explosion in 2017, with depths varying from ∼280 to ∼750 m.

Depth and structural index estimation of 2D magnetic source using correlation coefficient of analytic signal

Wireless sensor network (WSN) localisation has attracted significant research interest. The quality of time-of-flight ranging when used as the basis of localisation, has a direct impact on precision and accuracy. Recently significant attention has been devoted to modelling and analysis of range estimation error (REE) in indoor and cluttered environments. A Gaussianity assumption for the distribution of REE is a common practice in the literature. The motivation of this study is to analyse this assumption. To scrutinise this, rather than relying on computer generated data, a real IEEE 802.15.4 compliant WSN test-bed is used for collecting ranging data, covering outdoor and indoor environments for both line-of-sight and non-line-of-sight propagation conditions. The distribution of REE is analysed using both graphical and numerical goodness-of-fit (GOF) techniques, that is, quantile–quantile plotting, empirical cumulative distribution function plotting, probability density function plotting, linear correlation coefficient (γ) test, kurtosis (K) test, skewness (S) test, Anderson–Darling (A 2) test and chi-squared (χ 2) test. The GOF statistical analysis of the experimental results suggest that REE is not Gaussian distributed. A novel means of enhancement called the range filtration algorithm (RFA) is proposed. The RFA is based on the A 2 test, it filters out the range estimates with high errors.

Source depth estimation is always a problem in underwater acoustic area, because depth estimation is a nonlinear problem. Traditional depth estimation methods use a vertical line array, which has disadvantage of poor mobility due to the size of sensor array. In order to estimate source depth with a horizontal line array, we propose a matched-mode depth estimation method based on autoregressive (AR) wavenumber estimation for a moving source in shallow water waveguides. First, we estimate the mode wavenumbers using the improved AR modal wavenumber spectrum. Second, according to the mode wavenumber estimation, we estimate the mode amplitudes by the wavenumber spectrum, which is obtained by generalized Hankel transform. Finally, we estimate source depth estimation by the peak of source depth function wherein the data mode best matches the replica mode that is calculated using a propagation model. Compared with synthetic aperture beamforming, the proposed method exhibits a better performance in source depth estimation under low signal-to-noise ratio or the small range span. The robustness of the proposed method is illustrated by simulating the performance in mismatched environment.

Enhanced Cohen class time–frequency methods based on a structure tensor analysis: Applications to ISAR processing

In this study, we analyse both ground gravity and aeromagnetic data in order to delineate structural trends, fault systems and deduce sedimentary thicknesses within the Ajdabiya Trough in Libya’s northeast. A high-pass filter and a reduced-to-the-pole (RTP) transformation are applied to the gravity and aeromagnetic data respectively. Different filters are used to enhance the structural signatures and fault trends within the study area. The Werner deconvolution and source parameter imaging (SPI) techniques are applied to the RTP magnetic data for source depth estimation. Four well-data within the area are used as constraints in the two-dimensional forward modelling process. The results show that the Ajdabiya Trough is characterised by gravity anomaly highs and magnetic anomaly lows. The analysis of gravity data shows predominant Northeast–Southwest structural trends, whereas the analysis of magnetic data shows predominant North–South magnetic lineaments within the Ajdabiya Trough. The Euler deconvolution depth estimates of faults depths range between 1500 and 9500 m. The SPI estimates of the magnetic basement range between 2500 and 11,500 m beneath the study area (deepest beneath the Ajdabiya Trough). Constrained by the well-data, six major layers characterize the four profiles that are taken within the area. One of the profiles shows a high-density intrusion (about 4 km from the surface) within the sedimentary sequence. The intrusion may be the result of the rifting Sirt Basin which caused a weakening of the crust to allow for mantle intrusion.

The aperture of the towed horizontal line array is limited and the received signal is unstable in a complex ocean environment, making it difficult to distinguish the location of the sound source. To address this challenge, this paper presents a MoELocNet (Mixture of Experts Localization Network) for deep-sea sound source localization, leveraging interference structures in range-frequency domain signals from a towed horizontal line array. Unlike traditional correlation-based methods constrained by time-varying ocean environments and low signal-to-noise ratios, the model employs multi-expert and multi-task learning to extract interference periods from single-frame data, enabling robust estimation of source range and depth. Simulation results demonstrate its superior performance in the deep-sea shadow zone, achieving a range localization error of 0.029 km and a depth error of 0.072 m. The method exhibits strong noise robustness and delivers satisfactory results across diverse deep-sea zones, with optimal performance in shadow zones and secondary effectiveness in the direct arrival zone.

Multi sensor data fusion plays a significant role in addressing the research problems for mobile robot navigation. In this paper, a fusion procedure was developed, based on an Adaptive Neuro Fuzzy Inference Systems (ANFIS), which fuses the data from ultrasonic and LiDAR sensors for a better range estimation as well as environment perception. The root mean square error analysis between the measured and actual distance across different experiment trials indicates that, ultrasonic sensor could provide the data with a root mean square error (RMSE) which varies between 0.3118 cm and 4.9953 cm, whereas the LiDAR sensor could provide the same between 5.1503 cm and 10.4773 cm over 0–4 m range. The RMSE of the proposed fusion algorithm varies between 3.4831 cm and 6.1471 cm. It can be observed that, the fusion process could reduce the mean square error present with the high cost LiDAR sensing system by one-half, while fusing it with the low cost ultrasonic sensing system. The fusion algorithm discussed in this paper provides a guidance to define various operating/safety zones for initiating necessary control action during the navigation of the mobile robots.

This paper presents a technique for performing passive underwater acoustic ranging with data from a single hydrophone and builds upon earlier localization approaches which estimate the sound source position using times of arrival of acoustic energy traveling along direct and/or interface-reflecting paths between source and receiver. In this work, measured time differences between interface-reflecting and direct path arrival times are compared with a set of model-predicted time differences calculated over a set of candidate source ranges in a way that does not require measured arrival paths to be labeled (e.g., direct, surface bounce, bottom bounce, etc.). The modeled set with the best match to the measured data indicates the best estimate of source range. To enable the processing of multi-year data sets, the detection and localization steps are automated and, where possible, multi-threaded to improve computational efficiency on multi-core computer processors. This approach is demonstrated using 20-Hz fin whale (Balaenoptera physalus) calls recorded by the ALOHA Cabled Observatory (ACO), 100 km N of Oahu (Hawaii) in 4782 m of water.This paper presents a technique for performing passive underwater acoustic ranging with data from a single hydrophone and builds upon earlier localization approaches which estimate the sound source position using times of arrival of acoustic energy traveling along direct and/or interface-reflecting paths between source and receiver. In this work, measured time differences between interface-reflecting and direct path arrival times are compared with a set of model-predicted time differences calculated over a set of candidate source ranges in a way that does not require measured arrival paths to be labeled (e.g., direct, surface bounce, bottom bounce, etc.). The modeled set with the best match to the measured data indicates the best estimate of source range. To enable the processing of multi-year data sets, the detection and localization steps are automated and, where possible, multi-threaded to improve computational efficiency on multi-core computer processors. This approach is demonstrated using 20-Hz fin ...

Matched field processing (MFP) is frequently used to localize underwater sources, in range and depth, using vertical arrays recording ultra low frequency waves (/spl les/100Hz). In this paper, we use matched-field techniques, guided waves propagation and signal processing tools to estimate source depth in a shallow water environment with a horizontal array. MFP is performed in the frequency-wavenumber domain: frequency-wavenumber transform (f-k) is the square modulus of the 2D Fourier transform in time and radial distance. This f-k representation provides much information on guided propagation of Ultra Low Frequency waves in shallow water: on this representation, modes can be separate. Besides, modes excitation depends on the source depth. As a result, it is possible to estimate the source depth using MFP based on modes excitation factors.

Source localization with vertical arrays in shallow water has been a topic of intense research in the last 20 years. Although horizontal arrays can (and have) been used, vertical arrays are better suited for extracting signal modal structure and thus provide a source-location estimate in range and depth. It is well known that broadband signals have a localization capability superior to that of narrowband signals. One question that remains largely unresolved is whether frequency extent can compensate for the spatial diversity provided by sensor arrays, i.e. at the limit, can a broadband signal be localized with a single sensor? This communication presents theoretical, simulated, and practical evidence that a multipath-delay maximum-likelihood estimator does provide enough signal to background discrimination for localizing a 500 Hz bandwidth signal at 5 km range in a 130 m depth shallow water channel with a single hydrophone. The real data used in this study were collected during the INTIMATE'96 experiment which was conducted off the west of Portugal in June 1996 during an internal tide tomography experiment. A comparison with localization results provided by a direct match between the received and the model-generated arrival patterns is discussed.