Underwater Acoustic Localization Research Articles

Source localization in deep ocean based on complex convolutional neural network

To solve the problem that phase information cannot be used effectively in underwater acoustic localization, a deep learning network based on complex convolutional neural network is proposed in this paper. The complex convolution layer is used to effectively utilize the phase features favorable to the source localization problem, and the phase information is used to improve the feature extraction performance. Through simulation, the performance of the network for source localization in deep-sea direct arrival region under different SNR conditions is analyzed. The results show that the complex convolutional network proposed in this paper can locate the sound source with less computation and has better performance under the condition of low SNR.

Reflection-enabled HDOP-weighted underwater acoustic passive localization with a small aperture array

This paper addresses the localization problem of a noncooperative source in the multipath-rich environment. Unlike conventional schemes that only use the line-of-sight (LOS) and the once-surface-reflected (S1B0) path, the proposed scheme utilizes the complete spatial diversity of all multipaths by quantifying their positioning confidence. In the proposed scheme, all the measurement metrics (angle, relative delay and amplitude) extracted from multipath arrival structure, are combined to achieve localization with a small aperture array.Specifically, we first propose an algorithm for extracting multipath arrival structure based on spatial time-frequency analysis of underwater multipath signals. In this algorithm, each path is iteratively extracted, and the extraction error is close to the Cramér–Rao lower bound (CRLB). Then, the obtained CRLB is used to derive horizontal dilution of precision (HDOP) to quantify the positioning confidence of each path. Finally, the source location is estimated from an optimal HDOP-weighted multipath localization algorithm that assigns different weights to the LOS path and non-line-of-sight (NLOS) paths.Statistical-channel-model-based simulations and lake trial corroborate that the proposed scheme reduces the localization error by at least 128% compared with angle-only, relative-delay-only, and conventional hybrid metrics localization schemes.Index Terms--Underwater acoustic passive localization; Multipath arrival structure; Horizontal dilution of precision; Hybrid metrics.

High-performance towing cable hydrophone array with an improved ultra-sensitive fiber-optic distributed acoustic sensing system.

A high-performance towing cable hydrophone array based on an improved ultra-sensitive fiber-optic distributed acoustic sensing system (uDAS) with picostrain sensitivity is demonstrated and tested in sea trial, for the first time. A new composite transducer is designed and optimized to enhance the acoustic pressure sensitivity significantly. A sea trial is carried out to test the performances of such a hydrophone array, including flow noise, underwater acoustic signal capture capacity, beamforming processing and localization of artificial source targets. The array exhibits high sensitivity and low noise floor. An average sensitivity of -129.23 dB re rad/µPa at frequencies from 10 Hz to 1500 Hz has been achieved. The localization at distances of 5 km and 10 km is realized, respectively, validating the excellent remote detection and positioning capability of the hydrophone system. The proposed towing cable system, with high sensitivity, simple structure and remote target localization ability, may pave a way for development of the next generation of high-performance light-weighting hydrophone arrays for towing applications.

Efficient Underwater Acoustical Localization Method Based on TDOA with Sensor Position Errors

Underwater acoustic localization (UWAL) is extremely challenging due to the multipath nature of extreme underwater environments, the sensor position uncertainty caused by unpredictable ocean currents, and the lack of underwater observation data due to sparse array, which all affect localization performance. Addressing these issues, this paper proposes a simple and effective underwater acoustic localization method using the time difference of arrival (TDOA) measurements based on the multipath channel effect of the underwater environment. By introducing the calibration source, localization performance was improved, and the sensor position error was corrected. The Cramér–Rao lower bound (CRLB) was derived, and the proposed method was able to achieve the CRLB with small deviation. Numerical simulations confirm the improved performance of the proposed method, including (1) a 20 dB and 30 dB reduction in the CRLB for far and near source scenarios, respectively, indicating improved accuracy and reliability when estimating unknown sources; (2) better Mean Squared Error (MSE) performance compared to existing methods and an efficiency of over 90% in low noise and above 80% in moderate noise in several scenarios, with a delayed threshold effect; and (3) achieving CRLB performance with only three sensors in a 3D space, even under moderate noise, while existing methods require at least five sensors for comparable performance. Our results demonstrate the efficacy of the proposed method in enhancing the accuracy and efficiency of source localization.

On data-driven underwater acoustic direct localization: Design considerations of a deep neural network-based solution

High reverberation, depth-dependent speed of sound and intricate environmental structures are a few of the challenges that make the development of reliable underwater acoustic localization (UAL) methods hard. Since the nontrivial physics of underwater acoustics and the inherent random effects amount to a highly complex statistical model of the measured data, it is natural to consider deep neural networks (NNs) as a central tool in the development of UAL methods. Indeed, recent studies have shown that some properly designed NNs can lead to unprecedented localization capabilities and enhanced accuracy. However, out of the already myriad available possibilities, it is not immediately clear how to choose a proper NN architecture—input structure, layer types, training procedure, loss function(s), etc.—that will lead to successful operation. In this talk, we will present an analytically informed architecture that can learn (for now, in simulations) to localize in complex underwater environments. We will also provide analytical arguments that can explain the robustness of such architectures to certain types of propagation/environment model mismatches. The operation of the discussed architecture will be demonstrated via simulation results in a rich, reverberant channel. The results suggest that deep NNs could become a viable solution to the UAL problem.

Underwater acoustic localization via gradient-based optimization

In underwater acoustic localization via matched-field-processing, given a propagation model and a suitable environmental parameterization, one searches for the location (of the transmitter or receiver) whose replica field is closest to the observed one. The high computational complexity of such non-gradient-based optimization methods renders them infeasible for many real-time scenarios, especially when an accurate solution is desired due to resolution of the search grid required, or as the search dimensionality increases (e.g., when it is necessary to optimize over uncertain environmental parameters such as sound speed or bathymetry). In this talk, we propose a ray-based, differentiable model for acoustic propagation that can be exploited in a gradient-based optimization for localization. For localization applications in which accurate times of arrival might not be available (e.g., due to the signal's relatively small bandwidth), the proposed method does not directly rely upon times of arrivals. Rather, it seeks the location (and possibly environmental parameters) that minimize the squared-error between the observed signal and its estimation via the differentiable model. We leverage the PyTorch optimization and auto-differentiation tools for the implementation and demonstrate successful localization on synthetic data in a dense multipath environment.

Array invariant-based range-only target motion analysis for ship noise measured by two vertical line arrays in shallow water

A novel approach for underwater acoustic localization by using passive line array sonar will be proposed in this talk. In the case of the received signal having a finite pulse duration, the target location can be easily estimated using the time-difference-of-arrival (TDOA) method between sensors. However, it is difficult to use the TDOA method for signals that exist for a relatively long time, such as ship noise. An array invariant method has been proposed to measure the source-receiver distance, but it does not accurately estimate the target location. In this talk, target localization will be performed with the range-only target motion analysis method using non-linear target tracking after estimating the target range by the array invariant method. The proposed method is tested using the ship noise of the R/V Onnuri measured by two vertical line arrays during the Shallow-water Acoustic Variability Experiment (SAVEX-15), and the result will be discussed in this talk. [This research was supported by the Ministry of Oceans and Fisheries of Korea (PM63010, 2022).]

A Multiple Target Data Association Method for TDOA Passive Localization

In order to solve the multiple target data association problem in underwater acoustic passive localization system, a new algorithm focus on data association in Time Difference of Arriva l(TDOA) is studied. Firstly, the zero-sum condition is used to select TDOA measurements according to the continuous measurement values within a certain period of time. Then, the selected TDOA measurements are associated with the multiple targets and the correct correlation probability is measured by the correlation degree. The algorithm is used in sea trial. Processing results show that the method proposed in this paper can effectively solve the TDOA measurement association problem, and can locate multiple targets at the same time.

High-Precision Underwater Acoustic Localization of the Black Box Utilizing an Autonomous Underwater Vehicle Based on the Improved Artificial Potential Field

Underwater acoustic localization (UWAL) of the black box for a sunken airplane utilizing an autonomous underwater vehicle (AUV) is a useful technique in ensuring the traffic safety. Aiming at improving the localization precision, this paper proposes a new path planning algorithm based on the improved artificial potential field (APF). Compared with the conventional APF, we modify the conventional gravitation force and introduce a new localization precision force. Therefore, a balance between the localization precision and the obstacle avoidance is achieved, and the localization precision is significantly improved. Lake trail result validates the effectiveness of the proposed method.

Underwater Acoustic Localization of the Long Baseline Based on Track-Before-Detect

This work considers the long baseline (LBL) acoustic localization problem. In the classical LBL localization approach, the time of arrival (TOA) of direct sound from each buoy is estimated independently, and the estimated TOAs are then utilized to solve for the target’s position. This approach ignores the fact that all of the estimated TOAs from different buoys correspond to the same target and thus are intrinsically correlated. To address this shortcoming, a novel localization framework based on track-before-detect (TBD) theory is established in this work, which allows the target location to be directly determined. Since it is completely freed from the traditional localization paradigm, some of the persistent and challenging problems faced by classical localization methods, such as direct sound selection, and unacceptable false alarm/detection trade-offs caused by threshold setting, can be circumvented. The effectiveness of the TBD approach is validated on both simulated and real datasets. The experimental results indicate that the proposed approach can track targets continuously and accurately and can be effectively implemented even under challenging conditions.

Underwater target passive acoustic localization method based on Hanbury Brown–Twiss interference

PurposeAcoustic signals of the underwater targets are susceptible to noise, reverberation, submarine topography and biology, therefore it is difficult to precisely locate underwater targets. This paper proposes a new underwater Hanbury Brown-Twiss (HBT) interference passive localization method. This study aims to achieve precise location of the underwater acoustic targets.Design/methodology/approachThe principle of HBT interference with ultrasensitive detection characteristics in optical measurements was introduced in the field of hydroacoustics. The coherence of the underwater target signal was analyzed using the HBT interference measurement principle, and the corresponding relationship between the signal coherence and target position was obtained. Consequently, an HBT interference localization model was established, and its validity was verified through simulations and experiments.FindingsThe effects of different array structures on the localization performance were obtained by simulation analysis, and the simulations confirmed that the HBT method exhibited a higher positioning accuracy than conventional beamforming. In addition, the experimental analysis demonstrated the excellent positioning performance of the HBT method, which verified the feasibility of the proposed method.Originality/valueThis study provides a new method for the passive localization of underwater targets, which may be widely used in the field of oceanic explorations.

Design of a complete simulator for underwater acoustic localization systems based on spread-spectrum signals

Deploying prototype positioning systems in underwater environments is expensive and an especially challenging task, so a first common approach is to carry out simulated studies to evaluate the requirements and restrictions imposed by the environment. In this regard, it is helpful to have simulation models that allow the generation of a wide range of tests as a previous step to any experimental prototype implementation. For that purpose, this work focuses on the design of a simulation tool that facilitates research on underwater positioning systems by considering several parameters and features, such as the design of the signals emitted by the acoustic transducers (encoding techniques, modulation schemes, etc.), the frequency response and location of emitters and hydrophones, the bathymetry of the seabed, and the channel effects on the ultrasonic signal propagation, implemented in a model based on ray tracing for the propagation of acoustic signals. The simulation tool has been validated through a complete set of tests for different configurations and situations, analyzing the signals involved at different processing stages: baseband, modulated signals, received signals, and final estimated positions. This simulation tool is a valuable asset to research different positioning system configurations or to illustrate several concepts in a pedagogical context.

An evolved algorithm for underwater acoustic sensor node localization enhancement using reference node

Accurate localization of underwater acoustic sensor networks is one of the essential operations. The range-free DV-Hop technique is more acceptable due to its low hardware cost and simplicity than the range-based approach. But the main drawback of this basic DV-Hop method is low accuracy and link failure due to maintenance or short battery life during the continuous localization calculation in the underwater scenario. Most algorithms in this field do not pay enough attention to the mobility of the nodes. By examining the movement patterns of water, this manuscript uses a technique for UWSN localization based on a mobility prediction algorithm. This manuscript proposed a reference node-based multi-hop improved DV-Hop​ technique using School Topper Optimization (STO) and mobility prediction algorithms. This proposed scheme’s highlighted advantage is to reduce the localization error to identify the unknown target nodes using a reference node-based localization scheme using a mobility prediction algorithm without increasing the beacon nodes, and the multi-hop improved DV-Hop​ algorithm using STO significantly improved the link failure problem by iteratively selecting the different sets of the beacon nodes in the same scenario The simulation report has proved that our framework takes better computational time, has a better convergence rate by introducing the STO technique, provides 23.76% and 37.18% better avg. localization error and localization error variance, and 16.12% better coverage from PSODV-Hop and GSODV-Hop techniques.

Frequency-Diversity-Based Underwater Acoustic Passive Localization

This article considers underwater acoustic passive localization of a noncooperative broadband source in the presence of multipath propagation. A differential transmission-loss model-based passive localization method is proposed, in which 3-D localization can be implemented with only two hydrophones. Instead of spatial diversity, frequency diversity is exploited to achieve passive localization with a limited number of hydrophones and address the challenge of multipath propagation. Specifically, the received signals of both hydrophones are decomposed in the frequency domain, respectively, at first. Then, a set of differential transmission-loss model-based equations between the two hydrophones are established on multiple frequencies. Based on these equations, the passive localization is modeled as a multivariate optimization problem. Meanwhile, the nonline of sight (NLOS)-related parameters are also involved in the optimization problem. Therefore, accurate localization and NLOS mitigation can be accomplished simultaneously by solving the optimization problem. To simplify the multivariate optimization problem for a reliable solution, a cepstrum-autocorrelation-based multipath estimation algorithm is proposed, by which all the NLOS paths can be represented by an equivalent NLOS path. Consequently, the simplified optimization problem concerns only the desired coordinate of the source and a single NLOS-related parameter. Finally, a differential evolution algorithm is employed to solve the simplified optimization problem. Both simulation and lake trial results corroborate the effectiveness and the robustness of the proposed method.

Through-Ice Acoustic Source Tracking Using Vision Transformers with Ordinal Classification.

Ice environments pose challenges for conventional underwater acoustic localization techniques due to their multipath and non-linear nature. In this paper, we compare different deep learning networks, such as Transformers, Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, and Vision Transformers (ViTs), for passive localization and tracking of single moving, on-ice acoustic sources using two underwater acoustic vector sensors. We incorporate ordinal classification as a localization approach and compare the results with other standard methods. We conduct experiments passively recording the acoustic signature of an anthropogenic source on the ice and analyze these data. The results demonstrate that Vision Transformers are a strong contender for tracking moving acoustic sources on ice. Additionally, we show that classification as a localization technique can outperform regression for networks more suited for classification, such as the CNN and ViTs.

Research on passive localization method of shallow water acoustic source with single hydrophone based on hierarchical grid histogram filtering

An acoustic field model in the form of a probability density function is established to solve the problem of environmental instability in shallow water. A Bayesian underwater acoustic localization model with the state vector of the acoustic source as a posteriori probability is generated for the purpose of trading time for space in an iterative form and realizing localization of a moving source with a single hydrophone. A hierarchical grid histogram filtering algorithm is proposed, which solves the integral problem in the process of Bayesian filtering and improves the efficiency of the histogram filtering algorithm greatly. The SWellEx-96 experimental results show that in the range of 200 m * 10 km, the relative error of depth positioning is controlled at 12%, the relative error of distance positioning is controlled at 7%, and the efficiency of the algorithm is improved by N1/2 times where N is the number of grids of the histogram filter. The method proposed in this paper is suitable for concealed and low energy consumption platforms in shallow water.

Underwater acoustic localization using a modular differentiable model for acoustic wave propagation

Model-based algorithms for underwater acoustic localization are highly dependent upon environmental knowledge that is rarely available. Moreover, model-based localization usually involves optimization in high-dimensional spaces to find the best candidate for source localization. On the other hand, data-driven methods deliver poor generalization performance in data starved scenarios. Therefore, in order to improve generalization, we propose a modular, deep learning-based architecture that inherently learns the multipath structure, while learning the environmental properties from the training data. Furthermore, since ReLU-based fully connected networks cannot capture the high-frequency contents of the signals of interest properly, we use sinusoidal activations. Although the model needs an estimate of the carrier frequency as a hyper-parameter, it is observed that it can tolerate deviations about the true frequency. This model maps the source and receiver locations to received signals and is then used in a gradient descent manner exploiting the automatic differentiation toolbox in PyTorch, to find the source location. To evaluate the performance of the algorithm, we use Bellhop to generate data for a given set of environmental parameters. We show that our method outperforms matched-field processing and a deep fully connected network without a modular structure.

Underwater Localization Using Differential Doppler Scale and TDOA Measurements with Clock Imperfection

Underwater acoustic localization is an important, yet challenging problem: (1) node mobility issue, (2) Doppler effect, and (3) clock imperfection. To be specific, underwater nodes are not stationary in real-life due to unpredictable currents. Relative motion between a transmitter and a receiver causes the time scaling problem on the received signals, where the time scaling factor is termed as Doppler scale. Then, due to the slow acoustic signal propagation speed, the underwater Doppler scale becomes more severe compared with the one in terrestrial environments. Thus, the differential Doppler scale (DDS) measurements should also be collected, other than the time measurements like time-difference-of-arrival (TDOA), for enhancing the underwater localization. Since DDS/TDOA measurements and clock skew are tightly coupled, clock synchronization is essential for accurate localization. However, due to the stringent cost and power constrains of underwater nodes, low-cost clocks with relative low precision are normally employed, which makes it even more difficult to guarantee a perfect clock synchronization between transmitter/receiver pairs. In order to cope with those issues, we propose an algebraic underwater localization method using the hybrid DDS/TDOA measurements, which is particularly robust against the node clock imperfection. A new DDS/TDOA measurement model with clock imperfection is first presented by analyzing the received signals over underwater acoustic channels. Then, we devise a two-step weighted least square-based estimator, and the analytical study shows that our estimator can achieve the Cramer-Rao lower bound (CRLB) accuracy under small noise. Simulations corroborate the theoretical results and the good performance of the proposed method.

A Semi-Blind Method for Localization of Underwater Acoustic Sources

Underwater acoustic localization has traditionally been challenging due to the presence of unknown environmental structure and dynamic conditions. The problem is richer still when such structure includes occlusion, which causes the loss of line-of-sight (LOS) between the acoustic source and the receivers, on which many of the existing localization algorithms rely. We develop a semi-blind passive localization method capable of accurately estimating the source's position even in the possible absence of LOS between the source and all receivers. Based on typically-available prior knowledge of the water surface and bottom, we derive a closed-form expression for the optimal estimator under a multi-ray propagation model, which is suitable for shallow-water environments and high-frequency signals. By exploiting a computationally efficient form of this estimator, our methodology makes comparatively high-resolution localization feasible. We also derive the Cram\'er-Rao bound for this model, which can be used to guide the placement of collections of receivers so as to optimize localization accuracy. The method improves a balance of accuracy and robustness to environmental model mismatch, relative to existing localization methods that are useful in similar settings. The method is validated with simulations and water tank experiments.

High-Rate Underwater Acoustic Localization Based on the Decision Tree

Underwater acoustic localization (UWAL) is widely applied in ocean exploration, and a key issue for the high-rate time-of-arrival (TOA)-based UWAL method is to select the direct signal in the multipath acoustic channel. However, the existing direct signal selection methods suffer from a deterioration of the selection accuracy since the received signals from adjacent periods are mixed with each other. Addressing the problem, we propose a high-rate direct signal selection algorithm in this article. First, we choose the amplitude, TOA, bandwidth, and Doppler frequency of the detected pulses as the input features. Second, a direct signal classifier is established utilizing the decision tree. Finally, the localization model is built based on the TDOA information of the direct signal, and the localization precision analysis evaluates the performance of the UWAL method. Sufficient experiments, including simulation, tank experiment, and sea trial, are implemented to verify the effectiveness of the proposed method.