Sonar Signal Research Articles

A Novel Beam-Domain Direction-of-Arrival Tracking Algorithm for an Underwater Target

Underwater direction-of-arrival (DOA) tracking using a hydrophone array is an important research subject in passive sonar signal processing. In this study, a DOA tracking algorithm based on a novel beam-domain signal processing technique is proposed to ensure robust DOA tracking of an interested underwater target under a low signal-to-noise ratio (SNR) environment. Firstly, the beam-based observation is designed and proposed, which innovatively applies beamforming after array-based observation to achieve specific spatial directivity. Next, the proportional–integral–differential (PID)-optimized Olen–Campton beamforming method (PIDBF) is designed and proposed in the beamforming process to achieve faster and more stable sidelobe control performance to enhance the SNR of the target. The adaptive dynamic beam window is designed and proposed to focusing the observation on more likely observation area. Then, by utilizing the extended Kalman filter (EKF) tracking framework, a novel PIDBF-optimized beam-domain DOA tracking algorithm (PIDBF-EKF) is proposed. Finally, simulations with different SNR scenarios and comprehensive analyses are made to verify the superior performance of the proposed DOA tracking approach.

The acoustic complexity index (ACI): theoretical foundations, applied perspectives and semantics

The acoustic complexity index (ACI) is a commonly used metric in ecoacoustics, demonstrating reliability across diverse environments and ecological conditions. However, this index requires specific procedures to be applied correctly. Based on the Canberra metric, the ACI is an unsupervised metric formulated to extract information from fast Fourier transform (FFT) sonic matrices. The ACI measures contiguous differences in acoustic energy of each frequency bin along temporal steps (ACItf) and a temporal interval along the frequency bins (ACIft). Aggregating data after an FFT with a clumping procedure allows for better scaling of sonic signals before computing the ACI. A filter must be applied to reduce the effects of nonenvironmental signals produced by microphone electrical noise . Due to the singularity of the index for values of 0, ACI requires ad hoc procedures to exclude element pairs for which one of the elements is equal to 0 from the comparisons. The spectral and temporal sonic signatures are vectors obtained from the sequence of ACItf and ACIft values, respectively. The comparison between sonic signatures using the chord distance index returns spectral and temporal sonic dissimilarities, allowing the evaluation of sonic patterns at different temporal and spatial resolutions. Sonic variability, sonic evenness, and the effective number of frequency bins are further derivative metrics that help interpret sonic heterogeneity by distinguishing the temporal and spatial heterogeneity of sonoscapes. Moreover, this paper proposes changing the terminology of ‘acoustic complexity index' to ‘sonic heterogeneity index.'

Examination of representative manifolds existence in passive sonar dataset

Conventionally, passive sonar classification has been conducted to predict a ship size for given sample. Recent studies have shown high accuracy, far exceeding the performance of early baselines. However, the assumption that manifolds of passive sonar dataset are formed according to the ship size hasn't been studied deeply. In this paper, we investigate the assumption and attempt to analyze the dataset to observe the manifolds. We consider various attributes of passive sonar signals including, which could form the manifold. We employ unsupervised and supervised learning methods with various extracted features. In our analysis, it is difficult to find an attributes dominantly forming the manifolds because passive signals significantly changed by various facts including ocean environment and moving status. Our study wil provide a deeper understanding of the dataset, leading to a realistic and robust passive sonar classifier.

Reliable underwater multi-target direction of arrival estimation with optimal transport using deep models.

Multi-target direction of arrival (DoA) estimation is an important and challenging task for sonar signal processing. In this study, we propose a method called learning direction of arrival with optimal transport (LOT) to accurately estimate the DoAs of multiple sources with a single deep model. We model the DoA estimation problem as a multi-label classification task and introduce an optimal transport (OT) loss based on the OT theory to capture the intrinsic continuity within the angular categories. We design a cost matrix for the OT loss in LOT approach to characterize the order and periodicity of the angular grid. The LOT approach encourages reliable predictions closer to the ground truth and suppresses spurious targets. We also propose a lightweight channel mask data augmentation module for deep models that use items related to the covariance matrix as input. The proposed methods can be seamlessly integrated with different model architectures and we indicate the portability with experiments on several typical network backbones. Experiments across various scenarios using different measurements show the effectiveness and robustness of our approaches. Results on SwellEx-96 experimental data demonstrate the practicality in real applications.

Bio-inspired target detection method of active sonar based on multi-ping perception

Abstract The performance of the active sonar may degrade severely in real ocean environments, influenced by strong reverberation and multiple clutters. Inspired by the observation that dolphins always utilize a train of clicks when detecting underwater, we proposed an active target detection method by making full use of the multi-ping echo information which encompasses the differences between targets and reverberation/clutters. The detection performance of the click trains with different parameters were analysed first to give an insight on signal choice. Then we achieved apparent reverberation/clutter suppression result by utilizing the correlations and differences between multi-ping echoes. Further, the relative motion status between the target and the sonar platform could be perceived within multiple echoes, which strengthened the detection ability of weak target. We proposed two weighting strategies for multi-ping co-utilization. Either equal weights or unequal weights based on the different envelopes of dolphin echolocation click trains were exerted on multiple echoes, which can achieve different perception results for the relative target motion status. Both simulation and lake trials were conducted to verify the performance of the proposed method. The proposed method effectively suppress reveberaton and noise background by around 2-3 dB and achieve bionic detection of weak targets compared to traditional active sonar signal processing method.

A Web-Based Interactive Application to Simulate and Correct Distortion in Multibeam Sonars

Multibeam sonars are advanced scientific tools for estimating fish school volume and density, using multiple beams to provide comprehensive size measurements of detected targets. However, challenges remain in accurately estimating target dimensions due to beam geometric expansion and overlap, particularly in athwart-beam measurements, which tend to be overestimated with increasing distance from the transducer. We present an interactive web application that simulates distortion caused by beam overlap and expansion in multibeam sonars using simple geometric equations. Users can define sonar characteristics, such as the number of beams, swath opening, or degree of overlap, as well as specify an elliptical target’s dimensions, orientation, and distance from the transducer. The application estimates and visualises the true and distorted shapes of the target, calculating the level of distortion. It can run simulations in both forward and inverse directions, either simulating the distortion of a true school or correcting the shape of a distorted school. This tool aims to enhance the interpretation of multibeam sonar signals and improve the accuracy of target dimension estimates, facilitating more effective use of these sonars in scientific research.

Resistance to age-related hearing loss in the echolocating big brown bat ( Eptesicus fuscus ).

Hearing mediates many behaviors critical for survival in echolocating bats, including foraging and navigation. Most mammals are susceptible to progressive age-related hearing loss; however, the evolution of biosonar, which requires the ability to hear low-intensity echoes from outgoing sonar signals, may have selected against the development of hearing deficits in echolocating bats. Although many echolocating bats exhibit exceptional longevity and rely on acoustic behaviors for survival to old age, relatively little is known about the aging bat auditory system. In this study, we used DNA methylation to estimate the ages of wild-caught big brown bats ( Eptesicus fuscus ) and measured hearing sensitivity in young and aging bats using auditory brainstem responses (ABRs) and distortion product otoacoustic emissions (DPOAEs). We found no evidence for hearing deficits in aging bats, demonstrated by comparable thresholds and similar ABR wave and DPOAE amplitudes across age groups. We additionally found no significant histological evidence for cochlear aging, with similar hair cell counts, afferent, and efferent innervation patterns in young and aging bats. Here we demonstrate that big brown bats show minimal evidence for age-related loss of peripheral hearing sensitivity and therefore represent informative models for investigating mechanisms that may preserve hearing function over a long lifetime.

Sonic Signals and Symbolic “Musical Space”

Sonic Signals and Symbolic “Musical Space”

Fast Sparse Bayesian Learning Based on Beamformer Power Outputs to Solve Wideband DOA Estimation in Underwater Strong Interference Environment

Wideband direction-of-arrival (DOA) estimation is an important task for passive sonar signal processing. Nowadays, sparse Bayesian learning (SBL) attracts much attention due to its good performance. However, performance degrades in the existence of strong interference. This problem can be solved by combining the beamformer and the SBL. The beamformer is a useful tool to suppress interference. Then, the SBL can easily estimate the DOA of the targets from the beamformer power outputs (BPO). Unfortunately, the latter step needs to compute the matrix inversion frequently, which brings some computational burden to the sonar system. In this paper, the BPO-based SBL is modified. A sequential solution is provided for the parameters in the BPO probabilistic model. In this manner, only one signal precision parameter involved in the probabilistic model is updated in each iteration and the matrix inversion is avoided during the iteration, thus reducing the computational burden. Simulation and experimental results show that the proposed method maintains high estimation precision in the interference environment. At the same time, its computational efficiency is almost three times higher in comparison with state-of-the-art methods.

Vocalization Pattern and Echolocation Signal Characteristics of Yangtze Finless Porpoise (Neophocaena asiaeorientalis asiaeorientalis) in Captivity

The Yangtze finless porpoise (Neophocaena asiaeorientalis asiaeorientalis, YFP) possesses the ability to detect distance through echolocation signals, and its sonar signal signature is adjusted to detect different targets. In order to understand the vocal characteristics of YFPs in different behavioral states and their differential performance, we recorded the vocal activities of YFPs in captivity during free-swimming, feeding, and nighttime resting and quantified their signal characteristic parameters for statistical analysis and comparison. The results showed that the number of vocalizations of the YFPs in the daytime free-swimming state was lower than that in the feeding and nighttime resting states, and the echolocation signals emitted in these three states showed significant differences in the −10 dB duration, −3 dB bandwidth, −10 dB bandwidth, and root-mean-square (RMS) bandwidth. Analysis of the resolution of the echolocation signals of the YFPs using the ambiguity function indicated that their distance resolution could reach the millimeter level. These results indicate that the echolocation signal characteristics of YFPs present diurnal differences and that they can be adjusted with changes in their detection targets. The results of this study can provide certain scientific references and foundations for the studies of tooth whale behavioral acoustics, and provide relevant scientific guidance for the conservation and management of YFPs.

Data Compression and Damage Evaluation of Underground Pipeline With Musicalized Sonar GMM

The underground cement pipeline network is critical in urban infrastructure, such as sewage drainage, fluid transportation, etc. Pipelines are deeply buried, so the damage is difficult to be detected. It is of great significant to monitor the crack of pipeline in real time. In this research, a data compression and damage evaluation method of underground pipeline using musicalized sonar GMM (Gaussian Mixed Model) was proposed. First, the sonar echo signal was sampled and discretized into four musicalized indicators by musicalized model. Then, the distribution of multi parameters were analyzed by GMM. The Frechet similarity was constructed to compare the difference between each working condition and baseline condition, so that the damage state could be identified. The engineering experiment shown that this method could greatly compress the monitoring data and evaluate four crack levels.

Understanding model fidelity for training synthetic aperture sonar image classifiers

Convolutional neural networks (CNNs) are increasingly employed for classification tasks in automated target recognition (ATR) algorithms for synthetic aperture sonar (SAS) images. Training ATR algorithms, however, requires many unique observations of the targets of interest. Data available for training is often limited, and acquiring additional training data through experimentation is prohibitively expensive. Increasingly, training with simulated data is being considered as an alternative, but the fidelity required of the models that generate this data is not yet known. SAS imagery typically contains significant complexity from countless physical mechanisms. Simulating complex scenes requires multiple models to account for these different effects, but some effects may not require accurate modeling for the purposes of training an ATR algorithm. This presentation will describe a study to investigate the fidelity of models required so that simulated data may be used interchangeably with experimental data for training CNNs. Using in-air experimentation, a high-fidelity data set was developed with multiple degrees of complexity. A high-frequency sonar signal model was then used to generate complementary simulated data. This approach allows for specific physical features in the data to be individually isolated, enabling detailed exploration of the relationships between model fidelity and CNN architecture.

Physics-anchored masked autoencoder for efficient sonogram simulation

We present a technique for efficiently creating sonogram such as LOFAR and DEMON, by combining physical acoustic modeling and data-driven method of masked autoencoder. This technique involves two stages. First, in the given the ocean environment, the physical model accurately calculates a restricted portion of entire sonogram image. Next, the data-driven model based on masked autoencoder creates an image of remaining region. By employing an iterative decoding structure and controlling the weight of the physical loss term, the reconstruction accuracy of masked autoencoder is improved. The results are compared with those of a pure physical model, a hybrid model with original masked autoencoder, and a hybrid model with traditional PCA technique. We will discuss the practicality of the proposed technique in terms of accuracy and calculation performance, as well as the applicablity of real data using Deepship dataset. [Work supported by the Korea Research Institute for Defense Technology Planning and Advancement (KRIT) grant funded by the Korea government (Defense Acquisition Program Administration (DAPA)) (No. KRIT-CT-22-052, Physics-guided Intelligent Sonar Signal Detection Research Laboratory).]

Coupled models for designing and predicting performance of active sonar systems

Active sonar systems used to survey and image the seafloor typically transmit a pulse on a projector, receive echoes on an array of hydrophones, and condition and digitize the signals using on-board electronics. The quality of the data produced by these sensors is a function of the transducers, the electronics, and the environment. Simultaneous modeling of all these factors is necessary for accurate prediction of the system performance. Data products from these systems are often uncalibrated because absolute calibration is not required for their applications. However, modeling active sonar systems in their natural units within each domain provides clear benefits such as allowing system limitations to be accurately identified early during the design phase and easing validation of system performance in comparison with experimental data. This presentation will discuss the use of coupled multi-domain models in the design of two active sonar systems: the Sediment Volume Search Sonar and a multibeam echosounder architecture based on sigma-delta conversion. By coupling electro-mechanical transducer models and electrical hardware models with the Point-based Sonar Signal Model (PoSSM), it was possible to identify critical design parameters and sensitivities in the design of these sensors.

Acoustics at UMass Dartmouth

The University of Massachusetts Dartmouth has a long history spanning 5 decades of graduate courses offerings and research in underwater acoustics, transduction and signal processing leading to M.S. and Ph.D. degrees in Electrical Engineering. Collaborations between Marine Science, Physics, Mechanical, Bioengineering, and Electrical Engineering departments offer many interdisciplinary opportunities in Acoustics. Courses include Fundamentals of Acoustics, Vibrations, Underwater Acoustics, Electroacoustic Transducers, Medical Ultrasonics, Sonar, Digital Signal Processing, Array Processing, Random Signals, Information Theory, Communications, Detection and Estimation. Many unique facilities including an Underwater Acoustic Test Tank, Open-Ocean water access, and Unmanned Underwater Vehicles support both undergraduate and graduate projects. Research focuses include transducers and transduction science, materials characterization, calibration, array and sonar signal processing, animal bioacoustics, communications, detection and estimation, active and passive sonar funded by Office of Naval Research and Industry.

Underwater Object Prediction Using Sonar Waves

The detection and differentiation of potentially hazardous objects, such as mines and rocks, in underwater environments are crucial for the safe navigation of submarines during warfare scenarios. This study compares and contrasts different machine learning algorithms to determine whether an object detected by a submarine's sonar system is a mine or a rock. From the comparative analysis, we have proposed the Logistic Regression Model (LRM), which is used to train and test the models. To extract pertinent features like signal intensity, frequency content, and time-domain characteristics, sonar signals are essential to the labeled dataset. This dataset is made up of sonar signals and corresponding object classifications (mine or rock) based on the frequency range. These features served as inputs to the machine learning algorithms and enabled the features to learn the underlying patterns and make accurate predictions. In this paper, we have contributed a novel approach to underwater object detection using LRM, which is applied to sonar data. In summary, this research presents an innovative solution to the challenge of underwater object detection using sonar signals and machine learning algorithms. The developed model exhibits promising results, opening up new possibilities for advancing our understanding of underwater environments and enhancing underwater exploration and security capabilities.

A Bayesian generating function approach to adverse drug reaction screening.

Determining causality of an adverse drug reaction (ADR) requires a multifactor assessment. The classic Naranjo algorithm is still the dominant assessment tool used to determine causality. But, in spite of its effectiveness, the Naranjo algorithm is manually intensive and impractical for assessing very many ADRs and drug combinations. Thus, over the years, many "automated" algorithms have been developed in an attempt to determine causality. By-and-large, these algorithms are either regression-based or Bayesian. In general, the automatic algorithms have several major drawbacks that preclude fully automated causality assessment. Therefore, signal detection (or causality screening) plays a role in a "first pass" of large ADR databases to limit the number of ADR/drug combinations a skilled human further assesses. In this work a Bayesian signal detector based on analytic combinatorics is developed from a point of view commonly adopted by engineers in the field of radar and sonar signal processing. The algorithm developed herein addresses the commonly encountered issues of misreported data and unreported data. In the framework of signal processing, misreported ADRs are identified as "clutter" (unwanted data) and unreported ADRs are identified as "missed detections". Including the aforementioned parameters provides a more complete probabilistic description of ADR data.

CMAF: Cross-Modal Augmentation via Fusion for Underwater Acoustic Image Recognition

Underwater image recognition is crucial for underwater detection applications. Fish classification has been one of the emerging research areas in recent years. Existing image classification models usually classify data collected from terrestrial environments. However, existing image classification models trained with terrestrial data are unsuitable for underwater images, as identifying underwater data is challenging due to their incomplete and noisy features. To address this, we propose a cross-modal augmentation via fusion ( CMAF ) framework for acoustic-based fish image classification. Our approach involves separating the process into two branches: visual modality and sonar signal modality, where the latter provides a complementary character feature. We augment the visual modality, design an attention-based fusion module, and adopt a masking-based training strategy with a mask-based focal loss to improve the learning of local features and address the class imbalance problem. Our proposed method outperforms the state-of-the-art methods. Our source code is available at https://github.com/WilkinsYang/CMAF .

Multi-target direction-of-arrival estimation of deep models with frame-level permutation invariant training in marine acoustic environment.

Direction-of-arrival (DoA) estimation is an important part in sonar signal processing, providing a reliable foundation for tasks, such as underwater object detection and tracking. Although the deep learning model has powerful data fitting capabilities, accurately estimating the orientation of multiple targets with a single model remains a challenging task. To address this challenge, we enhance the permutation invariant training (PIT) technique and propose two different types of methods: multi-group classification with PIT (MC-PIT) and multi-group regression with PIT (MR-PIT). These two frame-level PIT schemes utilize a single model for both training and testing in multi-target scenarios. Furthermore, we evaluate the performance of MR-PIT and MC-PIT with different network backbones and demonstrate that the frame-level PIT has excellent portability. Compared with the model trained with the general multi-label strategy, simulation experiments show that our proposed methods have better multi-target DoA estimation performance. Finally, when the array configuration of simulated and recorded data are consistent, the model with frame-level PIT can achieve good performance on recorded data even only trained on simulation data.

Comparative Analysis of SVM and CNN for Sonar Signal Classification Using Sparse Arrays

Comparative Analysis of SVM and CNN for Sonar Signal Classification Using Sparse Arrays