Time-Frequency Fused Underwater Acoustic Source Localization Based on Contrastive Predictive Coding

Zhu, Xiaoyu; Dong, Hefeng; Rossi, Pierluigi Salvo; Landrø, Martin

doi:10.1109/jsen.2022.3179405

Cited by 7 publications

(4 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In (30), w w w 1,k , w w w 2,k , w w w 3,k are process disturbances that account for uncertainty on pressure, pressure derivative, and source position, respectively. Such process disturbances can be used to account for different sources of uncertainty including parameter uncertainties, time and space discretization errors, etc..…”

Section: A Multiple Model Approachmentioning

confidence: 99%

“…Additionally, this paper builds on large-scale field estimation of discretized PDE systems [23], [24], and previous work on source identifiability and estimation in such systems [25], [26]. Further related work focused on USL for shallow-water environments and highfrequency signals using a multi-ray propagation model [27], decentralized detection in underwater sensor networks [28], decentralized USL via generalized likelihood ratio test [29], a self-supervised learning architecture that exploits joint timefrequency processing for USL [30], and acoustic source localization and tracking using a cluster of mobile agents [31].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Underwater Source Localization via Spectral Element Acoustic Field Estimation

Manduzio,

Forti,

Sabatini

et al. 2023

IEEE Sensors J.

View full text Add to dashboard Cite

Underwater source localization from a passive array of acoustic sensors is a challenging problem, especially in complex environments characterized by multipath and reverberation effects, irregular seabed geometry, and low signal-to-noise ratio. This paper proposes a recursive Bayesian approach that propagates a spectral-element approximation of the wave equation to model the discretized space-time dynamics of the acoustic field conditioned on the position of the source, and sequentially estimates the field and the position of the radiating source from the acoustic measurements. We pursue a multiple-model approach where each model assumes either the source absence or its presence within a specific spectral element. To handle the high dimension of the large-scale field estimation problem and reduce the computational complexity, the multiple-model filter is implemented by using ensemble Kalman filters. Finally, the effectiveness of the proposed multiple-model spectral-element ensemble Kalman filter is demonstrated through simulation experiments in underwater acoustic environments with regular and irregular seabed geometry and via comparison with the standard matched-field processing method.

show abstract

Section: A Multiple Model Approachmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Underwater Source Localization via Spectral Element Acoustic Field Estimation

Manduzio,

Forti,

Sabatini

et al. 2023

IEEE Sensors J.

View full text Add to dashboard Cite

show abstract

“…Signal physics-based methods rely on basic characteristics, temporal features, and non-Gaussian characteristics of underwater acoustic signals (Yao X. et al, 2023). This includes time-domain features like zero-crossing distribution, frequency-domain features like cepstral analysis (Zhu et al, 2022), and joint time-frequency domain features such as wavelet transforms (Han et al, 2022;Tian et al, 2023). Brain-like computing features for underwater acoustic signals include Mel-frequency cepstral coefficients (MFCC) simulating nonlinear processing of the human ear (Di et al, 2023) and Gammatone filtering simulating peripheral auditory processing (Zhou et al, 2022).…”

Section: Introductionmentioning

confidence: 99%

MGFGNet: an automatic underwater acoustic target recognition method based on the multi-gradient flow global feature enhancement network

Chen,

Tang,

Qiu

et al. 2023

Front. Mar. Sci.

View full text Add to dashboard Cite

The recognition of underwater acoustic targets plays a crucial role in marine vessel monitoring. However, traditional underwater target recognition models suffer from limitations, including low recognition accuracy and slow prediction speed. To address these challenges, this article introduces a novel approach called the Multi-Gradient Flow Global Feature Enhancement Network (MGFGNet) for automatic recognition of underwater acoustic targets. Firstly, a new spectrogram feature fusion scheme is presented, effectively capturing both the physical and brain-inspired features of the acoustic signal. This fusion technique enhances the representation of underwater acoustic data, resulting in more accurate recognition results. Moreover, MGFGNet utilizes the multi-gradient flow network and incorporates a multi-dimensional feature enhancement technique to achieve fast and precise end-to-end recognition. Finally, a loss function is introduced to mitigate the influence of unbalanced data sets on model recognition performance using Taylor series. This further enhances model recognition performance. Experimental evaluations were conducted on the DeepShip dataset to assess the performance of our proposed method. The results demonstrate the superiority of MGFGNet, achieving a recognition rate of 99.1%, which significantly surpasses conventional methods. Furthermore, MGFGNet exhibits improved efficiency compared to the widely used ResNet18 model, reducing the parameter count by 51.28% and enhancing prediction speed by 33.9%. Additionally, we evaluated the generalization capability of our model using the ShipsEar dataset, where MGFGNet achieves a recognition rate of 99.5%, indicating its superior performance when applied to unbalanced data. The promising results obtained in this study highlight the potential of MGFGNet in practical applications.

show abstract

“…Given the environment in which the experimental data of interest were collected, simulated data in similar environments produced using these software packages can be run for specific ranges and used as the basis for training a deep learning model. Other ranging methods have also been used such as classifying received signals into discrete range bins [103] or using recurrent architectures or more advanced representation learning methods such as contrastive learning [104].…”

Section: Range Estimationmentioning

confidence: 99%

Automatic baleen whale detection and 2D localization using a network of unsynchonized passive acoustic sensors

Goldwater

View full text Add to dashboard Cite

Underwater acoustics is a powerful tool for learning about the ocean’s soniferous marine life. However, most modern acoustic sensing systems consist of expensive arrays of timesynchronized recorders which require a crewed research vessel and significant expertise to deploy, operate, and recover. Recently, there has been a growing corpus of research related to algorithms for low-cost and accessible acoustic hardware. Deep learning methods have shown great promise when applied to underwater acoustics inverse problems. While many signal processing or physics-based algorithms exhibit long run times and require manual labor to extract signals of interest, tune parameters, as well as visually verify the results, an appropriately trained neural network can quickly process data with no human supervision. Both low-cost passive acoustic monitoring (PAM) sensing platforms and algorithms that can analyze massive amounts of raw data are critical to accessible and scalable approaches in ocean acoustic monitoring. This thesis presents a method for detection and 2D (latitude-longitude) localization of underwater acoustic sources without requiring synchronized sensors. The signals of interest here are the dispersive low-frequency impulsive gunshot vocalizations of North Pacific and North Atlantic right whales (NPRWs, NARWs). In shallow-water channels, the timefrequency representation of the received signal is strongly dependent on source-receiver range, making these impulses ideal candidates for range-based localization. The first step in the localization pipeline uses a temporal convolutional network (TCN) to simultaneously detect gunshot vocalizations and predict their ranges. Trained on spectrograms of synthetic data simulated in a variety of environments, the TCN is applied to PAM data from moorings in the Bering Sea. Gunshots are detected with high precision, and the range estimates are comparable to those estimated using traditional physics-based processing. Both methods use a minimal set of a priori environmental information including water column depth, sound speed, and density. Depending on the sensor layout, the TCN may need to scan large windows of data, so the number of unique acoustic sources is unknown. To automatically associate and localize range measurements, the proposed method seeks subsets of measurements across unique sensors which are internally consistent. For every considered measurement subset, locations are estimated with single constituent measurements left out and checked to be sufficiently close to the excluded measurement’s set of potential locations. If a measurement subset is entirely consistent in this manner, the measurements are added as neighboring nodes in a graph-based representation, and strongly connected components are used to determine data associations and calculate the final source location estimates. Informed by the methods developed in this thesis, an array of low-cost TOSSIT moorings was deployed in Cape Cod Bay and used to collect experimental PAM data. The localization results are comparable to another similar physics-based inversion approach. Overall, this thesis aims to fill a gap in acoustic data processing methods where data from a low-cost network of unsynchronized acoustic sensors are fused to localize acoustic sources. The presented methods and data processing pipeline demonstrate the great potential of low-cost acoustic sensing systems.

show abstract

Time-Frequency Fused Underwater Acoustic Source Localization Based on Contrastive Predictive Coding

Cited by 7 publications

References 16 publications

Underwater Source Localization via Spectral Element Acoustic Field Estimation

Underwater Source Localization via Spectral Element Acoustic Field Estimation

MGFGNet: an automatic underwater acoustic target recognition method based on the multi-gradient flow global feature enhancement network

Automatic baleen whale detection and 2D localization using a network of unsynchonized passive acoustic sensors

Contact Info

Product

Resources

About