Underwater Acoustic Signals Research Articles

A light-weight neural network for marine acoustic signal recognition suitable for fiber-optic hydrophones

Fiber-optic hydrophones based on optical fiber sensing and communication technology have developed rapidly in recent years. With the advantages of high sensitivity, small size, and large dynamic range, they are easy to detect underwater acoustic signals, showing great potential in the field of marine ecological protection. Researchers can monitor the marine ecological environment by recognizing underwater optical fiber sensing signals collected by fiber-optic hydrophones in real-time. Deep learning models can achieve high recognition accuracy, but they usually have large amounts of parameters that makes them difficult to be deployed in mobile terminals of marine survey ships. To solve the key problem, this paper discussed the design of the Deep Light-Weight Attention Residual Network (DLA-ResNet). It introduces the depth-wise separable convolution (DSC) and the convolutional block attention module (CBAM) into ResNet18 to make its performance better. ResNet18 is a two-dimensional convolutional network (CNN) model applying residual blocks. Experiments verify that the accuracy of DLA-ResNet improved from ResNet18 is increased by about 2.04% and its size is reduced by 85.67%. The proposed scheme in this paper realizes an average recognition rate of 96.50% for eight types of underwater optical fiber sensing signals in the application of marine ecological environment monitoring. Moreover, its model size is only 6.12 MB and it only takes 27.90 ms, which is available to meet the needs of reliability and effectiveness.

Variable-step multi-scale fractal dimension and its application to ship radiated noise

Fractal dimension (FD) is a common tool for detecting dynamic changes in signals, however, it can only reflect the signal information at a single scale and has certain limitations. Currently, several scholars have proposed a few indexes based on multi-scale coarse-grained processing (MCP) and refined composite MCP (RCMCP). Nevertheless, the coarse-grained characteristics of MCP and RCMCP lead to their inadequate extraction of information from time series in the calculation process. To address this issue, variable-step multi-scale FD (VSMFD) was proposed, which not only reflects signal information from multiple scales, but also is more accurate and comprehensive when extracting signal information. The results of three sets of simulation experiments indicate that VSMFD is the most sensitive to the dynamic transform of chirped signals, and has the best separability for simulated noise signals and chaotic signals. In addition, both sets of ship radiated noise experiments indicate that VSMFD performs better in distinguishing ship radiated noise than indexes based on MCP and RCMCP, and has broad application prospects in the field of underwater acoustic signal processing.

Underwater Acoustic Signal Acquisition and Sensing Using a Ring Vector Sensor Communication Receiver: Theory and Experiments.

Signal acquisition and sensing in underwater systems and applications is typically a challenging issue due to the small signal strength within the background noise. Here, we present a ring vector sensor communication receiver that can significantly improve signal acquisition, by utilizing the underwater acoustic vector field components, compared to the scalar component. The vector sensor receiver is a multichannel receiver that measures particle velocities, which are vector components of the underwater acoustic field, in addition to the scalar field component. According to the combination of our measured experimental data with our signal acquisition performance analysis, the introduced ring vector sensor receiver exhibits higher signal acquisition probabilities for the vector components compared to the scalar component. This can be attributed to certain characteristics of the vector field components. Another advantage of this multichannel receiver is that combining all of its channels can further increase the signal acquisition and packet detection probability in underwater communication systems compared to a single-channel approach.

Stacking Ensemble Learning for the Modulation Recognition of Underwater Acoustic Communication Signals

Effective detection and recognition of underwater acoustic communication signals is of great significance in the field of ocean exploration. In this paper, a novel deep-learning recognition method of modulation modes of underwater acoustic communication signals is proposed. The stacking ensemble learning framework is established on the basis of several single deep-learning classifiers with different network structures and parameter configurations. The proposed method achieved an average recognition accuracy rate of more than 95% on the dataset generated under a simulation environment. The performance is also verified by means of a comparison with the classic SVM method.

Line spectrum extraction method of underwater targets based on the polarization characteristic

In order to suppress the environmental noise interference of underwater targets, a line spectrum extraction method is proposed based on the polarization characteristic of underwater acoustic signals received by a single vector sensor. In this paper, the polarization filtering function of a fourth-order cumulant diagonal slice is developed, and the fourth-order cumulant diagonal slice characteristic spectrum corresponding to the maximum singular value decomposition is used to extract the line spectrum. The simulation and experimental results show that the proposed method can extract the line spectrum signal adaptively, which has a good abatement effect on environmental noise interference.

Feature Fusion Based on Graph Convolution Network for Modulation Classification in Underwater Communication.

Automatic modulation classification (AMC) of underwater acoustic communication signals is of great significance in national defense and marine military. Accurate modulation classification methods can make great contributions to accurately grasping the parameters and characteristics of enemy communication systems. While a poor underwater acoustic channel makes it difficult to classify the modulation types correctly. Feature extraction and deep learning methods have proven to be effective methods for the modulation classification of underwater acoustic communication signals, but their performance is still limited by the complex underwater communication environment. Graph convolution networks (GCN) can learn the graph structured information of the data, making it an effective method for processing structured data. To improve the stability and robustness of AMC in underwater channels, we combined the feature extraction and deep learning methods by fusing the multi-domain features and deep features using GCN. The proposed method takes the relationships among the different multi-domain features and deep features into account. Firstly, a feature graph was built using the properties of the features. Secondly, multi-domain features were extracted from the received signals and deep features were extracted from the signals using a deep neural network. Thirdly, we constructed the input of GCN using these features and the graph. Then, the multi-domain features and deep features were fused by the GCN. Finally, we classified the modulation types using the output of GCN by way of a softmax layer. We conducted the experiments on a simulated dataset and a real-world dataset, respectively. The results show that the AMC based on GCN can achieve a significant improvement in performance compared to the current state-of-the-art methods. Our approach is robust in underwater acoustic channels.

Application of a Randomized Algorithm for Extracting a Shallow Low-Rank Structure in Low-Frequency Reverberation

The detection performance of active sonar is often hindered by the presence of seabed reverberation in shallow water. Separating the reverberations from the target echo and noise in the received signal is a crucial challenge in the field of underwater acoustic signal processing. To address this issue, an improved Go-SOR decomposition method is proposed based on the subspace-orbit-randomized singular value decomposition (SOR-SVD). This method successfully extracts the low-rank structure with a certain striation pattern. The results demonstrate that the proposed algorithm outperforms both the original Go algorithm and the current state-of-the-art (SOTA) algorithm in terms of the definition index of the low-rank structure and computational efficiency. Based on the monostatic reverberation theory of the normal mode, it is established that the low-rank structure is consistent with the low-frequency reverberation interference striation. This study examines the interference characteristics of the low-rank structure in the experimental sea area and suggests that the interferences of the fifth and seventh modes mainly control the low-rank structure. The findings of this study can be applied to seafloor exploration, reverberation waveguide invariant (RWI) extraction, and data-driven reverberation suppression methods.

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.

A deep-learning based high-gain method for underwater acoustic signal detection in intensity fluctuation environments

To achieve efficient underwater acoustic (UWA) signal detection in low signal-to-noise ratio (SNR) scenarios, the transmitted signals are designed with a large time-bandwidth product to get a high detection gain. The linear correlator (LC) is considered as the maximum SNR detector, whose detection gain is proportional to the time-bandwidth product. However, the detection performance of LC degrades significantly in time-variant multipath UWA channel and non-Gaussian UWA ambient noise. In this study, we present a deep-learning based two-stage UWA signal detection method in intensity fluctuation environments. This method takes the advantages of the Conv-TasNet and Encoder-Decoder network, which utilizes an encoder module to extract signal features, a separation module to enhance the signal components and then another decoder module to reconstruct the transmitted signal. To demonstrate the performance of the proposed method, the datasets used for training and testing originated from the ASIAEX 2001 South China Sea (SCS) experiment. The experimental results show that our model outperforms the classical LC and channel estimation based LC (CE-LC) in constant false alarm rate (CFAR) detection and also surpasses the TCDAE and Conv-Tasnet models as evaluated by DSI-SNR1 and DSI-SNR2.

Self-supervised learning–based underwater acoustical signal classification via mask modeling

The classification of underwater acoustic signals has garnered a great deal of attention in recent years due to its potential applications in military and civilian contexts. While deep neural networks have emerged as the preferred method for this task, the representation of the signals plays a crucial role in determining the performance of the classification. However, the representation of underwater acoustic signals remains an under-explored area. In addition, the annotation of large-scale datasets for the training of deep networks is a challenging and expensive task. To tackle these challenges, we propose a novel self-supervised representation learning method for underwater acoustic signal classification. Our approach consists of two stages: a pretext learning stage using unlabeled data and a downstream fine-tuning stage using a small amount of labeled data. The pretext learning stage involves randomly masking the log Mel spectrogram and reconstructing the masked part using the Swin Transformer architecture. This allows us to learn a general representation of the acoustic signal. Our method achieves a classification accuracy of 80.22% on the DeepShip dataset, outperforming or matching previous competitive methods. Furthermore, our classification method demonstrates good performance in low signal-to-noise ratio or few-shot settings.

Modulation Recognition Method for Underwater Acoustic Communication Signals Based on Passive Time Reversal-Autoencoder with the Synchronous Signals.

In the process of the modulation recognition of underwater acoustic communication signals, the multipath effect seriously interferes with the signal characteristics, reducing modulation recognition accuracy. The existing methods passively improve the accuracy from the perspective of selecting appropriate signal features, lacking specialized preprocessing for suppressing multipath effects. So, the accuracy improvement of the designed modulation recognition models is limited, and the adaptability to environmental changes is poor. The method proposed in this paper actively utilizes common synchronous signals in underwater acoustic communication as detection signals to achieve passive time reversal without external signals and designs a passive time reversal-autoencoder to suppress multipath effects, enhance signals' features, and improve modulation recognition accuracy and environmental adaptability. Firstly, synchronous signals are identified and estimated. Subsequently, a passive time reversal-autoencoder is designed to enhance power spectrum and square spectrum features. Finally, a modulation classification is performed using a convolutional neural network. The model is trained in simulation channels generated by Bellhop and tested in actual channels which are different from the training period. The average recognition accuracy of the six modulated signals is improved by 10% compared to existing passive modulation recognition methods, indicating good environmental adaptability as well.

Time of arrival estimation for underwater acoustic signal using multi-feature fusion

Underwater target detection (UTD) is widely applied in ocean exploration. A key issue for the time-of-arrival (ToA)-based UTD algorithm is to localize the first direct path in an underwater multipath channel. However, the conventional Matched Filter (MF) suffers from a deterioration of the estimation accuracy when the direct arrival of the received signal is non-strongest. Therefore, this paper proposes a multi-feature fusion ToA estimation algorithm. First, a multi-decision strategy including the coarse ToA range estimation and the fine ToA estimation is proposed to realize a balance between computational complexity and estimation accuracy. Second, the coarse ToA range estimation is realized by multi-level spectrum analysis. Then, the coarse ToA range is used to localize the fine ToA utilizing the multi-feature fusion ToA estimation. Finally, the simulation and experimental results are provided to validate the proposed ToA estimation algorithm. It clearly shows that without training data, the proposed algorithm can not only deal with the multipath acoustic channel problem but also improve the accuracy of the ToA estimation compared to other algorithms.

Design, development, fabrication and evaluation of the dynamics of a graphene based underwater acoustic vector sensor: A simulation and experimental study

In the last decade, the usage of underwater vector sensor has gained a lot of importance in the detection of low frequency underwater acoustic signals, as the noise level radiated by the submarine has been drastically reduced. Acoustic sensor based on Micro Electro Mechanical Systems (MEMS) technology is the recent advancement in the area of underwater acoustic vector sensors. We are reporting the design, fabrication, preliminary characterization and evaluation of the dynamics of a two-dimensional underwater acoustic vector sensor by simulation and experimental study. The vector sensor is realized using Reduced Graphene Oxide (rGO) as sensing element and Kapton as flexible substrate, to yield better sensitivity compared to the conventional acoustic vector sensor. Design of vector sensor is inspired by the biological sensing system of fish. The application of piezoresistive transduction principle and MEMS technology helps in miniaturisation and low frequency acoustic signal detection. The fabrication process is made simple, scalable and cost minimum by using the method of drop casting for the deposition of rGO on flexible substrate kapton, thereby eliminating the requirement of clean room. COMSOL Multiphysics 5.5 version is used to simulate acoustic vector sensors using polysilicon and rGO on substrates, silicon and flexible kapton respectively. The simulation results indicate that the sensitivity of rGO based acoustic vector sensor is −149.47 dB which is better in comparison with the sensitivity of −173.37 dB of conventional polysilicon vector sensor. By simulation, the natural frequency of the sensor is found to be 46.62 Hz, which closely matches with the theoretical value 41.06 Hz of the device as per the design criteria and hence, it could detect the low frequency acoustic signals. The displacement sensitivity of the fabricated sensor is found to be 0.0264 V/mm and 0.0276 V/mm for applied displacements in axial and radial directions respectively. The device is tested for its performance parameters i.e., sensitivity and directivity, by performing an acoustic test in the presence of an elastic medium, water. This test results infer that, the device is able to detect and give smooth output response for the acoustic signals in the low frequency range. The Y and X-channel receiving sensitivity exhibited by the sensor is found to be −128.04 dB to −134.93 dB and −130.61 dB to −136.78 dB respectively. Directivity pattern of the sensor is found to be in the shape of 8 with symmetry. As the device is a vector sensor, in addition to the scalar pressure measurement, it should be able to provide additional incentives such as acoustic particle displacement and acoustic particle velocity at any point in the acoustic field. In this paper, an experimental setup is made to generate deep and intermediate water waves, thereby attempting to simulate ocean conditions in the laboratory. Subsequently the numerical evaluation of the wave kinematics, i.e., fluid particle displacement and fluid particle velocity are performed by applying Airy's theory. An effort is also made to analyse the wave kinematics for various water depths, varying from Surface Water Level (SWL) till the sea bed. In this study, wave kinematics are analysed numerically by using experimental data and also by modelling Fluid-Structure Interactions in COMSOL Multiphysics environment. Our results show that the fluid flow of velocity 0.31 m/s results in fluid particle velocity of 0.125 m/s at time 0.215 s (around f = 5 Hz) by both the methods and fluid particle displacement is found to be 19.66 μm. The experimental test setup and videos are made publicly available1: 1https://sites.google.com/view/smithapaib/downloads

Characteristics of Internal Solitary Waves in the Timor Sea Observed by SAR Satellite

Internal solitary waves (ISWs) with features such as large amplitude, short period, and fast speed have great influence on underwater thermohaline structure, nutrient transport, and acoustic signal propagation. The characteristics of ISWs in hotspot areas have been revealed by satellite images combined with mooring observation. However, the ISWs in the Timor Sea, which is located in the outflow of the ITF, have not been studied yet and the characteristics are unrevealed. In this study, by employing the Synthetic Aperture Radar (SAR) images taken by the Sentinel-1 satellite from 2017 to 2022, the temporal and spatial distribution characteristics of ISWs in the Timor Sea are analyzed. The results show that most of ISWs appear in Bonaparte basin and its vicinity. The average wavelength of the ISWs is 248 m, and most of the wave lengths are less than 400 m. The peak line of ISWs is longer in deeper water. The underwater structures of two typical ISWs are reconstructed based on the Korteweg–de Vries (KdV) equation combined with mooring observation. This shows that, compared with the two-layer model, the continuous layered model is more suitable for reconstructing the underwater structures of ISWs. Further analysis shows that both the rough topography and the spring-neap tides contribute to the generation of ISWs in the Timor Sea. This study fills a gap in knowledge of ISWs in regional seas, such as the Timor Sea.

Absorption of echo signal for underwater acoustic signal target system using hybrid of ensemble empirical mode with machine learning techniques

Absorption of echo signal for underwater acoustic signal target system using hybrid of ensemble empirical mode with machine learning techniques

MSLEFC: A low-frequency focused underwater acoustic signal classification and analysis system

Many feature extraction techniques are employed to study the classification of underwater acoustic, however, most of these techniques result in the loss of detailed information in this field. A Multi-scale short-time Fourier transform (MS-STFT) method was proposed to improve the low-frequency information and maintain the detailed information by increasing the number of channels. An effective data augmentation approach was proposed and a Ladder-like Encode (LE) architecture was built to increase the generalizability of the model and the classification accuracy for the features extracted. Finally, A Frequency-CAM (FC) method is proposed to analyze the frequency band locations that the neural network is interested in when conducting classification tasks on various classes of data. The integration of the above technologies is called MSLEFC. The performance of this system has been tested on two ship radiation noise datasets and obtained 82.94% and 96.06% accuracy, respectively. On the ShipsEar dataset, the proposed method increases the accuracy by 0.7% over the previous state-of-the-art method. The proposed model architecture has significant improvements in parameter size and computation compared to ResNet50.

DOA Estimation of Underwater Acoustic Array Signal Based on Wavelet Transform With Double Branch Convolutional Neural Network

In underwater acoustic communications, a hydrophone array is often used to receive the underwater acoustic signals to improve the gain of the received signal. The direction of arrival (DOA) estimation is a basic task in underwater acoustic array signal processing. In underwater signal transmission, due to the channel complexity, a large signal transmission loss, and heavy noise interference, the received signal is seriously distorted, and the DOA estimation is poor. In order to improve the accuracy of DOA estimation for underwater acoustic array signals, in this work, we propose continuous wavelet transform with convolutional neural network (CWT-CNN) method. We use a linear factor to improve the calculation process of continuous wavelet transform. The characteristics of time-frequency fusion of signals are calculated by constructing time-frequency array model with modified wavelet factors. The fused features are then used to train the proposed double branch CNN. As compared with other DOA algorithms based on neural networks, the proposed algorithm shows a 5%–20% higher accuracy and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$0.479^{\circ }-1.172^{\circ }$</tex-math></inline-formula> lower RMSE at 7 different values of SNR. In addition, the proposed algorithm is less computationally complex.

Recognition of Whistle-like Signals Based on FM Harmonic Characteristics

While human hearing can easily identify whistle-like sounds made by dolphins and whales, manually identifying them from the increasingly variety and quantity of underwater acoustic signals (such as other marine life sounds and underwater acoustic communication signals) is undoubtedly laborious. To solve this question, a recognition method based on the frequency modulated feature and harmonic characteristic is analyzed in this paper. A harmonic energy analysis algorithm is proposed and is compared to a harmonic detection method. The experimental results prove that the proposed method could identify the whistle-like sounds with an accuracy over 90% and the traditional one could only reach 50%. The recognition method may be applied to the investigation of marine biological habits and the monitoring of marine ecological environment, and can also potentially be further applied to the detection of bionic communication signal.

Underwater multi-source DOA tracking using uniform linear array based on improved GM-PHD filter

Uniform linear array (ULA) is commonly used for collecting underwater acoustic signals, and the direction-of-arrival (DOA) is obtained by array signal processing methods. The ULA-based multi-source DOA tracking suffers from the performance degradation caused by the presence of false tracks from port-starboard ambiguity. To address this problem, we propose a bearing ambiguity discrimination algorithm. Firstly, the sum of real bearing motion velocity between truth and ambiguous targets is established with the Gaussian distribution. Secondly, an improved Gaussian mixture probability hypothesis density filter, which can adaptively generate target birth intensity, is utilized to select truth and ambiguous bearings generated by the same target from the estimated posterior intensity. Thirdly, truth and ambiguous bearings are discriminated according to the fact that the change of array heading will lead to a significant change in ambiguous bearing. Simulation and real experimental results illustrate that the proposed algorithm can provide unambiguous target-generated tracks and accurately estimate the number of targets.

Research on Very-Low-Frequency Hydroacoustic Acoustic Velocity Sensor Based on DFB Fiber Laser

We have conducted a study on a very-low-frequency acoustic-velocity sensor which is based on a cantilever of distributed-feedback (DFB) fiber laser immersed in castor oil. A mathematical model of the frequency dependent response of the proposed sensor to the acoustic pressure signal influenced by the fluid viscosity is established. We have fabricated the proposed sensor and conducted experimental measurements in the standing wave tube. The results show that the sensor has an average phase sensitivity of −179.5 dB (0 dB = 1 rad/μPa) with ±1.45 dB fluctuation over the frequency range of 20–38 Hz. It has good cosine directivity with a directivity index of 32.5 dB and axial maximum asymmetry of 0.4 dB. The sensor presents promising applications for detecting very-low-frequency underwater acoustic signals.