Source localization in the deep ocean using a convolutional neural network

Liu, Wenxu; Yang, Yixin; Xu, Ming; Lü, Ling; Liu, Zongwei; Shi, Yang

doi:10.1121/10.0001020

Cited by 41 publications

(11 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For source ranging, CNNs are built to learn the mapping relationship between the sound data and the source distance. Applications using CNNs for underwater acoustics [9,10] take advantage of the ability of CNNs to exploit the correspondence between sound field features and source location through training. Most CNN-based source localization methods can be roughly separated into regression tasks and classification tasks.…”

Section: Classic Cnns For Source Localizationmentioning

confidence: 99%

“…Various studies have found that ML achieved lower error in source ranging than MFP, especially in complex ocean environments with low SNR [5][6][7]. In addition, progressively increasing network architectures, such as FNN [5,6], TDNN [7,8], CNN [9,10], ResNet [11][12][13], and Inceptions [10], are applied to underwater acoustic localization to mine deep features and decouple sound source information from the underwater acoustic environment. This application of ML to source ranging has been demonstrated to achieve good localization results in both synthetic and experimental data sets collected on the sea.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Localization of Immersed Sources by Modified Convolutional Neural Network: Application to a Deep-Sea Experiment

Wang

Lin

et al. 2021

Sensors

View full text Add to dashboard Cite

A modified convolutional neural network (CNN) is proposed to enhance the reliability of source ranging based on acoustic field data received by a vertical array. Compared to the traditional method, the output layer is modified by outputting Gauss regression sequences, expressed using a Gaussian probability distribution form centered on the actual distance. The processed results of deep-sea experimental data confirmed that the ranging performance of the CNN with a Gauss regression output was better than that using single regression and classification outputs. The mean relative error between the predicted distance and the actual value was ~2.77%, and the positioning accuracy with 10% and 5% error was 99.56% and 90.14%, respectively.

show abstract

Section: Classic Cnns For Source Localizationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Localization of Immersed Sources by Modified Convolutional Neural Network: Application to a Deep-Sea Experiment

Wang

Lin

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…First, this method utilizes the deep neural network (DNN), a purely data-driven approach, to extract features from acoustic data for automatic depth estimation. Recently, many studies present applications of neural networks (NNs) to the underwater acoustic field, such as target localization [16][17][18] , classification [19][20][21][22] etc. According to the simulation and experimental results of the preceding literature, a well-trained NN has a competitive performance in comparison to conventional methods.…”

Section: Introductionmentioning

confidence: 99%

Target depth estimation by deep neural network based on acoustic interference structure in deep water

Guo

Duan

Yang

2022

IET Radar Sonar & Navi

View full text Add to dashboard Cite

Automatic and robust target depth estimation is an important issue of an active detection system working in the reliable acoustic path (RAP) environment. In this paper, the target depth-sensitive acoustic interference structure is used as input, and a deep neural network (DNN) method is proposed to realize automatic depth estimation. Furthermore, to improve the robustness of the network, both the sound speed profile (SSP) uncertainty and the target trajectory uncertainty are considered in the training dataset. The former is modelled as a random weighted sum of the empirical orthogonal function (EOF), and the latter is modelled by the partial-data-lacking of the network input. The method was evaluated through simulated data and compared with the conventional method of matching the measured interference structure with the replicas. The presented method exhibits a higher accuracy of depth estimate in uncertain environments when the signalto-noise ratio (SNR) is higher than −6 dB, but its performance deteriorates rapidly as the SNR decreases further. The method is also verified by semi-simulation with experimental data in deep water.

show abstract

“…Various deep learning (DL) based researches have been conducted in many applications such as image processing, signal detection, and estimation theory. DL-based SLs also VOLUME 4, 2016 have been studied [17][18][19][20][21][22][23][24][25][26]. The DLSLs were implemented by a deep neural network (DNN) or a convolutional neural network (CNN) with preprocessed data [17][18][19][20][21][22][23][24][25][26].…”

Section: Introductionmentioning

confidence: 99%

Deep Learning-Based Near-Field Source Localization Without a Priori Knowledge of the Number of Sources

et al. 2022

View full text Add to dashboard Cite

In this paper, we propose a high resolution grid-based deep learning source localization that precisely estimates the locations of near-field sources without a priori knowledge of the number of sources. The proposed method consists of a principal component analysis network (PCAnet) and a spatial spectrum network (Sp2net). The proposed PCAnet calculates the noise spaces of the received signals by convolutional layers without a priori knowledge or the estimation of the number of sources and has the lower complexity than eigenvalue decomposition (EVD). The proposed Sp2net calculates the spatial spectrum with a linear layer from the output of the PCAnet and classifies dense location grids with a convolutional neural network (CNN). From the spatial spectrum, this paper also proposes an activation function to enlarge the values at the grid points where the near-field sources exist, which are differentiable for all input values. Then, the direction of arrivals (DOAs) and the ranges of the near-field sources are estimated with high resolution. Computer simulations demonstrated that the proposed method had better DOA and range estimation performances than those of the conventional methods.

show abstract

Source localization in the deep ocean using a convolutional neural network

Cited by 41 publications

References 14 publications

Localization of Immersed Sources by Modified Convolutional Neural Network: Application to a Deep-Sea Experiment

Localization of Immersed Sources by Modified Convolutional Neural Network: Application to a Deep-Sea Experiment

Target depth estimation by deep neural network based on acoustic interference structure in deep water

Deep Learning-Based Near-Field Source Localization Without a Priori Knowledge of the Number of Sources

Contact Info

Product

Resources

About