Detection of underwater acoustic target using beamforming and neural network in shallow water. (28th February 2022)

• Record Type: Journal Article
• Authors: Jiang, Junjun; Wu, Zhenning; Huang, Min; Xiao, Zhongzhe
• Is Part Of: Applied acoustics. Volume 189(2022)
• Publication Date: 2022-02-28
• Subjects: Underwater acoustic target -- Beamforming -- Direction of arrival estimation -- Neural network; Acoustical engineering -- Periodicals; Periodicals; 620.2
• DOI: 10.1016/j.apacoust.2021.108626
• Languages: English
• ISSNs: 0003-682X

Abstract

Highlights: A two-stage approach is proposed for underwater acoustic target detection. Beamforming-based DOA estimation acts as a feature extractor. FNN is used as the kernel network to construct a reliable detection model. Underwater acoustic target can be detected effectively by using proposed method. Abstract: Due to the complexity of marine environment, underwater acoustic target detection based on experience of expert observers is not fully reliable. We proposed a two-stage approach for underwater acoustic target detection task. In the first stage, the features extracted by beamforming-based direction of arrival estimation can carry essential information of the targets, and are taken as the input of the second stage. In the second stage, a feedforward neural network (FNN) is used as the kernel network to discover the potential pattern and construct a reliable detection model. The performance of the proposed method is evaluated on ocean underwater acoustic data. Convolutional neural network (CNN) is adopted as the baseline method for comparison. In three different sea areas, the accuracy of detection with the proposed method can reach 93.81%, 97.94%, and 97.32%, respectively, while CNN presents an average accuracy of only 74.17%. The experimental results demonstrate that the proposed method is effective in complex and changing underwater environments with little precise environmental information.