Integrated Platform of Underwater Acoustic Signal Simulation and Classification

With the increase in marine activities, there is a growing need for efficient and accurate underwater acoustic signal processing techniques. In this study, MATLAB-based graphical user interfaces (GUIs) including Receiver Signal Simulation GUI, Database System GUI and Signal Classification GUI are developed for simulating, sorting, classifying and visualizing underwater acoustic signals with the aim of improving the efficiency and accuracy of acoustic signal processing. By converting one-dimensional acoustic signals into two-dimensional time-frequency spectrogram, this study applies image classification techniques in underwater acoustic signal processing using convolutional neural networks (CNNs), deep neural networks (DNNs), and random forests (RFs) for signal classification. Experimental results show that these machine learning models are able to effectively classify acoustic signals in a low signal-to-noise ratio environment, with CNN and RF showing particular robustness. In addition, the GUI provides a user-friendly platform that greatly simplifies the complex process of analyzing acoustic data. This study not only demonstrates effectiveness of the algorithms in classifying underwater acoustic signals, but also provides new tools and perspectives for marine science research and related application areas. The future work will explore the integration of more machine learning algorithms and extend the functionality of the GUI to support a wider range of acoustic signal processing applications.