Underwater sound source positioning method based on deep learning

Abstract

The invention discloses an underwater sound source positioning method based on deep learning, and the method comprises the steps: carrying out normalization operation of vector data simulated throughemploying a KRANEN program, carrying out the superposition of a 0-mean Gaussian random noise complex vector n, and obtaining the simulation sound field data p (f) at a frequency f; constructing a normalized covariance matrix H according to the analog sound field data p (f), performing Hermitian decomposition on the matrix H, and converting the complex matrix H into a real matrix capable of being processed by the convolutional neural network to obtain input data of the convolutional neural network; and training the convolutional neural network by using the input data to obtain an underwater sound source positioning prediction model, and predicting the distance and depth of the signal source according to the observed sound field data. LeNet-5 convolutional neural network and 56-layer deep residual network are used for underwater sound source positioning under single and multiple sound sources. An underwater sound source positioning algorithm with high precision and accuracy is acquired and the real-time performance of underwater sound source positioning is improved.

Description

technical field [0001] The invention relates to an underwater sound source localization method, in particular to an underwater sound source localization method based on deep learning, and belongs to the technical field of signal processing. Background technique [0002] Underwater sound source localization refers to the technology of using underwater sound waves and electronic technology to determine the direction and distance of underwater sound sources. According to the underwater acoustic signal data received by the receiving array composed of several hydrophones, the underwater sound source location is carried out after a certain data processing process. [0003] At present, Matching Field Processing (MFP) technology is a representative method for passive detection of underwater targets. It combines signal processing technology and underwater acoustic physics, and uses channel characteristics, narrow bandwidth and other related technologies to process The sound source d...

AI Technical Summary

Problems solved by technology

However, the shortcomings of the three methods used in this paper are also obvious (the three-layer feedforward neural network is a relatively simple neural network structure, which needs to use a more complex network structure to improve the prediction effect; When it is large, it will take up a lot of system memory and training time, and increasing the data set after the effect reaches a certain level cannot improve the experimental results better, and it cannot better realize the target detection of multiple sets of data; random forest is used in many data It has been demonstrated that if in some classification and regression problems, the data has large noise and there is a certain variable that needs to be divided into multiple categories, then the prediction effect of the random forest on the variable will be greatly reduced), and this paper only analyzes the sound The source distance and single sound source are predicted, which cannot meet the requirements of practical applications. Therefore, it is necessary to use a more complex and efficient neural network structure to design a sound source that can simultaneously predict the sound source distance, depth and have a sound source. Underwater sound source localization method with higher precision and accuracy

Images