基于改进型神经网络的水声目标声音分类研究

Underwater acoustic target classification based on Improved Neural Network

主要研究了采用模式匹配的方法对水声目标进行分类的技术,在模式匹配的过程中,引入了人工鱼群算法对传统神经网络进行改进。通过Matlab仿真试验表明:采用AFSA-神经网络可以大大提高对不同声源信号的分类正确率,并大幅降低检测的时间。

in this paper, the method of pattern matching for underwater acoustic target classification is studied. In the process of pattern matching, artificial fish swarm algorithm is introduced to improve the traditional neural network. The simulation results by Matlab show that the AFSA-neural network can greatly improve the classification accuracy of different sound source signals, and greatly reduce the detection time.