基于卷积神经网络对磁异常信号的识别研究

航空磁探反潜作为航空反潜的重要手段,在其中发挥了重要作用。针对目前航空反潜作战中,磁干扰信号极大地影响对水下目标磁探测效果这个问题,文章先对输入信号进行预处理,并使用卷积神经网络实现对2种信号的识别。实验结果显示,卷积神经网络的方法对信号的识别率达到了85%,能够有效对信号进行准确地识别。

Recognition of Magnetic Anomaly Signal Based on Convolutional Neural Network

As an important means of aviation anti-submarine, aviation magnetic anti-submarine played an important role.In current aviation anti-submarine warfare,the magnetic disturbance signal greatly affects the magnetic detection effect ofunderwater target. The article first preprocessed the input signal, and used the convolution neural network to recognize thetwo signals. The experimental results showed that the signal recognition rate of the convolutional neural network reached85%, which could effectively recognize the signal accurately.