低信噪比下的数字调制信号自动识别

提出了一种基于神经网络的数字调制信号识别。首先利用升余弦滤波器滤波，然后提取了5个用于识别的特征参数，利用神经网络分类器进行数字凋制识别。神经网络分类器采用了多层组合的神经网络分类器，不需要设定判决门限，而且在收敛速度、训练时间以及识别率方商都有很大改进。仿真结果表明，在信噪比大于4dB时，系统的正确识别率可达95％以上。这种低信噪比下快速有效的调制识别方法易于实时应用和工程实现。

Automatic Recognition of Digital Modulation Signal Under Lower SNR

A digital modulation signal recognition based on neural network is proposed in this paper. First, the signal is preprocessed by the raised cosine filter, and then five characteristic parameters for recognition are extracted, and then by neural network classifier, the digital signal is identified. The neural network classifier adopts muhilayer-combined neural network classifier and requires no decision threshold, and is improved much in convergence speed, training time and recognition ratio. Simulation results indicate that, when SNR is not less than 4 dB, the correct recognition rate is above 95%. This fast and efficient modulation recognition method under low SNR could be easily implemented in real lime and in engineering project.