基于K-means算法改进的SOM神经网络调制识别分类器

Classifier of Modulation Recognition based on K-means Algorithm to Improve SOM Neural Network

通信过程中,获得情报信息的关键步骤是清楚接收到的调制信号的调制方式。随着现代通信技术的高速发展,人工智能广泛应用于调制方式识别领域。提出将自组织特征映射(Self-Organizing feature Map,简称SOM网络)神经网络用于调制制式的识别。用K均值(K-means)聚类算法来寻找每类特征参数的两个聚类中心,并将此聚类中心作为SOM神经网络的初始权值向量。这样,可以降低神经网络的训练次数,同时提高正确识别率。

In communication process,it is very important for obtaining the intelligence information to clear know the modulation method of received modulated signals.With the rapid development of modern communication technology,artificial intelligence is widely used in modulation recognition field.In this paper,Self-Organizing feature Map neural network is proposed for modulation recognition.In order to decrease training time of the neural network and improve recognition probability,K-means clustering algorithm is used to find two clustering centers for each type of characteristic parameters.At the same time,the two clustering centers are as the right weight value vector.