摘要
针对具体的军用超短波信号,提出一种基于模糊神经网的改进信号识别算法。该算法利用模糊神经网实现模糊推理系统,采用分步识别的方法,将特征参数映射到模糊空间进行二分类。仿真表明:在具有高斯加性白噪声的环境中,信噪比高于15dB时,系统识别率高于95%。
An improved recognition arithmetic based on the fuzzy neural network is presented for the ultra-short signal used in the martial communications. This arithmetic forms an fuzzy illation system with the neural network, mapping the characters to the fuzzy space in order to classify the characters into two species. The simulation shows that all the ultra-short signals are classified with success rate more than 95% when SNR is higher than 15dB.
出处
《成都信息工程学院学报》
2007年第5期588-592,共5页
Journal of Chengdu University of Information Technology
关键词
调制
特征参数
模式识别
modulation
characters
pattern recognition
