摘要
该文介绍了一种基于小波分类特征的数字调制信号的识别方法,创新之处在于同时应用了连续小波变换和多层小波分解两种方法提取信号的特征,并且对于不同调制信号采用了不同的分类特征。算法实现时不需要进行码元周期估计以及同步时间估计,从而使分类器的设计变得简单,判决准则简化,提高了运算速度和识别率。
A new method of digital modulation identification with wavelet transform is introduced in this paper. There are two ways to get the characteristics. One is to get the local maximum with the continuous wavelet transform; the other is the multiresolution analysis. Both of them have been used. For different modulated signals, different characteristics have been used. Compared with others, the classifier is easy to realize and the decision is simple. It is not necessary to estimate the code period and the synchronization time. The percentage of correct identification is improved. The speed of modulation identification is increased as well.
出处
《电子与信息学报》
EI
CSCD
北大核心
2006年第11期2026-2029,共4页
Journal of Electronics & Information Technology
基金
国家部级基金资助课题
关键词
数字调制信号识别
连续小波变换
多层小波分解
Digital modulation identification, Continuous wavelet transform, Multiresolution analysis
作者简介
陈健:男,1968年生,副教授,博士生,主要研究方向为无线通信和信号处理.
阔永红:女,1967年生,副教授,博士,主要研究方向为智能信号处理.
李建东:男,1962年生,教授,博士生导师,通信工程学院院长、ISN国家重点实验室主任,主要研究方向为个人通信等.