一种稳健的通信辐射源个体识别方法

通信辐射源个体识别是目前通信对抗领域研究热点与难点问题,相对于雷达辐射源,通信辐射源信号弱、瞬时特征不明显导致个体识别更复杂、更困难。利用通信辐射源信号的长时谱统计特性,提取信号功率谱峰值特征和包络模板,构造通信辐射源个体特征向量,通过朴素贝叶斯分类算法与个体特征矢量相结合,在训练样本数目足够大的条件下可进行有效识别。测试实验表明,识别方法稳健有效,可在信噪比5 d B情况下实现93.7%的正确识别概率。...

Full description

Saved in:
Bibliographic Details
Published in电讯技术 Vol. 55; no. 3; pp. 321 - 327
Main Author 黄欣 郭汉伟
Format Journal Article
LanguageChinese
Published 中国西南电子技术研究所,成都,610036%北京科航军威科技有限公司,北京,100044 2015
Subjects
包络特征
峰值提取
模式识别
模板匹配
通信对抗
通信辐射源个体识别
communication countermeasure
通信辐射源个体识别
模式识别
specific communication emitter identification
包络特征
envelope features
template matching
峰值提取
通信对抗
peak extraction
模板匹配
pattern recognition
Online AccessGet full text
ISSN1001-893X
DOI10.3969/j.issn.1001-893x.2015.03.016

Cover

More Information
Summary:通信辐射源个体识别是目前通信对抗领域研究热点与难点问题,相对于雷达辐射源,通信辐射源信号弱、瞬时特征不明显导致个体识别更复杂、更困难。利用通信辐射源信号的长时谱统计特性,提取信号功率谱峰值特征和包络模板,构造通信辐射源个体特征向量,通过朴素贝叶斯分类算法与个体特征矢量相结合,在训练样本数目足够大的条件下可进行有效识别。测试实验表明,识别方法稳健有效,可在信噪比5 d B情况下实现93.7%的正确识别概率。
Bibliography:communication countermeasure ; specific communication emitter identification ; peak extraction ;envelope features ; template matching; pattern recognition
HUANG Xin, GUO Hanwei （ 1. Southwest China Institute of Electronic Technology, Chengdu 610036, China; 2. Beijing Kehangjunwei Technology Co. , Ltd. , Beijing 100044, China）
Specifc communication emitter identification is a hot topic and difficult issue in the field of com- munications countermeasures and specific emitter idendification of communication emitter is more difficult and complex than that of radar emitter due to weaker signal and transient characteristics. In this paper, the long-time spectrum feature of specific communication emitter is used to extract the peak and envelope tem- plates in power spectrum domain and construct the specific vector feature. Combining the Naive Bayes clas- sifier with the specific vector feature can effectively identify the communication emitters under the condition of enough training sample number. Actual test results
ISSN:1001-893X
DOI:10.3969/j.issn.1001-893x.2015.03.016