复杂电磁环境下多跳频信号盲检测  

Blind detection of multi-frequency hopping signals under complex electromagnetic environment

作　　者：李红光[1] 郭英[1] 齐子森[1] 苏令华[1] LI Hongguang;GUO Ying;QI Zisen;SU Linghua(Infomiation and Navigation College,Air Force Engineering University,Xi'an 710077,China)

机构地区：[1]空军工程大学信息与导航学院,陕西西安710077

出　　处：《华中科技大学学报（自然科学版）》2020年第7期13-19,共7页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(61601500);军队研究生资助课题(JY2018C169)。

摘　　要：为了在复杂电磁环境中实现多跳频信号盲检测,提出一种基于时频图连通域特征的多跳频信号检测算法.首先利用短时傅里叶变换与wigner-ville分布(STFT&WVD)组合时频方法完成时频变换,保证时频图的时频分辨率和交叉项抑制,并利用自适应二维维纳滤波去除背景噪声,提高算法抗噪性能;然后采用自适应阈值二值化算法对时频图二值化处理并进行8邻域连通域标记,提取每个连通域的特征组成分类特征集;最后利用改进的K均值聚类算法完成特征集分类,根据分类集统计结果和检测条件实现跳频信号检测.仿真结果表明:本文算法能够有效克服定频干扰、突发干扰和扫频干扰;在低信噪比条件下,算法聚类稳定性较好,跳频检测成功率较高.In order to realize blind detection of frequency hopping signals in complex electromagnetic environment,a multi-hopping signal detection algorithm based on connected domain feature of time-frequency diagram was proposed.Firstly,the time-frequency transform was implemented by short-time fourier transform&wigner-ville distribution(STFT&WVD)combined time-frequency method to ensure time-frequency resolution and cross-term suppression.The adaptive two-dimensional Wiener filtering was used to filter out background noise and improve the anti-noise performance of the algorithm.Then,the adaptive threshold binarization algorithm was used to binarize the time-frequency diagram and perform 8 neighborhood connected domain labeling,so as to extract the classification features of each connected domain.Finally,the improved K-means clustering algorithm was used to complete the feature set classification.The frequency hopping signal detection was implemented according to the statistical result of the classification set and the detection condition.Simulation results show that the proposed algorithm can effectively overcome fixed-frequency interference,burst interference and sweeping interference.Under low signal-to-noise ratio(SNR)conditions,the algorithm has good clustering stability and high detection probability.

关 键 词：跳频检测 时频图 二值化 连通域标记 K均值聚类 

分 类 号：TN914.41[电子电信—通信与信息系统]