一种采用卡尔曼滤波与正则匹配的信号盲检测法  

A Blind Signal Detection Method Using Kalman Filtering and Regular Matching

作　　者：史飞 牛慧莹[1,2] 王济 吕鹏[1,2] SHI Fei;NIU Huiying;WANG Ji;LYU Peng(Hebei Key Laboratory of Electromagnetic Spectrum Cognition and Control,Shijiazhuang 050081,China;The 54th Research Institute of CETC,Shijiazhuang 050081,China)

机构地区：[1]河北省电磁频谱认知与管控重点实验室,河北石家庄050081 [2]中国电子科技集团公司第五十四研究所,河北石家庄050081

出　　处：《无线电通信技术》2023年第1期182-186,共5页Radio Communications Technology

摘　　要：信号检测是电磁频谱感知的有效手段之一,但受电磁环境非平稳特性、样本数据有限性及傅里叶变换全域能量分布特性等诸多因素影响,使得基于快速傅里叶变换(Fast Fourier Transform,FFT)的频谱数据抖动较大,无法适应低信噪比、大动态范围的信号检测。为了提高信号检测性能,提出了一种采用卡尔曼滤波的频谱累积平滑方法,在此基础上提取了能够反映信号能量分布变化的曲率特征,设计了基于符号正则匹配的方法,实现对信号能量分布特征的检测。介绍了处理方法的主要原理并进行了仿真分析,结果表明在信噪比大于2 dB时,信号检测准确率达到100%,且相比能量检测门限法具有更优的搜索动态范围。Blind signal detection is an effective method for sensing electromagnetic spectrum.However,it is affected by many factors,such as non-stationary characteristics of electromagnetic environment,limited sample data and full-domain energy distribution of Fourier transform,which makes the estimation jitter high and unable to adapt to low SNR and large dynamic range.In order to improve the performance of spectrum detection,a spectrum cumulative smoothing method based on Kalman filtering is proposed.On this basis,curvature features which can reflect the changes of signal energy distribution are extracted.And a regular matching method based on symbols is designed to detect the energy distribution features of the signal.The processing flow and basic principle are introduced,and computer simulation and performance analysis are carried out.The results show that when the SNR is greater than 2 dB,the accuracy of signal detection reaches 100%.And it has better dynamic search range than the energy detection threshold method.

关 键 词：信号盲检测 卡尔曼滤波 曲率 正则表达式 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]