多跳频信号参数单通道盲估计算法  

Single channel blind estimation algorithm for parameters of multiple frequency-hopping signals

作　　者：张东伟[1,2] 梁晓龙 王鹏[1,2] 尹晨晓 ZHANG Dongwei;LIANG Xiaolong;WANG Peng;YIN Chenxiao(Air Traffic Control and Navigation School,Air Force Engineering University,Xi'an 710051,China;Shaanxi Key Laboratory of Meta-Synthesis for Electronic and Information System,Xi'an 710051,China;Unit 94654 of the Chinese People's Liberation Army,Nanjing 210046,China)

机构地区：[1]空军工程大学空管领航学院,陕西西安710051 [2]陕西省电子信息系统综合集成重点实验室,陕西西安710051 [3]中国人民解放军94654部队,江苏南京210046

出　　处：《华中科技大学学报（自然科学版）》2023年第6期152-159,165,共9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(61971434);陕西省自然科学基础研究计划项目(2023-JC-QN-0770)。

摘　　要：为了支撑跳频(FH)网台分选、支援干扰及信息恢复等任务,在单通道条件下,提出一种基于起跳时刻(SHT)序列检测分选的多跳频信号参数盲估计算法,并实现了异步网台频率自动分选.首先总结了三类侦察场景截获样本起跳时刻序列的理论分布,推导了基于起跳时刻序列分选结果的跳频信号时域参数理论表达;然后通过一种改进的谱熵算法,结合设计的自适应门限,实现对起跳时刻序列的准确检测/提取,据此将截获样本划分为若干频率驻留期(FRT),通过快速傅里叶变换(FFT)结合魏格纳-威利分布(WVD)实现了频率驻留期内频率分量的准确估计;接着充分利用跳频信号频域跳变属性,通过对相邻频率驻留期内频率分量进行跳频检测,实现起跳时刻序列分选,为各跳频信号时域参数盲估计提供了数据支撑;最后总结了所提算法完整流程.理论分析结合蒙特卡罗仿真结果表明:所提算法对不同跳频组网方式均具适应性,估计精度高,计算简单.In order to effectively support the tasks of frequency hopping(FH)network sorting,interference support and information recovery,a blind estimation algorithm for multi-frequency hopping signal parameters based on start hopping time(SHT)sequence detection and sorting was presented under single channel condition,and an asynchronous network frequency automatic sorting was implemented.Firstly,the theoretical distribution of the start hopping time sequences of three types of reconnaissance scenes was summarized,and the theoretical expression of the time domain parameters of frequency hopping signals based on the sorting results of the start hopping time sequences was deduced.Then,an improved spectral entropy algorithm combined with the designed adaptive threshold was used to detect/extract the start hopping time sequence accurately.Based on this,the intercepted samples were divided into several frequency residence times(FRT),and the accurate estimation of the frequency component in frequency residence times was achieved by combining FFT(fast Fourier transformation)with WVD(Wigner-Ville distribution).Secondly,making full use of the frequency domain hopping property of frequency hopping signal,the start hopping time sequence was sorted by“frequency hopping”detection of the frequency components in adjacent frequency residence times,which provides data support for blind estimation of time domain parameters of each frequency hopping signal.Finally,the complete flow of the proposed algorithm was summarized.Theoretical analysis and Monte Carlo simulation results show that the proposed algorithm is adaptive to different frequency hopping networking modes,with high estimation accuracy and simple calculation.

关 键 词：跳频信号 参数盲估计 单通道 起跳时刻 谱熵 

分 类 号：TN97[电子电信—信号与信息处理]