【作者】 刘杨；

【导师】 刁鸣；

【作者基本信息】 哈尔滨工程大学 ， 电子科学与技术， 2023， 硕士

【摘要】 跳频(Frequency Hopping,FH)信号的参数估计是侦察工作中的关键环节。跳频信号是典型的非平稳信号,其信号载波频率能够按照一定规律随机跳变,且通信频带中通常还夹杂着各种强干扰信号与噪声,因此如何从接收信号中准确提取跳频信号并精确估计出参数成为了研究的难点。随着电子对抗环境不断升级,跳频通信采用的载频频段不断升高、跳频范围不断扩大,这对阵列信号处理算法和相应的硬件设施提出了严峻考验,在采用宽频接收机进行侦察时,以现有的ADC采样技术实现高频信号时域奈奎斯特采样会非常昂贵复杂,且所得到的采样数据量庞大,直接对其进行处理会占用大量的资源和时间,同时对于空域数据的接收,高频信号也会限制接收阵列的最小间距,若以入射信号半波长的间距布置阵元会导致阵元间存在严重耦合,因此高频段跳频信号在时空域欠采样条件下的参数估计算法具有重要的研究价值。本文立足于这样的背景,基于分析非平稳信号的有力手段——时频分析,重点研究针对跳频信号的时、频、空三个维度参数的估计算法,具体工作内容如下:(1)将传统时频分析方法和时频后处理方法应用于跳频信号的分析,通过仿真定性定量比较不同时频分析方法在分析跳频信号时的时频聚集性及计算量,并给出了适用于跳频信号的时频分析策略:针对跳频信号的时频特征,可在传统时频分析方法所得结果的基础上利用时频后处理方法进行局部分析,在大幅降低计算成本的同时提高时频聚集性。(2)现有的跳频信号参数估计算法将时频分析方法与图像处理技术相结合,能够实现复杂电磁环境下的跳频信号参数估计,但因其大多采用传统时频分析方法,时频聚焦性的限制以及图像处理带来的数据区域扩大的问题,导致算法估计精度较低,为此本文利用基于区域能量重心思想的时频后处理方法优化传统跳频信号参数算法,实现跳频信号的时频空多维参数高精度估计,并基于此算法利用均匀线阵搭建了奈奎斯特采样下基于时频分析的跳频信号时频空多维参数估计框架。仿真实验证明了所提奈奎斯特采样下跳频信号时频空多维参数估计框架的高精度性。(3)基于闭式鲁棒中国余数定理的时空欠采样参数估计算法对接收信号存在限制,无法实现复杂电磁环境下的跳频信号参数估计,为此本文首次提出将时频分析与闭式鲁棒中国余数定理相结合,从复杂电磁环境下的接收数据中提取余数实现信号参数重构。同时因时频分析的频谱同样存在栅栏效应和频谱泄露的问题,且基于闭式鲁棒中国余数定理的数据重构算法对余数误差存在限制范围,为此本文引入频谱校正算法对通过基于时频分析的参数估计算法所得到的频谱进行高精度校正,以提高参数重构算法的成功率。本文基于以上算法利用非均匀阵列搭建了时空欠采样下基于时频分析的跳频信号时频空多维参数估计框架,仿真实验证明了所提框架的有效性。更多还原

【Abstract】 Parameter estimation of frequency hopping signal is a key link in reconnaissance work.The carrier frequency of the frequency hopping signal can jump randomly according to certain rules,and the communication frequency band is usually mixed with various strong interference signals and noise.Therefore,how to accurately extract the frequency hopping signal from the received signal and accurately estimate the parameters has become a research difficulty.With the continuous upgrading of the electronic countermeasure environment,the carrier frequency band used in frequency hopping communication is increasing and the frequency hopping range is expanding.It is very expensive and complicated to realize the time domain Nyquist sampling of high-frequency signals with the existing ADC sampling technology.At the same time,for the reception of spatial data,high-frequency signals will also limit the minimum spacing of the receiving array.Therefore,it is of great research value to study the parameter estimation algorithm of high-frequency frequency hopping signals under the condition of time-space domain undersampling.Based on this background,based on the powerful means of analyzing non-stationary signals-time-frequency analysis,this thesis focuses on the estimation algorithm of time,frequency and space parameters of frequency hopping signals.The specific work is as follows :(1)Applying the traditional time-frequency analysis method and time-frequency post-processing method to the analysis of frequency hopping signals,and comparing the time-frequency aggregation and computational power of different time-frequency analysis methods qualitatively and quantitatively through simulation,the time-frequency analysis strategy applicable to frequency hopping signals is given: for the time-frequency characteristics of frequency hopping signals,the time-frequency post-processing method can be used for local analysis on the basis of the results obtained by the traditional time-frequency analysis method,which can improve the time-frequency aggregation while significantly reducing the computational cost.(2)The existing parameter estimation algorithm of frequency hopping signal combines time-frequency analysis method with image processing technology,which can realize the parameter estimation of frequency hopping signal in complex electromagnetic environment.However,because most of them adopt traditional time-frequency analysis method,the limitation of time-frequency focusing and the problem of data area expansion caused by image processing lead to low estimation accuracy of the algorithm.Therefore,this thesis uses the time-frequency post-processing method based on the idea of regional energy center to optimize the traditional frequency hopping signal parameter algorithm,so as to realize the high-precision estimation of time-frequency-space multi-dimensional parameters of frequency hopping signal.Based on this algorithm,a time-frequency-space multi-dimensional parameter estimation framework for frequency hopping signals based on time-frequency analysis under Nyquist sampling is built by using uniform linear array.The simulation results show the high accuracy of the proposed multi-dimensional parameter estimation framework for frequency hopping signals under Nyquist sampling.(3)The time-space undersampling parameter estimation algorithm based on the closed robust Chinese remainder theorem has limitations on the received signal and cannot realize the parameter estimation of frequency hopping signal in complex electromagnetic environment.Therefore,this thesis proposes for the first time to combine time-frequency analysis with the closed robust Chinese remainder theorem to extract the remainder from the received data in complex electromagnetic environment to realize signal parameter reconstruction.At the same time,the spectrum of time-frequency analysis also has the problem of fence effect and spectrum leakage,and the data reconstruction algorithm based on the closed robust Chinese remainder theorem has a limit range for the remainder error.Therefore,this thesis introduces the spectrum correction algorithm to improve the success rate of the parameter reconstruction algorithm.Based on the above algorithms,this thesis uses non-uniform arrays to build a time-frequency-space multi-dimensional parameter estimation framework for frequency hopping signals based on time-frequency analysis under spatio-temporal undersampling.The simulation implementation proves the effectiveness of the proposed framework.更多还原