【作者】 李宏；

【导师】 黎湘；

【作者基本信息】 国防科学技术大学 ， 信息与通信工程， 2010， 博士

【摘要】 现代战争条件下,电子对抗与反对抗的斗争日趋激烈。对敌方雷达的侦察和定位是对其实施精确打击的前提,也是掌握战争主动权的关键。对大时宽带宽积线性调频(LFM)信号和多相编码信号等低截获概率(LPI)雷达脉冲压缩信号的分析处理是其中的核心问题。论文针对LPI雷达脉冲压缩信号的调制方式识别、调制参数估计及多普勒频率变化率高精度测量等关键技术展开了深入研究。主要研究内容概括如下:第二章研究LFM信号脉内调制参数估计问题,着重解决多分量LFM信号的脉内调制参数估计问题。(1)针对单分量LFM信号脉内参数估计二次相位函数(QPF)方法的性能分析问题,从信噪比积累增益角度对其抗噪性能进行了分析,给出了输出信噪比的表达式,得到了保证算法获得合理估计精度的约束条件,并给出了一阶扰动分析法所要求的“高信噪比”条件。此外还给出了实际应用中信号处理长度和调频率搜索间隔的选取方法。(2)针对多分量LFM信号的QPF存在的信号间互扰和伪峰问题,提出了积分二次相位函数(IQPF)方法,该算法能有效抑制多分量信号间互扰和伪峰,给出了IQPF的SNR积累增益和参数估计精度的解析表达式,并通过仿真实验验证了理论分析结果。第三章研究多相编码信号检测和调制参数估计问题,重点解决编码参数盲估计问题,为后续章节进行多相编码信号调制识别提供必要参数。(1)对多相编码信号QPF的特点进行了分析,提出了基于IQPF的检测方法,并通过接收机工作特性(ROC)曲线和特定虚警概率下的检测概率指标对其性能进行了分析。(2)推导了多相编码脉冲模糊函数(AF)的表达式,证明了AF幅度图像上距主脊线最近的脊线在时延和多普勒频率轴上的截距包含信号编码时长和子码率信息的结论。针对多相编码脉冲信号子码率和编码时长盲估计问题,提出了一种基于AF的估计算法,并对其性能进行了理论分析。(3)针对多相编码连续波信号,提出了基于AF局部特性的子码率和调制周期估计算法,算法计算效率高,易于工程实现。利用一阶扰动分析方法推导了子码率估计的理论均方误差的解析表达式。第四章研究多相编码脉冲信号调制识别问题,着重解决雷达脉内调制识别研究中相同调制类型内不同调制序列识别难题。(1)针对现有算法仅以编码长度和子码率构造参考信号可能导致编码类型识别错误的问题,提出了一种新的“三参数”参考信号构造方法,并考虑了编码序列正序和反序识别问题。通过数值仿真分析了载频估计误差对相关分类器性能的影响。(2)针对已有调制识别算法对载频估计误差敏感的缺点,在分析多相编码信号间互相关谱特点的基础上,提出了基于互相关谱的调制方式识别算法,该算法无需载频估计并且可采用高效的快速Fourier变换(FFT)实现,运算复杂度低。第五章研究单站无源定位中脉冲压缩信号多普勒频率变化率高精度估计问题,重点解决利用相参脉冲串精确估计脉压信号多普勒频率变化率的问题。(1)分析了目标辐射源与观测平台相对运动对观测平台接收信号的影响,建立了脉冲压缩信号观测信号模型。(2)针对LFM相参脉冲串信号,推导了多普勒频率变化率估计的克拉美-罗下界(CRLB),并将分数阶Fourier变换(FrFT)应用于LFM相参脉冲串信号多普勒频率变化率估计问题,提出了一种新的高精度估计算法。理论分析表明该算法估计精度逼近CRLB。(3)针对二相调制(BPSK)相参脉冲串雷达信号,提出了对接收信号进行±π相位补偿的高精度多普勒频率变化率估计算法,该方法通过相位补偿去除脉冲内BPSK调制,克服了进行平方或脉冲间相关运算会引入交叉项的问题;推导了不模糊多普勒频率变化率估计与脉冲重复频率、采样频率及观测时长间的约束关系。更多还原

【Abstract】 In modern warfare, the conflict between the electronic counter measures and anti electronic counter measures is becoming fiercer day by day. Gathering intelligence about hostile radars and locating their positions are a prerequisite to the precision strike and is also the master key to the initiative of war. One of the core issues is to analyze and process the low probability of intercept (LPI) radar pulse compression signals, etc. the large time-bandwidth product signals LFM and polyphase coded signals. We deeply study some key technologies/subjects in this dissertation, including the modulation rec-ognition, the modulation parameter estimation and the high-precision measurement of Doppler frequency rate for the LPI pulse compression signals. The main contributions of this dissertation are given as follows:In Chapter 2, the intra-pulse modulation parameter estimation problem is studied. We focus on solving the problem of intra-pulse modulation parameter estimation for multi-component linear frequency modulation (LFM) signals. (1) From the viewpoint of the output signal-to-noise-ratio gain, the anti-noise performance of the quadratic phase function (QPF) is analyzed and a closed-form expression for the output SNR is derived to evaluate the performance of the QPF method for single LFM signal. The constraint to guarantee a reasonably estimation accuracy is then obtained. Thereafter the high SNR condition required by the first order perturbation analysis is given. The selection of the signal processing length and the search step for the frequency rate is also given for prac-tical applications. (2) To deal with the problem of spurious peaks and interferences pre-sent in the QPF of multi-component LFM signals, a modified algorithm, named IQPF, is proposed. The method can suppress interferences and spurious peaks caused by the multi-component signals effectively. In what follows, the analytical expressions of the output SNR gain and the estimation accuracy are derived. The simulation results verify the theoretical results.In Chapter 3, the detection and intra-pulse modulation parameter estimation prob-lem are investigated. We are concentrating on solving the blind encoding parameter es-timation problem of the polyphase coded signals, which will be used by the following chapter for the purpose of the modulation recognition. (1) For the analysis of the char-acteristics of the polyphase coded signal’s QPF, an IQPF-based detector is proposed. The performance of the detector is evaluated by the receiver operating characteristic curve and the detection rate merits at a constant false-alarm rate. (2) The ambiguity function (AF) of the polyphase coded pulse signal is derived. A conclusion is drawn which states that, the intercepts, on the Doppler frequency and the time delay axes, of the ridge nearest to the main ridge contain information on the subcode rate and the time duration of the pulse signal. In addition, an AF based method is also proposed to obtain the blind estimation of the duration and the subcode rate. (3) We proposed a partial AF based algorithm for estimating the subcode rate and the modulation period of polyphase coded continuous wave signals. The algorithm is high efficiency and easy to be imple-mented. The theoretical mean square error of the subcode rate estimates is derived using the first order perturbation analysis.In Chapter 4, the modulation recognition of the polyphase coded pulse signal is studied. We focus on classifying the polyphase coded signals with the same modulation type but different code sequences. (1) To solve the error identification problem that may occur caused by using only the subcode rate and the code length to construct the refer-ence signals, a new reference signal generation method is proposed. The positive and reverse sequences are considered. The influence of the carrier frequency estimation er-ror on the performance of the correlation classifier is investigated by the numerical simulation. (2) The existing modulation recognition algorithms have a drawback that their performance is sensitive to the carrier frequency estimation error. Thus, a new cross-correlation spectrum based modulation recognition method, which doesn’t need estimating the carrier frequency, is proposed. The algorithm can be implemented by us-ing the fast Fourier transform (FFT). Thus it has low computational complexity.In Chapter 5, the high accuracy Doppler frequency rate estimation is studied for the single observer passive location application. We concentrate on the Doppler frequency rate estimation for the pulse compression signals by using the coherent pulse trains. (1) The influence of the relative motion between the observer platform and the object emit-ter on the characteristics of the observer-received signal is analyzed. The received signal model is described. (2) The Cramér-Rao lower bound (CRLB) of the Doppler frequency rate estimates for the LFM coherent pulse train signal is deduced. A new fractional Fou-rier transform (FrFT) based algorithm is proposed to estimate the Doppler frequency rate for LFM coherent pulse train signals. The theoretical analysis shows that the vari-ance of the estimates is approaching the CRLB at high SNRs. (3) A new Doppler fre-quency rate estimation algorithm is proposed which is based on the phase transition compensation. Different from the other algorithms that based on the square or the in-ter-pulse correlation operations, the proposed algorithm does not suffer the cross term problem. The constraints among the unambiguity Doppler frequency rate estimate, the pulse repetition frequency, the sampling frequency and the time duration of the observa-tion is given.更多还原