【作者】 朱晓；

【导师】 司锡才；

【作者基本信息】 哈尔滨工程大学 ， 通信与信息系统， 2008， 博士

【摘要】 脉冲压缩雷达发射大时宽-带宽积的雷达信号,在接收时采用脉冲压缩技术,将长脉冲压缩成窄脉冲,提高了雷达的距离分辨率的同时,降低了雷达信号的被截获概率。脉冲压缩雷达是一种低截获概率雷达,对现代反辐射导弹(ARM)导引头和其他被动侦察接收机提出了新的挑战。传统的晶体视频检波接收机不再适应截获脉冲压缩雷达信号的要求,急需发展宽带数字接收机。宽带数字接收机直接将中频信号或射频信号数字化,保留了脉冲压缩雷达信号的频率、相位调制信息,便于进行信号检测、存储、分析。对付脉冲压缩雷达信号首先是检测接收,然后是识别。在信号的载频、带宽均未知且时变的条件下,如何同时对不同体制的雷达信号进行“盲接收”,是宽带数字接收机首先要解决的问题。数字信道化接收机可以处理同时到达信号,均匀信道化方法可以采用高效的实现结构,是一种较为成熟的宽带数字接收机体制,但是由于信道划分是严格均匀且固定的,不适应电子战的环境,需要改进。完整接收到信号后,还需要提取脉冲压缩雷达信号的调制参数。相位编码信号、线性调频信号、非线性调频信号是脉冲压缩雷达信号普遍采用的信号类型,其中相位编码信号又分为二相编码信号和多相编码信号,四相编码信号是最常用的多相编码信号。相位编码信号的参数识别包括估计载频、码元速率、编码规律等。线性调频和非线性调频信号可以建模为多项式相位信号,前者的是二阶多项式相位信号,后者是高阶多项式相位信号。通过相位建模方法,估计出调频信号的相位多项式的系数就可以完成识别。本课题围绕宽带数字接收机平台截获和识别脉冲压缩雷达信号,对动态数字信道化方法、宽带数字接收机的动态范围、相位编码雷达信号和调频雷达信号的参数估计算法进行了深入细致的研究。在宽带信道化接收技术方面,针对接收宽带信号中存在多个非均匀分布且动态变化的信道的情况,分析了现有信道化接收机的缺陷,指出由于信道划分是严格均匀且固定的,不适应电子战的环境。提出了动态数字信道化方法,结合滤波器组理论和信号重构理论,推导出了两种高效的实现结构,基于DFT滤波器组的高效结构和基于短时快速傅立叶变换信号分析与综合的高效结构。两种结构可以对时变、非均匀信道进行划分,具有并行,实时的特点,运算效率高,具有一定的自适应能力。两种结构可以方便的映射到现有的FPGA器件上,文中给出了关键电路模块的实现方法。在扩展宽带数字接收机动态范围方面,采用中频数字AGC技术。研究了微波前端热噪声、放大器链路噪声因数及非线性特性、超高速ADC有限量化精度等因素对宽带数字接收机动态范围的限制。在分析数字AGC控制算法的基础上,提出了两种中频数字AGC系统以扩展宽带数字接收机动态范围。结合多速率数字信号处理技术,提出了高效希尔伯特变换结构提取信号的数字包络,将超高速ADC数据流通过采样率转换技术进行降速,给出了适于在FPGA上实现的高效信号包络提取电路。在相位编码雷达信号的参数估计算法方面,综合运用循环相关理论和连续小波变换方法,给出了一种较好的估计流程。根据PSK信号的循环平稳特性,首先采用循环谱方法估计其载频和码元速率。由于噪声不具有循环平稳特性,因而该方法具有良好的抗噪性能。利用载频估计值将PSK信号搬移到基带,分析了Haar小波变换提取基带PSK信号相位跳变点的性能,指出当残留载频趋近于0时,小波模极大值总是出现在相位跳变点处,根据此特征可以有效的提取相位跳变点的定时信息。在此基础上,分析了残留载频对算法性能的影响以及存在残留载频时,小波尺度的选取策略。在编码序列识别方面,提出了基于定时信息的直接相位法。采用大数判决对相位函数进行校正,有效地消除噪声引入的跳周现象,提出采用长度大于1的编码游程估计残留载频引入的累积相位,从而消除累积相位,大大降低了残留载频对算法的影响。在调频雷达信号参数估计算法方面,运用了多项式相位建模法。研究了乘积型高阶模糊度函数(PHAF)算法存在的漏检问题和误差传递效应。推导了PHAF算法漏检问题产生的条件,在此基础上提出了PAHAF算法。分析了误差传递效应的实质,提出通过Chirp-Z变换的方法提高频率分辨率,从而提高多项式相位信号的高阶相位参数的估计精度,减小误差传递效应对低阶相位参数估计精度的影响。更多还原

【Abstract】 The pulse compression radar emits large time-bandwidth product signals in the transmitter, while in the receiver it compresses the return signals with matched filter. The pulse compression techniques can result in high range resolution, and also provide a large processing gain for the radar. Pulse compression radar is a kind of low probability of intercept (LPI) radar. The term LPI is that property of radar that, because of its low power, wide bandwidth, or other design attributes, makes it difficult to be detected by means of a passive intercept receiver. The LPI requires the increase in capability of modern intercept receivers to detect and locate a radar emitter. The wideband digital receiver concerning its advantage over traditional crystal video receiver is in ergent need to develop. It digitizes the intermediate frequency or radio frequency radar signal directly, preserves its frequency and phase modulation information, and makes it easy to detect, store and analyze the radar signal.The contermeasures to pulse compression signal are detection and interception, and then identification. Modern electromagnetic environment is dense, interleaving with complex time-varying waveforms. There are simple pulse modulation signals with relatively narrow bandwidth, also pulse compression signals with large bandwidth. Their carrier frequencies and bandwidth are unkown to the intercept receiver. Therefore the reception of various types of radar signals is the first problem to deal with in wideband receiver. Digital channelization receiver can process simultaneous arriving signals. Uniform channelization receiver is a kind of mature wideband digital receiver. Its implementation is efficient. Unfortunately, because its channels are unchangeable and restrictly uniform, it dose not meet the EW environment and needs some improvements.Once the signals are intercepted, modern signal processing algorithms can be used to analyze the intrapulse information and extract the modulation parameters. There are three kinds of widely used pulse compression signals, phase shift keying (PSK) signals, linear frequency modulation (LFM) signals, and nonlinear frequency modulation (NLFM) signals. PSK signals include BPSK signals and MPSK signals, and QPSK signals are frequently used MPSK signals. Parameters of PSK to be estimated include carrier frequency, code rate and coding sequence. LFM and NLFM signals can be modeled as polynomial phase signals (PPS). LFM signals are second order PPS signals, while NLFM signals are high order PPS signals. Therefore FM signals can be identified by estimating each order phase coefficient.In order to intercept and recognize pulse compression radar signals on wideband digital receiver (WDR) platform, several subjects were studied in detail, including dynamic digital channelization techniques, dynamic range extension of WDR, modulation parameter estimation algorithms for PSK and FM radar signals.In terms of wideband channelization techniques, the drawback of recent proposed channelization receiver when receiving multi-channels which are non-uniformly distributed and dynamically changing in wideband received signal was pointed out. Dynamic channelization method was proposed. Two efficient implementation structures based on DFT filter banks and short time discrete Fourier transform respectively were presented. Both structures can be used to channelize dynamic changing and non-uniformly distributed channels adaptively. All channels can be processed in parallel and real-time. Both structures can be implemented in field-programmable gate array (FPGA) chips. Implementation schematics of key modules were also presented.In terms of extending dynamic range of WDR, digital automatic gain control (DAGC) was used. The limitation factors of WDR dynamic range such as microwave front-end thermal noise, amplifier-link noise factors and nonlinear characteristics, super high speed ADC limited quantification accuracy were thoroughly researched. Two intermediate frequency DAGC systems were presented based on analysis of DAGC control algorithms. In combination of multi-rate signal processing technology, an efficient Hilbert transform method for envelope extraction was proposed. This method converts the high speed ADC data flunt to low speed data flunt, which then can be processed by FPGA.In terms of PSK signal parameters estimation, a good estimation procedure with comprehensive application of cyclic correlation theory and continous wavelet transform was presented. Firstly cyclo-spectrum method was used to carry out blind estimation of PSK signal carrier frequency and coding rate for its good anti noise performance. Then the PSK signal was shifted to base band with frequency estimate. Haar wavelet transform method was used to extract phase jump points of PSK signal, and its performance was studied in detail. It was pointed out that wavelet transform maxima point always appears at phase jump point when carrier residual approaches zero. Therefor the phase jump points can be extracted via wavelet transform maxima. Influence of carrier residual on the algorithm was thoroughly studied. Selection of wavelet scale when carrier residual exists was given. An algorithm for PSK coding sequence recognition based directly on phase function was presented. Large number decision was introduced to eliminate phase-wrapping phenomenon caused by noise. The phase difference between neighboring codes was estimated according to phase function. The accumulative phase induced by residual carrier was estimated and subtracted from the phase difference. Then the phase jump of neighboring codes was estimated.In terms of FM radar signal parameter estimation, polynominal phase modeling method was used. Product high-order-ambiguity function to estimate phase coefficients of PPS signals suffers from detection missal and error propagation effect. Detection missal condition was deduced, and an improved method called PAHAF was presented. The error propagation effect was induced by limited frequency resolution of FFT, which can be improved by Chirp-Z transform.更多还原