【作者】 王赞；

【导师】 陈伯孝；

【作者基本信息】 西安电子科技大学 ， 信号与信息处理， 2014， 博士

【摘要】 高频地波雷达具有超视距探测能力，不仅能够覆盖常规微波雷达和天波雷达的探测盲区，而且在反隐身、反低空突防、抗反辐射导弹和远程预警等方面具有突出的优势。但由于发射信号的波长较长，现役高频地波雷达普遍存在“阵地大、选址难、角度分辨率低、机动性差”等问题，干扰和杂波也是影响高频地波雷达目标检测的主要因素，特别是短波通信和广播电台等射频干扰以及电离层杂波等问题最为突出，亟待进一步解决。因此，结合“岸舰双基地波超视距雷达”课题，论文围绕分布式双基地波雷达展开相关研究，包括岸基收发阵列分布式阵列结构设计、分布式阵列下的DOA估计方法研究、岸基收发阵列校准、射频干扰抑制方法研究、双基地体制下电离层杂波建模以及杂波抑制方法研究，具体工作概括如下：1．分布式双基地波雷达方案设计及其DOA估计方法研究。首先针对现役高频地波雷达存在的“阵地大、选址难、角度分辨率低、机动性差”等四大问题，提出分布式双基地波雷达的设计思想：分布式双基地波雷达保留岸舰双基地波超视距雷达的综合脉冲孔径技术以及海上接收平台机动灵活的优点，同时引入分布式阵列技术，改进岸基发射阵列结构为分布式收发阵列，使分布式双基地波雷达具有单基地MIMO和双基地MISO两种工作模式；其次给出了分布式双基地波雷达的发射波形，并通过对信号处理流程的分析，得到两种工作模式下的等效接收阵列模型；然后对分布式双基地波雷达DOA估计进行着重研究，通过对信号处理流程和等效阵列模型的分析，指出雷达在单基地工作模式下用CBF、Capon和MUSIC等常规方法作DOA估计会导致雷达的接收数据处理量过大，在双基地工作模式下，因其等效接收阵列具有分布式结构，阵列方向图存在栅瓣，运用上述常规的DOA估计方法会出现角度模糊问题。针对这两个问题，论文提出一种基于压缩感知的DOA估计方法，利用压缩感知通过少量观测数据即可高概率获取目标准确信息的优势，在保证DOA估计精度的同时，降低单基地工作模式下接收数据的处理量和DOA估计的运算量，避免了双基地工作模式下等效阵列方向图栅瓣引起的角度模糊问题。仿真实验分析说明了分布式双基地波雷达的优点及其在DOA估计方面应用常规方法时出现的问题，验证了论文提出的基于压缩感知的DOA估计方法的有效性及正确性。2．分布式双基地波雷达岸基收发阵列校准。对两种工作模式下的阵列误差进行分析，在此基础上探讨利用这两种工作模式对岸基收发阵列的发射阵元和接收阵元分别进行误差校准的方案：首先在双基地工作模式下对发射阵元进行校准，采用岸舰双基地波超视距雷达的校准方式，利用海上接收平台接收到的直达波信号，并通过数据协方差矩阵拟合（CMF）方法或子空间拟合（SF）方法校准发射阵元的幅相误差；然后在单基地工作模式下对接收阵元进行校准，由于该模式下的等效阵列的阵元数较多，导致幅相误差校准的运算量大，因此采用对等效阵列中部分阵元进行校准的策略，利用大型舰船目标回波、有方向性的强射频干扰和强电离层杂波，并通过基于MUSIC的阵列自校准方法校准接收阵元的幅相误差。论文还给出了阵列幅相误差校准的克拉美罗界。最后通过仿真实验验证了岸基收发阵列误差校准方案的可行性以及所用到的校准方法的有效性。3．高频地波雷达射频干扰抑制研究。论文以分布式双基地波雷达为研究对象，首先分析了射频干扰在该雷达下表现出的时域、距离域、多普勒域以及空域特性；其次介绍了射频干扰时域剔除法，并指出该方法检测阈值的设定会直接影响干扰抑制效果，且方法只能用于射频干扰较少的白天，在干扰密集的夜间该方法失效，针对上述缺陷，论文提出一种基于压缩感知的射频干扰抑制方法，同样利用射频干扰的时域特性，方法通过对每个调频周期选取少量“干净”的时域采样数据，利用压缩感知重构算法分别获得目标准确的距离信息和速度信息。此外，论文还探讨运用盲信号处理的方法来分离射频干扰的构想，根据单载频射频干扰在各调频周期间具有很强的相关性且与目标回波不相关的特点，在单接收通道对多个调频周期采用独立分量分析算法将射频干扰与目标回波分离开。最后通过仿真实验分别验证了基于压缩感知和基于独立分量分析的射频干扰抑制方法的有效性。4．双基地体制下的电离层杂波研究。首先介绍了高频地波雷达电离层杂波的成因，并针对双基地体制下电离层杂波的空域特性和多普勒特性进行分析，给出了杂波在双基地体制下方位、俯仰和锥角之间的关系，分析了海上移动接收平台接收的回波信号中电离层杂波多普勒频率的组成，综述了国内外已经研究的电离层杂波特性；其次，论文根据电离层杂波特性，采用随机信号建模法分别建立了电离层镜像折射杂波和电离层散射杂波模型；然后利用特征值分解方法，提出“时域信号”抑制方法，对电离层镜像折射杂波这类强杂波进行抑制。论文给出了电离层杂波仿真结果以及“时域信号”抑制方法对电离层镜像折射杂波的抑制结果，仿真结果表明了“时域信号”抑制方法的有效性。更多还原

【Abstract】 Much attention has been paid to HF surface wave radar for its ability to detect theover the horizon targets. The low altitude and near ranges are blind spots ofconventional microwave and Sky Wave Over the Horizon Radar respectively, HFSWRcan compensate these spots completely. The long distance and early radar surveillance,anti stealth, anti altitude penetration, and anti radiation missiles were the decidedadvantages of HFSWR. However, long wavelength of transmit leads to the problems oflarge position, difficulty of location chosen, low of the angle resolution, and badflexibility. The detection capability of HF surface wave radar is also influenced byinterferences and clutters especially the radio interference and the ionospheric clutter.Based on the discussion of coast ship bistatic surface wave over the horizon radar,this dissertation focuses on the design of distributed subarrays, the DOA estimation fordistributed subarrays, calibration of arrays which are on the coast, the methods of radiointerference suppression, the ionospheric clutter modeling and the study of suppressionmethod for ionospheric clutter in bistatic HFSWR with distributed subarrays.Themain content of our studies are summarized as follows:1. Distributed bistatic surface wave radar is designed and the DOA estimation inthis new radar is studied. Firstly, the concept of distributed bistatic surface wave radar isproposed for the problem of large position, difficulty of location chosen, low of theangle resolution, and bad flexibility. The new radar reserves the Synthetic Impulse andAperture Radar mechanism (SIAR) and the moving receive plat over ocean, the transmitarray using the distributed subarrays mechanism which makes the new radar havemonostatic MIMO system and bistatic MISO system. Secondly, the waveform oftransmit is proposed. Based on the analysis of signal processing, we can obtain theequivalent receive array models in two systems. Lastly, this dissertation focuses on theDOA estimation methods in the new radar. After the analysis of the DOA estimationproblem in two systems, the new DOA estimation method based on compressed sensingis proposed. The new method can estimate the target’s information with a few of sampledata, reduce the processing complication, and avoid the angle ambiguity. Simulationshows the advantages of distributed subarrays bistatic surface wave radar, the problemsusing the conventional DOA estimation methods and the validity of the new DOAestimation method based on compressed sensing.2. Transmit receive array of distributed bistatic surface wave radar is calibrated. Based on the analysis of array error in two systems, a strategy of transmit array andreceive array error calibration in these systems is proposed. Firstly, transmit array erroris calibrated in monostatic MIMO system, the error calibration technique is same ascoast ship bistatic surface wave over the horizon radar, by using the direct wave, themoving receive plat can calibrate the transmit array error with Covariance MatrixFitting(CMF) method and Subspace Fitting(SF) method respectively; Receive arrayerror is calibrated in bistatic MISO system, by using the echo from large ship, the strongradio interference and the ionospheric clutter, the receive array on the coast can achievearray error calibration with auto calibration method based on Multiple SingalClassification (MUSIC) method. Because of the large number of equivalent receiveantennas, this dissertation adopts the strategy that only a part of the equivalent receiveantennas are calibrated. This dissertation also presents the Cramer Rao Bound(CRB) ofarray error calibration. Simulation shows validity of the transmit array and receive arrayerror calibration strategy.3. Radio interference suppression of high frequency radar is studied. Thisdissertation focuses on distributed subarrays bistatic surface wave radar. After theanalysis of radio interference’s characteristics in time domain, range domain, dopplerdomain and space domain, we introduce the excising method in time domain, indicatethe several disadvantages of this method. Then, the radio interference suppressionmethod based on compressed sensing is proposed. Also, by using the interference’scharacteristics in time domain, this new method can obtain target’s range informationand doppler information accurately with a few of "cleanly" data. This dissertation alsoproposes a new radio interference suppression concept by using Blind Signal Processing(BSP).Because of the independence between single frequency radio interference andtarget echo in a periodicity, we can separate the single frequency radio interference fromthe target echo. Simulation shows validity of the radio interference suppression methodbased on compressed sensing and the radio interference separation method.4. Ionospheric clutter in bistatic system and suppression method is studied. Firstly,we analyze the cause for the form of ionospheric clutter in high frequency radar, thespace domain and doppler domain characteristics in bistatic system, and show therelation with azimuth, elevation and cone. We also analyze the components of dopplerincluded in the echo which are received by moving receive flat, and sum up theionospheric clutter transmission characteristics. Secondly, we achieve the ionosphericmirror refraction clutter model and ionospheric diffuse reflection clutter modelrespectively by using random signal modeling method. Lastly, we propose "time domain signal" clutter suppression method based on the eigenvalue decompose. This methodcan suppress ionospheric mirror refraction clutter availably. This dissertation shows thesimulation result of ionospheric clutter and validity of "time domain signal" cluttersuppression method.更多还原