【作者】 张涛；

【导师】 刘鲁涛；

【作者基本信息】 哈尔滨工程大学 ， 信息与通信工程， 2024， 硕士

【摘要】 随着电磁环境的复杂化,基于传统阵列的信号处理系统越发不能满足电子侦察的需求。极化敏感阵列是一类矢量传感器阵列,相较于传统阵列,可以额外感知到电磁信源的极化特性,扩展可利用信息维度。利用该特性,可以有效的提高波达方向估计的精度,提高信号处理系统的性能,在高尖端电子战中取得信息优势地位。而复杂电磁环境下阵列接收信号模型扰动也相当复杂,阵列通道的不一致性带来的幅相误差、阵元非理想质点带来的位置误差、阵元间互耦响应、阵元极化敏感失配、环境噪声等影响因子都会大幅影响接收信号模型,进而使各类基于理想模型下的波达方向估计(Direction of arrival,DOA)算法性能下降甚至完全失效。为了减少误差影响,提高侦察系统性能,对极化敏感阵列进行校正具有重要的意义。本文针对复杂环境下的极化敏感阵列进行如下研究:首先,介绍了极化信源的分解和阵列接收模型,然后基于各类误差影响因子,分别给出了其相关的误差信号模型。研究了秩亏降维MUSIC算法,给出角度参数和极化参数的分离估计方法。针对极化敏感阵列特有的敏感失配误差,提出一种基于泰勒展开的敏感失配误差有源校正方法,构建了敏感失配误差模型,只需对任意一角度的两个不同极化信源进行采样,即可求取敏感失配角度。通过仿真实验表明,敏感误差校正后将有效提高信源估计的分辨率和估计均方误差。考虑复杂环境下的误差复杂铰链关系,构建方位依赖的误差信号模型,给出WF迭代自校正算法和基于最大似然估计的旋转迭代自校正方法,通过迭代逼近的方式同时完成参数估计和误差校正。针对实际应用中校正数据存储量庞大的问题,提出基于核的基函数曲线拟合数据压缩方法,通过将保留的部分存储数据拟合成函数曲线,大幅减少了数据存储量,有效减轻了信号处理系统负担。针对协方差矩阵估计受到有限采样和背景噪声的影响问题,提出基于正交极化阵列的矩阵重构DOA估计方法。构建均匀正交极化阵列,利用结构特点构建特殊的重构矩阵,并针对正交极化阵列完成子阵划分和重构分组,大幅降低了重构运算量。该算法将重构理论扩展到了极化敏感阵列领域,提高了阵列在恶劣环境下的参数估计能力。更多还原

【Abstract】 With the complexity of the electromagnetic environment,signal processing systems based on traditional arrays are increasingly unable to meet the needs of electronic reconnaissance.Polarization sensitive arrays are a type of vector sensor array that,compared to traditional arrays,can additionally perceive the polarization characteristics of electromagnetic sources and expand the available information dimensions.By utilizing this feature,the accuracy of direction of arrival estimation can be effectively improved,the performance of signal processing systems can be improved,and information advantages can be obtained in advanced electronic warfare.In complex electromagnetic environments,the disturbance of the array received signal model is also quite complex.The amplitude and phase errors caused by the inconsistency of array channels,the position errors caused by non ideal particles of array elements,the mutual coupling response between array elements,the polarization sensitivity mismatch of array elements,environmental noise and other influencing factors will greatly affect the received signal model,thereby causing the performance of various Direction of Arrival(DOA)estimation algorithms based on ideal models to decline or even completely fail.It is of great significance to calibrate polarization sensitive arrays in order to reduce the impact of errors and improve the performance of reconnaissance systems.This article conducts the following research on polarization sensitive arrays in complex environments:Firstly,the decomposition of polarization sources and the array reception model were introduced.Then,based on various error influencing factors,the relevant error signal models were provided.Studied the rank deficient dimensionality reduction MUSIC algorithm and provided a separate estimation method for angle and polarization parameters.Aiming at the unique sensitivity mismatch error of polarization sensitive array,an active correction method for sensitivity mismatch error based on Taylor expansion is proposed,and the sensitivity mismatch error model is constructed.Only two different polarization sources at any angle are sampled,the sensitivity mismatch supporting role can be obtained.Simulation experiments have shown that sensitive error correction can effectively improve the resolution and mean square error of source estimation.Considering the complex hinge relationship of errors in complex environments,a directional dependent error signal model is constructed.The WF iterative self correction algorithm and the rotation iterative self correction method based on maximum likelihood estimation are proposed,and both parameter estimation and error correction are completed simultaneously through iterative approximation.Aiming at the problem of large storage capacity of calibration data in practical applications,a kernel based basis function curve fitting data compression method is proposed.By fitting the retained stored data into a function curve,the data storage capacity is greatly reduced,effectively reducing the burden on signal processing systems.A matrix reconstruction DOA estimation method based on orthogonal polarization array is proposed to address the impact of limited sampling and background noise on covariance matrix estimation.Construct a uniform orthogonal polarization array,utilize the structural characteristics to construct a special reconstruction matrix,and complete sub array division and reconstruction grouping for the orthogonal polarization array,significantly reducing the amount of reconstruction operations.This algorithm extends the reconstruction theory to the field of polarization sensitive arrays,improving the parameter estimation ability of arrays in harsh environments.更多还原