Fast ICA 盲分离算法在雷达抗主瓣干扰中的应用研究

压制干扰信号从主瓣进入雷达天线,会严重影响雷达的性能,通常的副瓣抗干扰技术难以奏效。文中首先给出了Fast ICA 应用于雷达抗主瓣干扰的信号模型;在高信噪比的均匀噪声环境中,利用基于寻找峭度的局部极值点的Fast ICA盲分离算法分离接收到的主瓣干扰混合信号,通过脉压找出目标信号。仿真验证了算法用于抗主瓣干扰的有效性,该算法具有良好的抗干扰性能,在分离效率上具有较明显的优势。     

一种对MIMO雷达侦察识别的新方法

多输入多输出（MIMO）雷达作为一种新体制雷达,能够利用分集技术来改善雷达的性能。介绍了MIMO雷达的特点和常规的雷达信号分选方法,分析了它们对MIMO雷达信号进行分选的适用性,提出了一种采用奇异值分解和盲信号处理的方法对MIMO雷达进行侦察识别,通过仿真结果验证了该方法的有效性。     

一种对MIMO雷达的侦察方法

多输入多输出（MIMO）雷达作为一种新体制雷达,在低截获概率和抗干扰等方面具有优异的性能。介绍了MIMO雷达的分类和共址MIMO雷达的特点,讨论了对共址MIMO雷达进行侦察的方法,对其信号的特性做了仿真分析。     

盲分离算法在MIMO雷达和通信中的应用研究

将盲分离算法应用于多输入多输出(MIMO)雷达抗干扰和MIMO通信符号检测中。首先,利用信号相互之间以及与干扰之间的独立性,通过盲源分离算法,将各个信号分离出来;然后,雷达中通过匹配处理,完成信号检测;通信中利用少量的训练序列完成信号的匹配以及相位和幅度的校正。仿真结果表明:无论在雷达或通信中,均可获得优良的性能。     

基于ICA的电磁态势监测信号分离研究

简要地对ICA及其快速定点算法（Fast／CA）进行介绍的同时,着重地探讨将该算法运用到电磁监测领域分离空间复杂电磁信号的可行性。随后,文中通过仿真比较了FastlCA算法和Informax算法在这一领域运用中分离信号的性能,实验结果显示FastlCA算法性能更佳。     

Fast ICA应用于雷达抗主瓣干扰算法研究

压制干扰信号从主瓣进入雷达天线,会严重影响雷达的性能。通常的副瓣抗干扰技术难以奏效。本文首先给出了盲源分离应用于雷达主瓣抗干扰的信号模型,在此基础上提出了均匀噪声环境中,基于负熵最大化的快速固定点独立成分分析(Fast ICA)盲源分离算法,并用其分离接收到的干扰混合信号,最后脉压找出目标信号。仿真验证了算法用于主瓣抗干扰的有效性,并对其抗干扰性能进行了评价。仿真结果表明了算法良好的抗干扰性能,以及在分离效率上较明显的优势      

基于数据统计算法和信号分离技术的Massive MIMO系统半盲解码方案

为了解决Massive MIMO系统中的导频污染问题,文中提出了一种半盲解码方案,该方案使用数据统计算法和复数Fast ICA算法降低接受信号矩阵的维度并提取独立成分。同时完成信道估计,然后利用信号分离技术MMSE将发送信号解码出来。     

基于相关性测度的雷达分选算法

雷达信号分选利用信号特征参数的相关性来实现不同雷达信号的分离。在实际雷达信号分选的处理过程中,由于信号参数的无规律性以及先验知识的缺乏,采用独立分量分析(In-dependent Component Analysis, ICA)的方法对雷达信号进行处理。针对 ICA 算法分离非平稳信号性能下降的问题,提出一种相关性测度变步长 ICA 算法,将其应用于雷达信号分选中。计算机仿真表明,这种算法不仅减少了信号的分选达到收敛的迭代次数,而且可以有效地提高信号分选的稳态性。     

一种自适应算法的语音信号盲分离

盲信号处理算法主要有批处理算法和自适应算法两类,本文导出了一种批处理和自适应相结合的快速独立分量分析（Fast Independent Component Analysis, Fast ICA）算法,将该算法应用于语音信号盲分离处理,通过综合实验,从分离前后的波形、频谱图和主要评价参数说明该算法具有良好的信号分离效果。与扩展联合对角化（The Joint Approximative Diagonalization ofEigenmatrix,JADE）算法和自然梯度（Natural Gradient,NG）算法比较, fast ICA算法具有更好的分离效果。      

对MIMO雷达信号测频效果比较

为了测试典型雷达对抗侦察接收机对多输入多输出(MIMO)雷达信号的测频效果,首先分析了MIMO雷达信号的特点,建立了信号模型,然后分析了4种典型接收机对MIMO雷达信号侦察的处理过程,最后对几种接收机的侦察效果进行了仿真对比。     

Esprit算法在单基地MIMO雷达中的应用

多输入多输出（MIMO）雷达是近年来提出的一种新体制雷达,由于发射正交信号,不同位置目标的回波信号相互独立,因此在DOA估计和目标分辨方面,比传统雷达具有较大优势。文中将Esprit算法应用于单基地MIMO雷达系统中,并对两种结构的MIMO雷达分别进行一维和二维DOA估计,仿真证明了算法的优越性。     

Angle Estimation Using Quaternion-ESPRIT in Bistatic MIMO-Radar

In this letter; we present a novel two-dimensional angle estimation for bistatic multiple-input multiple-output (MIMO) radar. We reconstruct the received signal of MIMO radar to model with quaternion theory, and then angle estimate using quaternion estimation of signal parameters via rotational invariance technique for MIMO radar is proposed. The proposed algorithm can obtain automatically paired two-dimensional angle estimation in MIMO-radar. The proposed algorithm has much better angle estimation performance than the Wang’s quaternion algorithm, which has a much heavier computational load than the proposed algorithm. Simulation results verify the usefulness of our algorithm.     

波形相关矩阵未知情况下单基地MIMO雷达中一种改进MUSIC的DOA估计算法

该文研究了波形相关矩阵未知情况下多输入多输出(MIMO)雷达中的角度估计问题,提出了一种单基地MIMO雷达中改进多重信号分类(MUSIC)的到达角(DOA )估计算法。该算法可以在波形相关矩阵未知的情况下工作且性能优于传统的传播算子(PM)和借助旋转不变技术估计信号参数(ESPRIT)算法以及基于接收信号重构的MUSIC算法。该文算法可以扩展到任意阵列结构的MIMO雷达中进行角度估计。该文还给出了单基地MIMO雷达中DOA估计的克拉美罗界(CRB)。仿真结果验证了该算法的有效性。     

基于协方差矩阵重构的阵元失效MIMO雷达DOA估计

在实际应用中由于恶劣环境或人为干扰等因素而导致多输入多输出(Multiple-Input Multiple-Output,MIMO)雷达部分阵元失效,使得其接收数据缺失及其协方差矩阵秩亏,从而导致子空间类算法的波达方向(Direction of Arrival,DOA)估计性能恶化甚至完全失效。针对上述问题,提出了一种接收阵元失效下基于协方差矩阵重构的MIMO雷达DOA估计方法。该方法根据MIMO雷达协方差矩阵中以接收阵元数划分的子方块矩阵具有Toeplitz特性,利用正常工作接收阵元的协方差矩阵元素来恢复相应的缺失元素,从而重构出完整的数据协方差矩阵,提高阵元失效MIMO雷达的DOA估计性能。仿真结果验证了所提方法的有效性。     

基于块正交匹配追踪预处理的米波多输入多输出雷达测高方法研究

米波雷达具有很好的反隐身性能。多输入多输出(MIMO)雷达的波形分集具有高自由度特点,使MIMO雷达在检测和参数估计等方面具有更多优势,故米波MIMO雷达受到广泛研究。而测高是米波MIMO雷达最重要的问题之一。针对米波MIMO雷达测高问题,最大似然和广义多重信号分类方法是米波MIMO阵列雷达测高方法行之有效的算法,但其计算量大,工程中难以接受。该文提出一种基于块正交匹配追踪(BOMP)预处理的方法来降低计算量。首先对MIMO阵列接收数据稀疏化处理,然后通过数学操作将其变形至适合于BOMP算法的信号模型,然后利用粗栅格得到角度粗估计。并以此为初始值中心,取MIMO雷达波束宽度作为搜索范围。仿真结果表明该算法能有效降低搜索类测高算法的计算量。      

基于协方差匹配SLO算法的MIMO雷达DOA估计

利用加权平滑l0范数(Smoothed l0, SL0)算法估计MIMO雷达目标DOA时,需要把协方差矩阵进行矢量化来获得相应的稀疏重构模型,并利用信号和噪声子空间的正交性来构造加权向量。然而当存在相干信源时,MIMO雷达协方差矩阵的秩将退化,这会使得稀疏重构模型的误差较大以及无法正确区分信号和噪声子空间,导致加权SL0算法的DOA估计性能恶化。针对上述问题提出了一种基于协方差匹配SL0算法的MIMO雷达DOA估计方法。该方法利用协方差匹配准则重构出一个满秩的协方差矩阵,恢复MIMO雷达协方差矩阵的Toeplitz特性,并利用协方差逆矩阵的高阶幂来近似噪声子空间从而计算加权向量。仿真分析表明,该方法能够在无需预知信源数目的情况下有效地完成对相干信号的DOA估计。     

Joint Angle and Array Gain-Phase Errors Estimation Using PM-like Algorithm for Bistatic MIMO Radar

This paper discusses the problem of the direction of departure (DOD) and the direction of arrival (DOA) estimation for multi-input multi-output (MIMO) radar with array gain-phase errors. In this paper, we propose a propagator method (PM)-like algorithm for joint angle and array gain-phase errors estimation in MIMO radar. The proposed method not only yields automatically paired estimates of the angles and gain-phase errors but also has much better gain-phase errors estimation performance than the estimation of signal parameters via rotational invariance techniques (ESPRIT)-like algorithm; this has higher computational cost than the proposed algorithm. Furthermore, the proposed algorithm has angle estimation performance very close to ESPRIT-like algorithm. We also derive the Cramér–Rao bound (CRB) for MIMO radar with array gain-phase errors. Simulation results present the usefulness of our approach.     

Iterative Adaptive Approaches to MIMO Radar Imaging

Multiple-input multiple-output (MIMO) radar can achieve superior performance through waveform diversity over conventional phased-array radar systems. When a MIMO radar transmits orthogonal waveforms, the reflected signals from scatterers are linearly independent of each other. Therefore, adaptive receive filters, such as Capon and amplitude and phase estimation (APES) filters, can be directly employed in MIMO radar applications. High levels of noise and strong clutter, however, significantly worsen detection performance of the data-dependent beamformers due to a shortage of snapshots. The iterative adaptive approach (IAA), a nonparametric and user parameter-free weighted least-squares algorithm, was recently shown to offer improved resolution and interference rejection performance in several passive and active sensing applications. In this paper, we show how IAA can be extended to MIMO radar imaging, in both the negligible and nonnegligible intrapulse Doppler cases, and we also establish some theoretical convergence properties of IAA. In addition, we propose a regularized IAA algorithm, referred to as IAA-R, which can perform better than IAA by accounting for unrepresented additive noise terms in the signal model. Numerical examples are presented to demonstrate the superior performance of MIMO radar over single-input multiple-output (SIMO) radar, and further highlight the improved performance achieved with the proposed IAA-R method for target imaging.      

基于Group lasso的分布式MIMO雷达参数估计与能量优化

针对压缩感知算法在分布式MIMO雷达参数估计性能上易受噪声影响而出现伪峰、定位不准等问题,结合目标散射系数所满足块稀疏的前提条件,提出了一种基于Group lasso模型框架下压缩感知算法的参数估计。Group lasso作为一种块稀疏模型,可以有效解决感知算法在低SNR时参数估计性能差的问题,有效抑制了噪声对稀疏信号的破坏,其性能明显优于感知算法中常用的凸松弛CVX方法。此外针对MIMO雷达目标散射系数准确估计的前提,提出了基于线性规划的发射能量优化分配策略,此时系统模型转化为单发多收的MISO雷达,且集中发射功率于散射系数较大的路径。仿真结果表明,优化发射能量后的MISO雷达性能优势明显,尤其是当路径起伏较大时,性能尤为突出。仿真实验验证了理论分析的正确性和算法模型的有效性。      

Joint parameter estimation of DOD/DOA/polarization for bistatic MIMO radar

In the paper,polarization-sensitive array is exploited at the receiver of multiple input multiple output (MIMO) radar system,a novel method is proposed for joint estimation of direction of departure (DOD),direction of arrival (DOA) and polarization parameters for bistatic MIMO radars. A signal model of polarimetric MIMO radar is developed,and the multi-parameter estimation algorithm for target localization is described by exploiting polarization array processing and the invariance property in both transmitter array and receiver array. By making use of polarization diversity techniques,the proposed method has advantages over traditional localization algorithms for bistatic MIMO radar. Simulations show that the performance of DOD and DOA estimation is greatly enhanced when different states of polarization of echoes is fully utilized. Especially,when two targets are closely spaced and cannot be well separated in spatial domain,the estimation resolution of traditional algorithms will be greatly degraded. While the proposed algorithm can work well and achieve high-resolution identification and accurate localization of multiple targets.