【作者】 蔡云霄；

【导师】 解小华；

【作者基本信息】 吉林大学 ， 控制理论与控制工程， 2010， 硕士

【摘要】 阵列信号处理是空域信号分析和处理的一种重要手段,它广泛应用于雷达、声纳、通信、地震勘探和医学成像等多个领域。阵列信号处理的本质是利用空间按一定规则布置的传感器阵列和多通道来接收获取信号的时域和空域等多维信息,达到检测信号和提取其参数的目的。对于空间电磁波信号而言,波达方向和极化状态是其重要的特征参数,携带了空间电磁波信号的重要信息。信号波达方向描述了信号源在空间中的位置,极化状态描述了电磁波的矢量运动特征,是电磁波本身固有属性。传统的标量传感器阵列能够估计信号的功率、频率、时间延迟和到达方向,但不能估计出波信号的极化状态。电磁矢量传感器阵列不仅能获得信号的空间到达角信息,而且也能得到信号的极化信息。和传统标量传感器阵列相比较,电磁矢量传感器阵列具有以下几个优点：具有较强的抗干扰能力；具有稳健的监测能力；具有较高的系统分辨能力；具有极化多址能力。在实际工程中,经常会碰到相干源的问题。在雷达或声纳系统中,当目标信号存在多径或有多个目标信号时,接收到的回波是相干的。关于相干源的波达方向估计问题一直是雷达、通信和声纳等系统中的关键问题之一。当来波信号相干时,经典的MUSIC算法和ESPRIT算法中关键的协方差矩阵不满秩,从而无法划分信号子空间和噪声子空间,导致算法失效。在此,本文利用电磁矢量传感器阵列的优秀性能,讨论研究了相干信号源DOA和极化参数的估计问题。针对循环平稳信号中的相干源问题,本文首先利用极化均匀线阵,结合空域平滑技术,介绍了循环MUSIC平滑算法。其后,研究了在Toeplitz矩阵重构基础上的相干源DOA估计方法。对相干循环平稳信号源的DOA给予了有效估计,而且在不影响阵列孔径的情况下,进一步提高了估计性能。最后,分析了非均匀线阵的信号模型,讨论了非均匀线阵的阵元配置、阵列信号的估计模糊的问题,并结合电磁矢量传感器平滑的循环MUSIC算法,研究了在稀疏线性阵列下的循环相干信号的DOA和极化参数估计问题,并进行仿真实验予以验证。更多还原

【Abstract】 Array signal processing is important means of the signal analysis.It is widely applied in radar, sonar, communication, seismic exploration and medical imaging etc. Array signal processing is using plenty of sensors in different locations in the space composition to form a sensor array, through which signals of the space are received and processed to get the temporal and spatial multi-dimensional signal. That’s how detection and extraction of its parameters is completed. The polarization state and DOA are the most important characteristic parameters for space of electromagnetic signals and they carry important information of electromagnetic signals. Signal DOA describes the position of signal source in space, while the polarization describes the characteristics of electromagnetic wave motion vector, which is the intrinsic property of electromagnetic wave. The traditional scalar sensor array can estimate the signal power, frequency, time delay and to estimate direction, but not the polarization. Electromagnetic vector sensor array can not only obtain the angle of signal in the space, but also can get information of signal polarization. Compared with traditional scalar sensor array, electromagnetic vector sensor array, has several advantages:strong anti-jamming ability, stable capability. The system has high resolution and polarization multi-access ability.In practical engineer project, we often encounter with coherent sources. In radar or sonar system, when there is target with multipath or multiple target signals, receipt of the echo is key issues of coherence estimation problem. When the wave signals are coherent, the covariance of the classical MUSIC algorithm and the ESPRIT algorithm cannot divide signal subspace and noise sub-space, which causes algorithm failure. So this paper studies the DOA of coherent sources and the polarization parameters estimation problem.Firstly, based on the polarized uniform linear array, we combined with spatial smoothing technique, described the cycle MUSIC smoothing algorithm for the coherent cyclostationary signals. Subsequently, the basis of Toeplitz matrix regrouping DOA coherent sources estimate method is studied. The smooth circulation of coherent sources estimate DOA gives effective estimates, and in the circumstances, the pore arrays to further improve the performance.Furthermore, the signal sparse linear array model is also analyzed, the inhomogeneous linear array elements configuration and the estimation of the fuzzy array signals are discussed. At last, based on the cycle of electromagnetic vector sensor smoothing algorithm, DOA and polarized coherent signal parameter estimation problem is discussed under the sparse linear array with simulation experiment verified.更多还原