基于阵列转换的二维MUSIC算法的二维DOA估计

发布时间：2019-02-25 16:12

【摘要】：空间谱估计作为信号处理技术研究的主要内容，其目的就是要获知信号发射源的准确位置，也就是信源定位。“空间谱”表示信号在空间各个方向上的能量分布，空间谱估计具有很强的空间分辨能力，其主要研究方向在于空间中的多个传感器阵列所构成的处理系统，就是利用一组传感器，这组传感器在空间按一定的规则分布，同时接收外界信源所发出的信号，从而提取出所需要的信号参数，其中最主要的估计参数是：空域参数、信源的俯仰角和方向角。在一个立体空间内建立一个球坐标系，只需要知道信号源相对于坐标系原点的方向角、俯仰角和距离这三个参数，就完全可以确定一个信源的准确位置。一维空间谱估计就是在一个平面内对信源的一个参数进行估计，但是在现实环境中，信源处在三维立体环境中，用一维参数不能估计信源的精确位置，因此需要用二维波达角来进行估计，所以相对于一维波达角估计二维波达角就更精确。二维DOA在军事和民用等领域起着重要的作用，在对一维DOA估计算法和阵列研究的基础上，对远场窄带二维DOA估计作了深入研究，分别提出了阵列变换和减小计算量的方法。对平行六边形阵列进行阵列变换，平行六边形阵列的子阵列不满足旋转不变性，变换为满足旋转不变性这一条件的阵列。然后再用平滑算法进行空间谱估计，变换后的阵列比矩形阵列在相同的信噪比下具有很小的均方根误差。针对MUSIC算法计算量大的问题，提出了一种快速空间谱估计改进算法。该算法就是将角度域波达角转换到变换域，，从而有效地解决了计算量大的问题，改进算法比传统MUSIC算法的计算量大大减少，估计成功的概率也略高于传统MUSIC算法。
[Abstract]:As the main research content of signal processing technology, spatial spectrum estimation aims to obtain the exact location of the signal source, that is, the location of the signal source. The "spatial spectrum" represents the energy distribution of the signal in all directions in space. The spatial spectrum estimation has a strong spatial resolution ability. The main research direction of the spatial spectrum estimation is the processing system composed of multiple sensor arrays in space. It is the use of a set of sensors, which are distributed according to certain rules in space and receive signals from outside sources, thus extracting the required signal parameters, the most important of which are: spatial parameters, The pitch angle and direction angle of the source. To establish a spherical coordinate system in a solid space, we only need to know the direction angle, pitch angle and distance of the signal source relative to the origin of the coordinate system, and the exact position of a signal source can be completely determined. One-dimensional spatial spectral estimation is to estimate a parameter of a source in a plane, but in a real environment, the source is in a three-dimensional environment, and one-dimensional parameters cannot be used to estimate the exact location of the source. Therefore, it is necessary to estimate the two-dimensional wave angle, so it is more accurate to estimate the two-dimensional wave-arrival angle than the one-dimensional wave angle. Two-dimensional DOA plays an important role in military and civilian fields. Based on the research of one-dimensional DOA estimation algorithm and array, the far-field narrow-band two-dimensional DOA estimation is deeply studied, and the methods of array transformation and reduction of computation are proposed respectively. The array transformation of parallel hexagonal array is carried out. The sub-array of parallel hexagonal array does not satisfy the rotation invariance and is transformed into an array which satisfies the condition of rotation invariance. Then using smoothing algorithm to estimate the spatial spectrum, the transformed array has a small root mean square error (RMS) compared with the rectangular array at the same signal-to-noise ratio (SNR). In order to solve the problem of large computational complexity of MUSIC algorithm, an improved fast spatial spectrum estimation algorithm is proposed. The algorithm transforms the angle-domain radar angle into the transform domain, which effectively solves the problem of large computational complexity. Compared with the traditional MUSIC algorithm, the improved algorithm reduces the computational complexity greatly, and the probability of successful estimation is slightly higher than that of the traditional MUSIC algorithm.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN911.7

【参考文献】