短快拍阵列测向技术研究

发布时间：2018-06-22 14:39

本文选题：DOA估计 + 单快拍　；参考：《电子科技大学》2014年硕士论文

【摘要】：在现代军事或卫星通信中,由于跳频通信与突发短时通信技术的广泛应用,以及复杂多变的电磁环境影响,阵列接收的有效数据往往不足。对于基于统计理论的经典测向算法来说,上述情况导致可利用快拍数据不足以准确估计样本协方差矩阵,进而导致子空间相互泄漏,算法性能下降甚至失效。本文分析了短快拍对传统子空间类测向算法的性能影响,将短快拍阵列测向技术分为极限的单快拍和非极限的少快拍两种情况分别进行了分析和探讨,并在原有算法的基础上做了如下工作:提出了一种无需已知信源个数基于空域平滑理论的单快拍测向算法。该算法基于有效提高阵列孔径的空域平滑理论,利用子空间的旋转不变特性,在较少的先验条件下,可先对信源个数进行比较准确的估计,再应用样本协方差矩阵的幂来近似子空间,进而完成DOA估计。实验结果表明其测向性能与现有空域平滑理论下的SSIA ESPRIT算法相当。与其他现有单快拍测向算法如相关域单快拍重构伪协方差矩阵的Power ESPRIT相比测向精度更高,适用于对目标信号已知条件较少的非合作通信侦察中。提出一种在短快拍下基于伪噪声重采样技术并结合酉Root-MUSIC算法的快速DOA估计算法。该算法借助伪噪声重采样技术提高测向算法性能使其适用于短快拍。由于利用计算复杂度较低的QR分解来求解噪声子空间,故不需计算复杂的特征分解。同时利用酉Root-MUSIC算法来进行DOA估计,将复数运算转化为实数运算,降低了算法的计算复杂度。与只使用PR技术的酉Root-MUSIC算法相比,性能相当,但计算复杂度较低;且相比调整子空间类算法及离网稀疏贝叶斯算法等现有的短快拍测向算法具有更低的计算复杂度,适用于信号处理中的实时系统。
[Abstract]:In modern military or satellite communication, due to the wide application of frequency-hopping communication and burst short-time communication technology, as well as the complex and changeable electromagnetic environment, the effective data received by array are often insufficient. For the classical direction-finding algorithm based on statistical theory, the situation mentioned above leads to insufficient estimation of the sample covariance matrix by using the rapid-beat data, which leads to the leakage of subspace, and the performance of the algorithm even fails. In this paper, the effect of short-shot on the performance of traditional subspace-like direction-finding algorithm is analyzed, and the short-shot array direction-finding technology is divided into two cases, which are limited single-shot and non-limit low-shot respectively. On the basis of the original algorithm, the following work is done: a single beat direction finding algorithm based on spatial smoothing theory without known number of sources is proposed. This algorithm is based on the spatial smoothing theory which can effectively improve the aperture of the array. By using the rotation invariance of the subspace, the number of sources can be estimated accurately under the condition of less prior information. Then the power of the sample covariance matrix is used to approximate the subspace and the DOA estimation is completed. The experimental results show that the direction finding performance is comparable to the SSIA Esprit algorithm based on the existing spatial smoothing theory. Compared with other existing single shot direction-finding algorithms such as Power Esprit with pseudo-covariance matrix reconstruction in correlation domain, it is suitable for non-cooperative communication reconnaissance with less known conditions for target signals. A fast DOA estimation algorithm based on pseudo-noise resampling technique and unitary Root-music algorithm is proposed in this paper. The algorithm improves the performance of the direction finding algorithm by using pseudo-noise resampling technique and makes it suitable for short shot. Because the QR decomposition with low computational complexity is used to solve the noise subspace, there is no need to compute the complex characteristic decomposition. At the same time, the unitary Root-music algorithm is used to estimate the DOA, and the complex operation is transformed into real operation, which reduces the computational complexity of the algorithm. Compared with the unitary Root-MUSIC algorithm which only uses PR technique, the performance of the algorithm is comparable, but the computational complexity is lower, and the computational complexity is lower than the existing short-shot direction-finding algorithm, such as adjusting subspace class algorithm and off-network sparse Bayesian algorithm. It is suitable for real-time system in signal processing.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN911.7

【相似文献】