改进的MUSIC算法在信号DOA估计中的研究

发布时间：2018-01-14 06:08

本文关键词：改进的MUSIC算法在信号DOA估计中的研究　出处：《长安大学》2014年硕士论文　论文类型：学位论文

【摘要】：阵列信号处理技术成为信号处理领域的一个重要的学科，它在通信、雷达、声纳、电子对抗等对多领域有着极为广阔的应用前景。基于阵列信号处理的波达方向（DOA）估计方法，可以同时对空间不同方向上的多个信号源实现高分辨率的方向估计。经过40过年的研究，，各国学者分析了各种类型的算法，信号波达方向的研究已经越来越完善。本文采用了理论和计算仿真实验相结合的的方法对算法和阵列结构的有关问题进行了研究。主要工作内容如下：本文从经典的MUSIC算法入手，在理想高斯白噪声条件下，分析了影响MUSIC算法性能的客观因素并用MATLAB进行仿真。由于MUSIC算法在相干信源进行估计时性能严重下降甚至是失效，因此，本文基于对MUSIC算法原理的改进，详细地介绍了空间平滑法、修正MUSIC算法和基于Toeplitz矩阵重构的去相干算法。空间平滑法，虽然解决了相干信号源的问题，但是对阵列孔径有一定的损失。修正的MUSIC算法，计算量小而且对阵列孔径没有损失，而基于Toeplitz矩阵重构的去相干算法，不仅有修正MUSIC算法的优点，还实现了少阵元多信源的DOA估计，并且在低信噪比时性能表现较好。在传统DOA估计算法的基础上，对近几年DOA估计中出现的高阶累积量信号分析与处理技术进行了研究。基于高斯色噪声条件下的DOA估计做了详细的介绍，本文应用了基于四阶累积量的MUSIC算法，为了去除算法中的大量数据冗余，提出了改进的四阶MUSIC算法。仿真结果表明，该在未知高斯噪声下，可以准确的估计出信号的入射角度，且阶数越高，估计性能越好。
[Abstract]:Array signal processing technology has become an important subject in the field of signal processing, it is in communication, radar, sonar. Electronic countermeasures have a wide application prospect in many fields. The DOA estimation method based on array signal processing is presented. High resolution direction estimation can be achieved for multiple signal sources in different directions at the same time. After 40 years of research, scholars from all over the world have analyzed various kinds of algorithms. The research on the DOA of the signal has become more and more perfect. In this paper, the algorithm and the array structure are studied by combining the theory with the simulation experiment. The main work is as follows: This paper starts with the classical MUSIC algorithm, under the condition of ideal Gao Si white noise. The objective factors that affect the performance of MUSIC algorithm are analyzed and simulated with MATLAB. Because the performance of MUSIC algorithm in the estimation of coherent sources is seriously degraded or even failed. Based on the improvement of MUSIC algorithm, spatial smoothing, modified MUSIC and Toeplitz matrix reconstruction are introduced in detail in this paper. Although the problem of coherent signal source is solved, there is a certain loss to array aperture. The modified MUSIC algorithm has less computation and no loss to array aperture. The decoherence algorithm based on Toeplitz matrix reconstruction not only has the advantage of modified MUSIC algorithm, but also realizes the DOA estimation of fewer array elements and more sources. And the performance is better when the signal-to-noise ratio is low. Based on the traditional DOA estimation algorithm. The signal analysis and processing technology of high-order cumulant in DOA estimation in recent years is studied. The DOA estimation based on Gao Si color noise condition is introduced in detail. In this paper, MUSIC algorithm based on fourth-order cumulant is applied. In order to remove a lot of data redundancy in the algorithm, an improved fourth-order MUSIC algorithm is proposed. The simulation results show that the algorithm is under unknown Gao Si noise. The incident angle of the signal can be estimated accurately, and the higher the order, the better the estimation performance.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN911.23

【参考文献】