LANZHOU UNIVERSITY OF TECHNOLOGY
毕业论文
题 目： 高分辨方位估计算法研究
学生姓名 学 号 专业班级 指导教师 学 院 答辩日期 2014年6月15日
高分辨估计算法研究
High Resolution Estimation Algorithm Research
xxx xxx
兰州理工大学毕业论文
摘 要
阵列信号处理是信号处理领域内的一个重要分支，在近些年来得到了迅速发展。波达方向（Direction of Arrival，DOA）估计是阵列信号处理的一个重要的研究领域，在雷达、通信、声纳、地震学等领域都有着广泛的应用前景。在DOA估计的发展过程中，人们对高分辨DOA估计算法一直有很大的研究兴趣，并在这一领域取得了很多重要的进展。本文主要研究经典的多重信号分类（Multiple signal Classification，MUSIC）算法。
最经典的超分辨率空间谱估计方法是Schmidt在1979年提出的MUSIC(Mukiple Signal Classification)算法，在模型准确的条件下，该算法能精确地估计空间上互不相关信号的波达方向。然而，对空间上相隔比较近的小信噪比信号，MUSIC算法却不能很好地分辨出它们的波达方向，解决的途经是充分利用目标信号的信息。基于信号循环平稳特性的循环互相关MUSIC算法就是将信号的循环平稳特性用于阵列信号处理中，这样就进一步提高信号的处理质量，具有更好的噪声抑制特性和分辨能力，但是它对循环相关信号的分辨力则受到限制。本文在循环互相关MUSIC算法的基础上，同时利用了信号的循环互相关矩阵和循环共轭互相关矩阵的信息划该算法进行改进。仿真结果表明，这种改进的算法有更好的抑制噪声能力，能够分辨循环相关的信号，具有更好的分辨能力。
本文首先回顾了空间谱估计技术的发展过程及现状，比较详细的介绍了空间谱估计基础和DOA估计模型，研究了DOA估计中的MUSIC算法，给出了MUSIC算法的原理和步骤，并通过一系列MATLAB仿真实验，得出了MUSIC算法的性能分析。最后做了全文总结，归纳了本文所做的工作和结论。
关键词: DOA估计; 阵列信号处理; MUSIC算法
兰州理工大学毕业论文
Abstract
Array signal processing is an important branch of the field of signal processing, in recent years it has been developing rapidly. Direction-of-arrival（DOA）estimation is one of the important research of array signal processing ,which has found wide applications in radar, communication , sonar , seismology and other fields . During the development process of DOA estimation, high-resolution DOA estimation techniques have long been of great research interest and many significant progresses have been made in this field. This paper mainly studies the classical Multiple-signal-classification（MUSIC ）algorithm.
The MUSIC(Multiple Signal Classification)algorithm，presented by Schmidt in 1979， is the most classical super-resolution algorithm．Based on an exact model of antenna array , and the algorithm can estimate the DOA of uncorrelated signals accurately．However，as far as adjacent signals with small SNR(Signal to Noise Ratio)is considered，MUSIC algorithm cannot estimate their DOA. The main approach to solve this problem is full utilization of signal sources information．The MUSIC algorithms based on the cyclostationarity property have a better performance in noise suppressing and signal selectivity by using the cyclosationarity property of signal into array signal processing. However as far as signal with cyclic correlation is considered，the signal resolution of this algorithm is limited．To solve this problem，an improved cyclic MUSIC algorithm was presented in this paper．Both cyclic correlation matrices and conjugate cyclic correlation matrices are considered in this improved algorithm. Emulation results have indicated the improved algorithm can estimate the DOAs of cyclic correlation signals and adjacent signals and has a better resolution and noise suppressing ability．
This paper first reviewed the development process and the present situation of the spatial spectrum estimation; A more detailed introduction to the basis of the spatial spectrum estimation and to the model of DOA estimation; Studied the MUSIC algorithm of DOA estimation, given the MUSIC algorithm’s principles and steps; And through a number of MATLAB obtained the performance analysis of the MUSIC algorithm. Finally summarizes all the main work and results of the whole dissertation.
Keywords: DOA estimation; array signal processing; MUSIC algorithm
兰州理工大学毕业论文
目 录
第1章 绪论 ........................................................................................................................................ 1
1.1 研究背景及意义 .................................................................................................................... 1 1.2 DOA估计发展概述 ............................................................................................................... 2 1.3 论文的主要工作及内容安排 ................................................................................................ 3 第2章 DOA估计基础知识 .............................................................................................................. 5
2.1 DOA估计原理 ....................................................................................................................... 5
2.1.1 空间谱估计的系统结构 .............................................................................................. 5 2.1.2 DOA估计的基本原理 ................................................................................................. 6 2.2阵列信号DOA估计的常用方法 .......................................................................................... 7 2.3影响DOA估计结果的因素 .................................................................................................. 8 2.4 MATLAB简介 ....................................................................................................................... 9 2.5其他相关知识 ....................................................................................................................... 10
2.5.1分辨力 ......................................................................................................................... 10 2.5.2 Hermite矩阵............................................................................................................... 10 2.5.3协方差及协方差矩阵 ................................................................................................. 11
第3章 MUSIC算法 ........................................................................................................................ 12
3.1 MUSIC算法的提出 ............................................................................................................. 12 3.2波达方向估计问题中的阵列信号数学模型 ....................................................................... 12 3.3阵列协方差矩阵的特征分解 ............................................................................................... 15 3.4 MUSIC算法的原理及实现 ................................................................................................. 16 第4章 MUSIC算法的DOA估计仿真 ......................................................................................... 19
4.1 MUSIC算法的基本仿真 ..................................................................................................... 19 4.2 MUSIC算法DOA估计与阵元数的关系 .......................................................................... 20 4.3 MUSIC算法DOA估计与阵元间距的关系 ...................................................................... 20 4.4 MUSIC算法DOA估计与快拍数的关系 .......................................................................... 21 4.5 MUSIC算法DOA估计与信噪比的关系 .......................................................................... 22 4.6 MUSIC算法DOA估计与信号入射角度差的关系 .......................................................... 23 4.7 信号相干时MUSIC算法与改进MUSIC算法的仿真比较 ............................................ 23 第5章 MUSIC算法在应用中存在的问题及解决措施 ................................................................ 26
5.1通道失配对算法的影响 ....................................................................................................... 26