阵列信号处理论文：波束域特征空间波束形成及其方向图低旁瓣实现.doc

阵列信号处理论文：波束域特征空间波束形成及其方向图低旁瓣实现

【中文摘要】波束域处理方法借鉴并继承了某些阵元域的处理方法,是阵元域处理方法的拓展与创新。目前,随着研究的深入,广大学者逐渐对波束域的自适应波束形成越来越感兴趣,学者们将众多算法应用于波束域,研究其在波束域的性能,特征空间方法是阵元域常用的方法之一,基于此方法的波束形成器有着良好的收敛性和稳健性,通过结合波束域处理方法和特征空间方法获得一种兼具波束域性能和特征空间法性能的波束形成器是一个值得研究的课题。本文首先介绍了波束域处理方法和特征空间法的国内外研究进展,有许多阵元域处理方法在波束域得到成功应用,波束形成器性能得以大幅提高,特征空间方法在阵元域有着极好的表现,基于特征空间的各种算法也不计其数。其次,介绍了与阵列信号处理相关的基础知识,包括阵列的基本原理与阵列波束形成。在阵列基本原理中介绍了空间信号与信号的调制解调；在阵列波束形成中分析波束形成器中输出的信号的信噪比。接着,总结了产生波束的波束形成算法,包含两个大的部分：第一个大的部分是波束形成一般原理,其中又包括常规波束形成和最佳波束形成；第二个大的部分是自适应波束形成一般实现,其中又包括自适应采样矩阵和自适应加权。再次,提出了一种新的波束域波束形成算法,研究工作包含两个大的部分：第一个大的部分提出在波束域运用特征空间法,其中又包括阵列信号的转换问题,线性约束最小方差准则下的波束域处理问题,转换矩阵问题等。将波束域特征空间方法产生的波束方向图与传统方法产生的波束方向图进行比对,分析了本论文所提出的算法在稳健性与收敛性方面的性能改进、主副比的改进、旁瓣电平的改进、阵列输出信号干扰噪声比增益的改进、对阵列采样数数要求改进等。第二个大的部分提出了三种波束域低旁瓣方向图实现方法,第一种方法的仿真结果表明直接加窗截取阵列加权矢量的方法不再适用于波束域处理,第二种方法的仿真结果表明加窗截取波束域转换矩阵T的方法适用于当输入信号SNR相对于输入INR较大的情形,当输入信号SNR相对于输入INR较小时,得到的旁瓣水平不理想,第三种方法的仿真结果表明虚拟干扰法得到的波束方向图旁瓣随着虚拟的INR包络改变而改变,而通过程序编程可以任意改变虚拟的INR包络,是比较理想的降低旁瓣的方法。
【英文摘要】Beam-space processing method inherits and learns from some of the array-space methods, which is a development and innovation of array-space processing.Currently, with continuous in-depth study, more and more scholars are increasingly interested in beam-space adaptive beamforming. Many algorithms have been applied to beam-space, and their performance in the beam-space is also studied; Eigen-space method is one of the methods commonly used in array-space, the beamformer based on these methods has a good convergence speed and robustness, by combining the beam-space and eigen-space processing methods to obtain a beamformer with both beam-space and eigen-space properties is a research subject worthy of study.This paper firstly introduces the beam-space and eigen-space processing methods researth progress at home and abroad, there are many array-space methods having been successfully applied in beam-space, which has greatly improved the performance of the beamformer; The eigen-space method has an excellent performance in array-space, algorithms based on eigen-space are also countless.Secondly, basic knowledge related to array signal processing is introduced, including the basic principles of the array and the array beamforming. Spatial signal and its modulation-demodulation are introduced in the part of array basic principles; the output SNR from beamformer is analyzed in the part of array beamforming.Then, a summary of algorithms for beamforming is maked, containing two major parts:the first part is the general principles of beamforming, which includes the conventional beamforming and optimum beamforming; the second part is the general implementation methods of adaptive beamforming, which includes adaptive weight and adaptive sampling matrix.Finally, a new beamspace beamforming algorithm is proposed, the research work includes two major parts:Application of the eigen-space method in the beamspace is proposed in the first part, including the conversion of array signals, the beamspace processing under the criterion of linear constrained minimum variance, the transform matrix and so on. By comparing the beam patterns generated by beamspace eigen-based method and the conventional method, analysis about the new algorithm is proposed in terms of the robustness and convergence performance, the ratio of main and side lobe level, the gain of the array output signal to interference plus noise ratio, the required number of array samples and so on.Three methods of getting beam-space patterns are proposed for low-sidelobe in the second part. The simulation results of the first method show that the directly Window taper method on array weight vector is no longer suitable for beam-space processing; the simulation results of the second method show that Window taper method on beamspace transform matrix T is applicable to the case that the input signal SNR is relatively greater than input INR, when the input signal SNR is relatively smaller than the input INR, the sidelobe level obtained is not ideal; the simulation results of the third method show that beam pattern sidelobe obtained from the virtual-interference-method changes with the changes of virtual-interference INR envelope, a ideal level of side lobe can be obtained by changing virtual-interference INR envelope through programming, which is an ideal way to reduce the sidelobe.
【关键词】阵列信号处理 自适应波速形成 波束空间 特征空间 低旁瓣
【英文关键词】Array signal processing Adaptive beamforming Beam-space eigen-space Low sidelobe
【目录】波束域特征空间波束形成及其方向图低旁瓣实现

摘要

3-5

Abstract

5-6

目录

7-9

第一章 绪论

9-16

1.1 引言

9-11

1.2 国内外发展情况

11-15

1.3 论文的主要工作

15-16

第二章 阵列信号处理基础

16-29

2.1 阵列的基本原理

16-21

2.1.1 空间信号

18-20

2.1.2 信号的调制与解调

20-21

2.2 阵列波束形成简介

21-28

2.2.1 阵列信号模型

21-24

2.2.2 空间采样

24-25

2.2.3 波束形成

25-28

2.3 本章小结

28-29

第三章 自适应波束形成算法研究

29-41

3.1 波束形成一般原理

29-39

3.1.1 常规波束形成算法

30-33

3.1.2 最佳波束形成算法

33-39

3.2 自适应波束形成一般原理

39-40

3.2.1 自适应采样矩阵

39-40

3.2.2 自适应加权

40

3.3 本章小结

40-41

第四章 波束域特征空间自适应波束形成算法研究

41-79

4.1 波束域特征空间自适应波束形成

41-67

4.1.1 阵列信号向波束空间转换

42-43

4.1.2 线性约束最小方差准则下的波束域处理

43-51

4.1.3 转换矩阵

51-54

4.1.4 仿真结果与性能分析

54-67

4.2 波束域低旁瓣方向图实现

67-78

4.2.1 低旁瓣方向图实现的一般方法

67-68

4.2.2 波束域特征空间低旁瓣方向图实现

68-70

4.2.3 仿真结果与性能分析

70-78

4.3 本章小结

78-79

第五章 总结展望

79-81

参考文献

81-86

致谢

86-87

攻读硕士学位期间发表的学术论文

87