基于分数低阶统计量的子空间拟合目标角度估计算法

毕业设计说明书（论文）外文摘要
Title   Estimation algorithm based on the fractional    Lower order moment statistics subspace fitting  Target angle.
Abstract
DOA（Direction of Arrival,DOA）estimation is huge prospects attracted more
And more attention in array passive detection,radar sonar communication etc.multiple signal classification（Multiple Signal Classification,MUSIC）algorithm,the rotation invariant subspace(Estimation of Signal Parameters Via Rotational invariance Techniques,ESPRIT)algorithm and weighted subspace fitting(weighted subspace fitting,WSF)algorithm are classified as DOA algorithm.However,these algorithms are based on the ideal of noise , when the noise is impulsive noisethe algorithm performance degradation.
The principles and steps of MUSIC、ESPRIT、WSF algorithm are introduced

in this paper and then the impulsive noise are definited and characterized by the introduction,and the fractional lower-order moments(fractional lower-order moment,FLOM),screening proportion (screened Ratio,SR) solution of impulsive noise algorithm are studied,and finally through the
Matlab simulation,the performance of WSF,FLOM-WSF,SR-WSF are compared.

Keywords: DOA estimation,music algorithm,esprit algorithm,wsf algorithm
1 绪论1
1.1 研究的背景与意义1
1.2 国内外研究现状1
1.3 本论文主要工作及内容安排2
2 空间谱估计理论基础3
2.1 阵列信号统计模型3
2.2 特征子空间分析5
3 基于特征子空间的DOA估计算法8
3.1 MUSIC算法8
3.2ESPRIT算法9
3.3 WSF算法10
4 冲击信号12
4.1  稳定分布的定义12
4.2  稳定分布的性质13
4.3  函数概率密度函数14
5 冲击噪声背景下的WSF算法15
5.1 FLOM-WSF算法15
5.2 Screened Ratio-WSF算法17
5.3 仿真性能对比19

1 绪论
1.1 研究的背景与意义
DOA 估计在阵列无源探测、雷达、声纳、通信等方面产生了巨大的应用和作用受到了很多科学家和学者重视[1].传统的DOA算法有多重信号分类（Multiple Signal Classification ,MUSIC）算法和旋转子空间不变（Estimation of Signal Parameters Via Rotational Invariance Techniques,ESPRIT）算法,其中music法是通过整合噪声子空间的特征矢量和导向矢量矩阵A的列矢量 的正交性来进行DOA估计的,esprit法是利用2个子阵之间的旋转不变关系进行DOA估计.但是缺点是精度都不高,不能适应在现代DOA估计之中.其中本次任务重点研究WSF算法, 加权子空间拟合(WSF)算法由Viberg首先提出,其分为加权信号子空间拟合（weighted subspace source fitting,WSSF ）算法和加权噪声子空间拟合(weighted subspace noise fitting,WNSF)算法.WSSF算法是将加权的信号子空间与阵列流型张成的空间进行拟合,WNSF算法则是利用加权的噪声子空间与阵列流型张成的空间正交的关系整合进行拟合. 基于分数低阶统计量的子空间拟合目标角度估计算法:http://www.lwfree.cn/tongxin/20171113/15863.html
------分隔线----------------------------