基于压缩感知理论的信源定位算法研究
发布时间：2019-06-03 13:59
【摘要】：阵列信号处理作为现代信号处理领域当中的一个重要分支,广泛应用于军用和民用领域,本文着重研究其中的信源定位算法。根据信源与阵列天线间的距离,信源分为远场信源和近场信源,远场信源仅由波达方向(Direction of Arrival,DOA)表征而近场信源同时需要DOA和距离参数表征。基于压缩感知理论的信源定位算法能够降低所需快拍数并提高定位精度,为解决信源定位问题提供了新的思路。论文主要工作如下:(1)研究远近场信源定位模型及经典方法。研究远场、近场和远近混合场信源的阵列接收数据模型,仿真验证经典的远场DOA估计算法和混合场信源定位算法的性能;仿真三种信源定位场景下参数估计的克拉美罗界。(2)研究基于压缩感知理论的信源定位模型及算法。研究将压缩感知理论用于信源定位问题的稀疏模型,验证凸优化算法和贪婪算法用于DOA估计的性能以及仿真验证一种结合压缩感知和四阶累积量的混合场定位算法。本文创新点如下:(1)改进一种基于OMP(Orthogonal Matching Pursuit)算法的DOA估计方法。通过研究导向矢量间的相关性,分析压缩感知理论用于信源定位的不确定性,以及通过对接收数据进行向量化,研究协方差域的操作对重构算法的性能影响。改进一种基于OMP算法的改进DOA估计算法,该算法能够自适应利用先验信息,在低信噪比和少快拍情况下的估计性能优势明显。(2)改善两种传统定位算法的空间谱并提出一种新的混合场信源定位算法。对两种传统混合场信源定位算法进行改进,用稀疏重构算法改善其空间谱。运用压缩感知理论提出一种适应混合场定位算法,该算法只适用于多频点信号,通过对接收数据的不同频率分量联合稀疏表示,可以提高定位精度并且能够定位与阵元数相同的信源。该算法对于纯近场信源定位的情况同样适用。
[Abstract]:Array signal processing, as an important branch of modern signal processing, is widely used in military and civil fields. This paper focuses on the source location algorithm. According to the distance between the source and the array antenna, the source is divided into far field source and near field source. The far field source is only represented by the direction of arrival (Direction of Arrival,DOA), while the near field source needs DOA and distance parameter representation at the same time. The source location algorithm based on compressed perception theory can reduce the number of fast beats and improve the positioning accuracy, which provides a new idea for solving the problem of source location. The main work of this paper is as follows: (1) the far and near field source location model and classical methods are studied. The array receiving data model of far-field, near-field and far-near mixed field sources is studied, and the performance of classical far-field DOA estimation algorithm and hybrid field source location algorithm is verified by simulation. The Keramero bound of parameter estimation in three source localization scenarios is simulated. (2) the source location model and algorithm based on compressed perception theory are studied. In this paper, the sparse model of compressed sensing theory for source location is studied, the performance of convex optimization algorithm and greedy algorithm for DOA estimation is verified, and a hybrid field location algorithm combining compressed perception and fourth-order cumulant is verified by simulation. The innovations of this paper are as follows: (1) an improved DOA estimation method based on OMP (Orthogonal Matching Pursuit) algorithm. By studying the correlation between guidance vectors, the uncertainty of compressed perception theory used in source location is analyzed, and the effect of covariance domain operation on the performance of reconstruction algorithm is studied by vector quantification of received data. An improved DOA estimation algorithm based on OMP algorithm is improved, which can adaptively make use of prior information. In the case of low signal-to-noise ratio (SNR) and low fast beat, the estimation performance has obvious advantages. (2) the spatial spectrum of the two traditional localization algorithms is improved and a new hybrid field source location algorithm is proposed. Two traditional hybrid field source localization algorithms are improved, and their spatial spectrum is improved by sparse reconstruction algorithm. Based on the compressed sensing theory, an adaptive hybrid field location algorithm is proposed, which is only suitable for multi-frequency point signals, and is represented by sparse representation of different frequency components of the received data. It can improve the positioning accuracy and locate the same number of sources as the array elements. The algorithm is also suitable for the location of pure near-field sources.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN911.7
本文编号：2491984
[Abstract]:Array signal processing, as an important branch of modern signal processing, is widely used in military and civil fields. This paper focuses on the source location algorithm. According to the distance between the source and the array antenna, the source is divided into far field source and near field source. The far field source is only represented by the direction of arrival (Direction of Arrival,DOA), while the near field source needs DOA and distance parameter representation at the same time. The source location algorithm based on compressed perception theory can reduce the number of fast beats and improve the positioning accuracy, which provides a new idea for solving the problem of source location. The main work of this paper is as follows: (1) the far and near field source location model and classical methods are studied. The array receiving data model of far-field, near-field and far-near mixed field sources is studied, and the performance of classical far-field DOA estimation algorithm and hybrid field source location algorithm is verified by simulation. The Keramero bound of parameter estimation in three source localization scenarios is simulated. (2) the source location model and algorithm based on compressed perception theory are studied. In this paper, the sparse model of compressed sensing theory for source location is studied, the performance of convex optimization algorithm and greedy algorithm for DOA estimation is verified, and a hybrid field location algorithm combining compressed perception and fourth-order cumulant is verified by simulation. The innovations of this paper are as follows: (1) an improved DOA estimation method based on OMP (Orthogonal Matching Pursuit) algorithm. By studying the correlation between guidance vectors, the uncertainty of compressed perception theory used in source location is analyzed, and the effect of covariance domain operation on the performance of reconstruction algorithm is studied by vector quantification of received data. An improved DOA estimation algorithm based on OMP algorithm is improved, which can adaptively make use of prior information. In the case of low signal-to-noise ratio (SNR) and low fast beat, the estimation performance has obvious advantages. (2) the spatial spectrum of the two traditional localization algorithms is improved and a new hybrid field source location algorithm is proposed. Two traditional hybrid field source localization algorithms are improved, and their spatial spectrum is improved by sparse reconstruction algorithm. Based on the compressed sensing theory, an adaptive hybrid field location algorithm is proposed, which is only suitable for multi-frequency point signals, and is represented by sparse representation of different frequency components of the received data. It can improve the positioning accuracy and locate the same number of sources as the array elements. The algorithm is also suitable for the location of pure near-field sources.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN911.7
【参考文献】
相关期刊论文 前2条
1 梁国龙;韩博;林旺生;王丹;;基于稀疏信号重构的近场源定位[J];电子学报;2014年06期
2 蒋佳佳;段发阶;陈劲;常宗杰;;一种高精度的近场与远场混合源定位算法[J];天津大学学报(自然科学与工程技术版);2013年12期
，本文编号：2491984
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2491984.html
