利用信号非圆特性的阵列测向技术研究
发布时间：2018-07-20 20:15
【摘要】：现代雷达、通信系统中,经常使用具有非圆特性的调制信号,在阵列测向技术领域中,信号具有非圆特性会使得其伪协方差矩阵中也含有信息。因此如果合理地利用信号非圆特性将可以增加信息利用效率,提高阵列测向算法各项估计性能,因此,利用信号非圆特性的阵列测向方法已经成为近年来的一个研究热点。目前已有的适用于非圆信号的阵列测向方法多用于估计只有最大非圆率信号的情况,而当入射信号中同时存在圆信号、非圆信号时,会存在对信源数目估计准确性不高、估计精度不足、分辨能力较差等现象。本文针对利用信号非圆特性的阵列测向方法做出了较为深入的研究,主要内容如下:1.分析了现有信源数目估计方法的不足和局限性,提出了可以在色噪声背景下对信源数目进行估计的基于特征空间投影的CEP算法,并将其应用于虚拟扩展阵列,提出了可以在圆信号和非圆信号同时入射的情形下对等效信源数目进行估计的基于均值聚类的MCEP算法。与目前已有方法相比,MCEP算法无论在信噪比和快拍数较低还是较高情况下均有较良好的估计性能,且在色噪声背景下仍然能够保持有效估计。2.针对DOA参数估计问题,分析了现有方法在入射信号中同时存在圆信号、一般非圆率信号和最大非圆率信号时会面临存在伪峰和分辨力不足等问题的原因。提出了可以提高分辨力的HRNC-MUSIC算法,该方法可以对不同类别的信号依次做出估计,减少了各类信号间相互影响,提高分辨能力。仿真结果表明,当混合入射信号中各类信号之间的空间角距离相距较小时,HRNC-MUSIC算法可以明显提高对入射信号的分辨能力和估计精度。3.针对应用于极化敏感阵列的DOA估计算法运算量高、阵列扩展能力不强等现象,提出了基于平面双极化敏感阵列的DRL-MUSIC算法,并研究了同时利用信号非圆特性和极化信息的测向方法,提出了基于四元数的QNC-MUSIC算法和基于降维长矢量的DRLNC-MUSIC算法,填补了极化敏感阵列下适用于非圆信号测向方法的空白。仿真结果表明,当入射信号只含最大非圆率信号时,QNC-MUSIC算法的估计性能优于DRLNC-MUSIC算法;而当入射信号中既存在非圆信号又存在圆信号时,DRLNC-MUSIC算法的估计性能要优于QNC-MUSIC算法,且DRLNC-MUSIC算法可以同时对极化参数做出联合估计,不耗费额外的计算量。
[Abstract]:In modern radar and communication systems, modulation signals with non-circular characteristics are often used. In the field of array direction finding technology, the non-circular characteristics of signals make the pseudo-covariance matrix contain information. Therefore, if the signal non-circular characteristics are used reasonably, the efficiency of information utilization can be increased and the estimation performance of the array direction finding algorithm can be improved. Therefore, the array direction finding method based on the non-circular characteristics of signals has become a hot research topic in recent years. Most of the current array direction finding methods for non-circular signals are used to estimate only the maximum non-circular rate signals, but when there are circular signals and non-circular signals in the incident signals, the estimation of the number of sources is not accurate. The estimation accuracy is insufficient, the resolution ability is poor and so on. In this paper, the array direction-finding method based on the non-circular characteristics of signals is studied in depth, the main contents are as follows: 1. In this paper, the shortcomings and limitations of the existing methods for estimating the number of sources are analyzed. A CEP algorithm based on feature space projection is proposed to estimate the number of sources under the background of color noise, and it is applied to the virtual extended array. An MCEP algorithm based on mean value clustering is proposed to estimate the number of equivalent sources when both circular and non-circular signals are incident at the same time. Compared with the existing methods, the MCEP algorithm has better estimation performance in the case of low SNR and high shot number, and can still maintain an effective estimation of .2. under the background of color noise. Aiming at the problem of DOA parameter estimation, this paper analyzes the reasons why the existing methods have circular signals in incident signals simultaneously, and the problems such as false peaks and insufficient resolution will be encountered in general non-circular rate signals and maximum non-circular rate signals. The HRNC-MUSIC algorithm which can improve the resolution is proposed. This method can estimate the different kinds of signals in order to reduce the interaction between different signals and improve the resolution ability. The simulation results show that the HRNC-music algorithm can obviously improve the resolution and estimation accuracy of the incident signals when the distance between the different signals is small. A novel DRL-MUSIC algorithm based on planar dual polarimetric sensing array is proposed to solve the problem of high computational complexity and low expansion ability of DOA estimation algorithm applied to polarization-sensitive arrays. At the same time, the method of direction finding based on the non-circular characteristic and polarization information is studied. The QNC-MUSIC algorithm based on quaternion and the DRLNC-MUSIC algorithm based on reduced dimension vector are proposed, which fills the blank of the direction finding method for non-circular signals under polarization sensitive array. Simulation results show that the estimation performance of QNC-MUSIC algorithm is better than that of DRLNC-MUSIC algorithm when the incident signal contains only the maximum non-circular rate signal, and the estimation performance of DRLNC-MUSIC algorithm is better than that of QNC-MUSIC algorithm when there are both non-circular and circular signals in the incident signal. Moreover, the DRLNC-MUSIC algorithm can simultaneously estimate the polarization parameters without additional computation.
【学位授予单位】：哈尔滨工程大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN911.7
[Abstract]:In modern radar and communication systems, modulation signals with non-circular characteristics are often used. In the field of array direction finding technology, the non-circular characteristics of signals make the pseudo-covariance matrix contain information. Therefore, if the signal non-circular characteristics are used reasonably, the efficiency of information utilization can be increased and the estimation performance of the array direction finding algorithm can be improved. Therefore, the array direction finding method based on the non-circular characteristics of signals has become a hot research topic in recent years. Most of the current array direction finding methods for non-circular signals are used to estimate only the maximum non-circular rate signals, but when there are circular signals and non-circular signals in the incident signals, the estimation of the number of sources is not accurate. The estimation accuracy is insufficient, the resolution ability is poor and so on. In this paper, the array direction-finding method based on the non-circular characteristics of signals is studied in depth, the main contents are as follows: 1. In this paper, the shortcomings and limitations of the existing methods for estimating the number of sources are analyzed. A CEP algorithm based on feature space projection is proposed to estimate the number of sources under the background of color noise, and it is applied to the virtual extended array. An MCEP algorithm based on mean value clustering is proposed to estimate the number of equivalent sources when both circular and non-circular signals are incident at the same time. Compared with the existing methods, the MCEP algorithm has better estimation performance in the case of low SNR and high shot number, and can still maintain an effective estimation of .2. under the background of color noise. Aiming at the problem of DOA parameter estimation, this paper analyzes the reasons why the existing methods have circular signals in incident signals simultaneously, and the problems such as false peaks and insufficient resolution will be encountered in general non-circular rate signals and maximum non-circular rate signals. The HRNC-MUSIC algorithm which can improve the resolution is proposed. This method can estimate the different kinds of signals in order to reduce the interaction between different signals and improve the resolution ability. The simulation results show that the HRNC-music algorithm can obviously improve the resolution and estimation accuracy of the incident signals when the distance between the different signals is small. A novel DRL-MUSIC algorithm based on planar dual polarimetric sensing array is proposed to solve the problem of high computational complexity and low expansion ability of DOA estimation algorithm applied to polarization-sensitive arrays. At the same time, the method of direction finding based on the non-circular characteristic and polarization information is studied. The QNC-MUSIC algorithm based on quaternion and the DRLNC-MUSIC algorithm based on reduced dimension vector are proposed, which fills the blank of the direction finding method for non-circular signals under polarization sensitive array. Simulation results show that the estimation performance of QNC-MUSIC algorithm is better than that of DRLNC-MUSIC algorithm when the incident signal contains only the maximum non-circular rate signal, and the estimation performance of DRLNC-MUSIC algorithm is better than that of QNC-MUSIC algorithm when there are both non-circular and circular signals in the incident signal. Moreover, the DRLNC-MUSIC algorithm can simultaneously estimate the polarization parameters without additional computation.
【学位授予单位】：哈尔滨工程大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN911.7
【相似文献】
相关期刊论文 前10条
1 司伟建;崔冬槐;司锡才;;阵列测向多值模糊问题研究[J];弹箭与制导学报;2004年S8期
2 高昭昭;李世文;张晓芸;;基于分布式稀疏理论的高分辨阵列测向[J];电子信息对抗技术;2013年05期
3 罗朗;;高分辨阵列测向中信号相关性问题的研究[J];实验科学与技术;2006年02期
4 陈超;蔡科;梁剑;;单通道阵列测向系统的设计与实现[J];舰船电子对抗;2007年06期
5 毛健,郭振民;高分辨阵列测向的最小二乘自校正算法[J];声学与电子工程;1995年04期
6 屈金佑;游志刚;张剑云;;基于相位补偿的单通道接收阵列测向方法[J];现代防御技术;2007年05期
7 施龙飞;王伟;赵研;任季中;顾明超;;基于径向对称二元阵的非同向阵列测向方法[J];宇航学报;2013年04期
8 毛健;葛利嘉;陈天麒;;高分辨阵列测向的全局优化自校正算法[J];电子科技大学学报;1993年05期
9 陈悦龙;;高分辨阵列测向技术及其仿真[J];四川职业技术学院学报;2014年02期
10 刘章孟;周一宇;吴海斌;;非圆信号的贝叶斯稀疏重构阵列测向方法[J];航空学报;2014年03期
相关会议论文 前1条
1 司伟建;;相关辐射源特性及产生方法研究[A];中国造船工程学会电子技术学术委员会2006学术年会论文集（下册）[C];2006年
相关博士学位论文 前4条
1 朱f，
本文编号：2134647
本文链接：https://www.wllwen.com/kejilunwen/wltx/2134647.html
