摘要
实际阵列测向系统不可避免地存在各种系统误差,使得经典的超分辨率测向算法的性能极大地恶化。利用正交级数展开来描述理想阵列响应和实际阵列响应的关系,通过引入球形调谐函数作为正交基并求出多个校正矩阵,能够使得校正矩阵作用于理想阵列响应后更好地逼近实际阵列响应。本文以此为基础,研究了阵列响应与球形调谐函数之间的内在联系,并对球形调谐基函数进行了优化,以进一步提高原方法测向精度。蒙特卡罗仿真表明,在相位误差为±20°、幅度误差为±2dB条件下,采用三个校正矩阵时,改进后的方法比原方法方位角均方根误差减小89%,俯仰角均方根误差减小52%。本文提出的改进测向算法,对提高实际阵列测向系统的鲁棒性具有一定指导作用。
Practical arrays can＇t avoid system errors,which dramatically degrade the performance of the classical super resolution direction finding methods. To deal with this problem,using series expansion to describe the relationship between ideal array response and real response can make the ideal response approach more accurately to the real response when applying error matrices through introducing spherical harmonics and multiple calibration matrices,thereby getting high estimation precision. Based on this method,this paper optimizes the spherical harmonics functions based on relationship between array response and spherical harmonics in order to further improve the angle estimation precision. Monte Carlo simulations show that when phase errors lie between-20 deg and 20 deg,gain errors lie between-2 dB and 2 dB,the azimuth＇s Root Mean Squared Error （RMSE） is decreased by 89% via optimization while the elevation＇s RMSE decreased by 52%,provided that three calibration matrices are applied. The newly proposed direction finding algorithm can give some instructions about improving practical system＇s robustness to system errors.
出处
《微计算机信息》
2010年第36期162-164,63,共4页
Control & Automation
关键词
鲁棒测向
阵列误差
级数展开
基函数
robust direction finding
array error
series expansion
base functions
