基于宽带欠采样阵列的频率估计算法研究  

Algorithm of Frequency Estimation Based on Wide-band Under-sampled Array

作　　者：赵曼[1] 陈辉[2] 王俊[2] 邓钰[2] 

机构地区：[1]空军预警学院研究生管理大队,武汉430019 [2]空军预警学院三系,武汉430019

出　　处：《空军预警学院学报》2013年第4期263-266,共4页Journal of Air Force Early Warning Academy

基　　金：新世纪优秀人才支持计划资助项目(NCET-10-0903);电子信息控制重点实验室资助项目

摘　　要：鉴于宽带欠采样阵列的常规FFT频率估计算法只适用于输入信噪比较高且精度要求不高的单信号频率估计,给出了多组采样频率联合解模糊的方法,在无需增加复杂配对算法的情况下,实现了多信号频率的无模糊估计.针对常规FFT法估计精度不高的问题,提出了一种分级高分辨频率估计方法(CMUSIC算法).该算法估计精度可达到10kHz级,相对于常规FFT法的MHz级提高了至少一个数量级.最后,仿真结果验证了本文所提方法的有效性.Seeing that the conventional FFT frequency estimation algorithm for the wide-band under-sampling can only apply to the case of higher input SNR and the single-signal frequency estimation that is used for the precision requirement being not higher, this paper puts forward a method of the joint solution ambiguity of multiple group sampling frequencies, which could implement the unambiguous estimation of multiple signal frequencies under the circumstances of no need to increase the complex paring algorithm. Aimed at the estimation precision of conventional FFT method being lower, a frequency estimation algorithm with graded high resolution is proposed. The estimation precision of this algorithm can be up to the level of 10 kHz, and an order of magnitude is at least improved, with respect to the order of MHz of conventional FFT algorithm. And also the performance analysis is performed on the FFT frequency measurement of single array element, the FFT frequency measurement of synthetic arrays and the high resolution frequency measurement of arrays. Finally, the effectiveness of this proposed method is verified by simulation results.

关 键 词：宽带欠采样 时域欠采样 频率估计 快速傅里叶变换 多重信号分类算法 

分 类 号：TN957[电子电信—信号与信息处理]