基于半实值Capon的高效波达方向估计算法

发布时间：2018-04-17 03:28

本文选题：波达方向估计 + 信号个数　；参考：《电子与信息学报》2015年04期

【摘要】：子空间类超分辨波达方向(DOA)估计算法需预先估计信号个数,当信号个数估计错误时,其性能会严重下降。该文提出一种新颖的半实值Capon(SRV-Capon)DOA估计算法。该算法继承了Capon算法无需信号个数估计的优点并克服了现存实值算法仅适用于中心对称阵列(CSA)的缺点。相比于Capon算法,SRV-Capon仅利用阵列接收数据协方差矩阵的实部求逆构建空间谱函数,实现了谱值计算的半实值化并将谱搜索的范围压缩至原来的一半,从而至少降低约75%的计算量。理论分析和仿真实验证明了该算法的有效性。
[Abstract]:The subspace-like super-resolution DOA (DOA) estimation algorithm needs to estimate the number of signals in advance. When the number of signals is estimated incorrectly, its performance will be seriously degraded.In this paper, a novel semi-real valued Capon(SRV-Capon)DOA estimation algorithm is proposed.This algorithm inherits the advantages of Capon algorithm which does not need to estimate the number of signals and overcomes the shortcoming that the existing real-valued algorithm is only suitable for centrosymmetric arrays (CSAs).Compared with the Capon algorithm, SRV-Capon only uses the real part of the array received data covariance matrix to construct the spatial spectral function, which realizes the half-real value of the spectral value calculation and reduces the range of spectral search to half of the original value, thus reducing the computation cost by at least 75%.Theoretical analysis and simulation results show the effectiveness of the algorithm.
【作者单位】：哈尔滨工业大学(威海)信息与电气工程学院;哈尔滨工业大学航天学院;
【基金】：山东省自然科学基金(ZR2014FQ003) 哈尔滨工业大学(威海)校科学研究基金(HIT(WH)201411)资助课题
【分类号】：TN911.23

【参考文献】