摘要
提出了一种基于Matching Pursuit算法的阵列信号降噪方法,有效改进了DOA估计的性能。该算法首先依据信号模型构建过完备原子库,然后通过梯度跟踪的稀疏分解方法找到最优原子。利用最优原子重构的信号在有效消除噪声的同时保留了信号的全部空间方位特征。计算机仿真证明,新方法与传统的DOA估计方法相结合有效地提高了阵列信号DOA估计的精度和准确度。
An effective scheme to remove noises from the array signal using matching pursuit algorithm is proposed.The denoised array signal which can be used for Direction of Arrival（DOA） estimation can improve the performance of algorithm greatly.Any signal can be decomposed adaptively into a linear weighed expansion of atomic vectors which are selected from an over-complete dictionary.This kind of denosing method can remove noises and remain characteristic of array signal at the same time,which can improve the performance of the DOA estimation no matter what kind of method is used.The simulation results show that the proposed method is effective.
出处
《电子信息对抗技术》
2011年第3期16-19,78,共5页
Electronic Information Warfare Technology
关键词
阵列信号处理
降噪
DOA估计
匹配跟踪
array signal processing
de-noising
direction of arrival estimation
matching pursuit
