摘要
ESPRIT算法在有效性和稳定性方面都有非常突触的表现,所以其受到了广泛的研究。同时不可避免的要进行特征值分解,无法满足实时系统的要求。因此,提出了一种提取信号子空间的快速算法,通过对旋转矩阵的改进,不需要进行特征值分解和繁琐的迭代计算,节约了计算时间,使得所提算法可以运用于实时处理,并且取得了良好的效果。
ESPRIT algorithm is very synapses in the effectiveness and stability, so it has been a wide range of research as a major superresolution algorithm. At the same time inevitable to carry out eigende-composition, and unable to meet the requirements of real-time systems, based on this consideration, a extracting signal subspace of ESPRIT fast algorithm, through the improvement of the rotation matrix, so it need not eigende-composition and iterativ calculation, saving the time of calculation and making it apply to real-time processing, and receiving good perforce.
出处
《计算机工程与设计》
CSCD
北大核心
2009年第13期3057-3059,共3页
Computer Engineering and Design
关键词
有效性
特征值分解
实时系统
信号子空间
实时处理
effectiveness
eigende-composition
real-timesystems
signal subspace
real-timeproeessing
