Multiple shift maximum element sequential matrix diagonalisation for parahermitian matrices

Multiple shift maximum element sequential matrix diagonalisation for parahermitian matrices
复制标题

准埃尔米特矩阵的多移位最大元素顺序矩阵对角化

DOI:
10.1109/ssp.2014.6884638
复制
发表时间:
2014
期刊:
2014 IEEE Workshop on Statistical Signal Processing (SSP)
影响因子:
--
通讯作者:
I. Proudler
I. Proudler
中科院分区:
--
文献类型:
--
作者:
J. Corr;K. Thompson;Stephan Weiss;J. McWhirter;Soydan Redif;I. Proudler

文献摘要

被引文献

相似文献

Paraher-mitian矩阵的多项式特征值分解可以使用迭代方法近似计算,例如顺序矩阵对角化(SMD)算法。在本文中,我们提出了一种改进的SMD算法,相比现有的SMD方法,消除更多的非对角能量每一步。这导致了更快的收敛,同时只会导致复杂性的边际增加。我们激励的方法,证明其收敛性,并证明了一些结果,强调算法的性能。
A polynomial eigenvalue decomposition of paraher-mitian matrices can be calculated approximately using iterative approaches such as the sequential matrix diagonalisation (SMD) algorithm. In this paper, we present an improved SMD algorithm which, compared to existing SMD approaches, eliminates more off-diagonal energy per step. This leads to faster convergence while incurring only a marginal increase in complexity. We motivate the approach, prove its convergence, and demonstrate some results that underline the algorithm's performance.