1 引言矩阵满秩分解是线性代数的基本分解方法之一,在广义逆矩阵的求解过程中起着重要的作用.矩阵A的满秩分解及A的Moore-Penrose逆不仅有着广泛的现实应用,而且也有理论研究意义,尤其是在数理统计,系统理论,优化计算和控制论等许多领域应用十分广泛.如用计算机对具有对称性质的图像进行采样,所得到的数据矩阵具有行或列对
In this paper, we prove the quantitative correspondence of F*and G*matrices of row or column symmetric matrices and its mother matrix A. We also analysis the Hermite standard form and address two new algorithms for the symmetric matrix, based on, we can save dramatically the CPU time and memory. At the same time, we analysis the Moore-Penrose inverse, and give the algorithms of Moore-Penrose inverse for row or column symmetric matrix.