The maximal positive definite solution of the nonlinear matrix equation \(X + A^*X^{-1}A+B^*X^{-1}B = I \) (Q6066823)

The authors propose a new inversion free variant of the fixed point iteration method for finding the maximal positive definite solution of matrix equation \[ X + A^* X^{-1} A + B^* X^{-1} B = I , \tag{1} \] where \(A, B \in \mathbb C^{n\times n}\) are invertible matrices, \(A^*\) and \(B^*\) are conjugate transposes of \(A\) and \(B\), and \(I\) is the identity matrix. Namely, their method is as follows: first take \[ X_0=Y_0=I, \] then iterate \[ Y_{r+1}=-\frac{1}{2}I+Y_r\left( \frac{5}{2}I+\frac{1}{2}X_r-\frac{3}{2}X_rY_r\right), \] \[ X_{r+1}=I-A^*Y_{r+1}A-B^*Y_{r+1}B, \qquad r=0,1,2,\ldots \] The authors explain the idea behind this method and prove some theorems concerning necessary and sufficient conditions for equation (1) to have solutions. Then, they present numerical examples showing the behavior and efficiency of their iterative method. They claim that the comparison of the results shows that the new algorithm is more accurate and requires fewer operations than other algorithms.