1: \begin{abstract}

2:  When a linear system $A x = y$ is solved by means of iterative methods (mainly CG and

3: GMRES) and the convergence rate is slow, one may consider a preconditioner $P $ and move

4: to the preconditioned system $P^{-1}Ax = P^{-1}y$. The use of such preconditioner changes

5: the spectrum of the matrix defining the system and could result into a great

6: acceleration of the convergence rate. The construction of \textit{optimal rank} preconditioners is

7: strongly related to the possibility of splitting $A$ as $A = P + R + E$, where $E$ is a

8: small perturbation and $R$ is of low rank \cite{tyrty}. In the present work we extend the black-dot 

9: algorithm  for the computation of such splitting for $P$ circulant (see \cite{ot}), to the case where $P$ is in $\mss A$, for several known low-complexity matrix

10: algebras $\mss A$. The algorithm so obtained is particularly efficient when $A$ is Toeplitz plus Hankel like. 

11: %We also show that for a suitable class of matrices, the splitting $A = P + R + E$

12: %is actually ensured, being $P$ an element of a $\vphi$-circulant or Hartley-type algebra.

13: We finally discuss in detail the existence and the properties of the decomposition  $A=P+R+E$ when $A$ is Toeplitz, also extending to the $\vphi$-circulant and Hartley-type cases some results previously known for $P$ circulant.

14: \end{abstract}

15: