1: \begin{abstract}

2: In this note, we study   the fast solution of Toeplitz linear

3: systems with coefficient matrix  $T_n(f)$, where the generating

4: function $f$ is nonnegative and has a unique zero at zero of any

5: real positive order $\theta$. As preconditioner we choose a matrix

6: ${\tau}_n(f)$ belonging to the so-called $\tau$ algebra, which is

7: diagonalized by the sine transform associated to the discrete

8: Laplacian. In previous works, the spectral equivalence of the

9: matrix sequences $\{{\tau}_n(f)\}_n $ and $\{T_n(f) \}_n$ was

10: proven under the assumption that the order of the zero is equal to

11: $2$: in other words the preconditioned matrix sequence

12: $\{{\tau}^{-1}_n(f)T_n(f) \}_n $ has eigenvalues, which are

13: uniformly away from zero and from infinity. Here we prove a

14: generalization of the above result when $\theta<2$. Furthermore,

15: by making use of multiple step preconditioning, we show that the

16: matrix sequences $\{{\tau}_n(f)\}_n $ and $\{T_n(f) \}_n$   are

17: essentially spectrally equivalent for every $\theta>2$, i.e., for

18: every $\theta>2$, there exist $m_\theta$ and a positive interval

19: $[\alpha_\theta,\beta_\theta]$ such that all the eigenvalues of

20: $\{{\tau}^{-1}_n(f)T_n(f) \}_n $ belong to this interval, except

21: at most $m_\theta$ outliers larger than $\beta_\theta$. Such a

22: nice property, already known only  when  $\theta$ is an even

23: positive integer greater than 2, is coupled with the fact that the

24: preconditioned sequence has an eigenvalue cluster at one, so that

25: the convergence rate of the associated preconditioned conjugate

26: gradient method is optimal. As a conclusion we discuss possible

27: generalizations and we present selected numerical experiments.

28: 

29: \end{abstract}

30: