1: \begin{abstract}

2: In this paper we investigate the computational aspects of some recently

3: proposed iterative methods for approximating the canonical tight and

4: canonical dual window of a Gabor frame $\left(g,a,b\right)$. The

5: iterations start with the window $g$ while the iteration steps comprise

6: the window $g$, the $k^{th}$ iterand $\gamma_{k}$, the frame operators

7: $S$ and $S_{k}$ corresponding to $\left(g,a,b\right)$ and $\left(\gamma_{k},a,b\right)$,

8: respectively, and a number of scalars. The structure of the iteration

9: step of the method is determined by the envisaged convergence order

10: $m$ of the method. We consider two strategies for scaling the terms

11: in the iteration step: norm scaling, where in each step the windows

12: are normalized, and initial scaling where we only scale in the very

13: beginning. Norm scaling leads to fast, but conditionally convergent

14: methods, while initial scaling leads to unconditionally convergent

15: methods, but with possibly suboptimal convergence constants. The iterations,

16: initially formulated for time-continuous Gabor systems, are considered

17: and tested in a discrete setting in which one passes to the appropriately

18: sampled-and-periodized windows and frame operators. Furthermore, they

19: are compared with respect to accuracy and efficiency with other methods

20: to approximate canonical windows associated with Gabor frames.

21: \end{abstract}

22: