1: \begin{abstract}{Time--frequency analysis for non--linear and non--stationary signals is extraordinarily challenging.

2: To capture features in these signals, it is necessary for the analysis methods  to be local, adaptive

3: and stable. In recent years, decomposition based analysis methods, such as the empirical mode

4: decomposition (EMD) technique pioneered by Huang et al.,  were developed by different research

5: groups. These methods decompose a signal into a finite number of components

6: on which the time--frequency analysis can be applied more effectively.

7: 

8: In this paper we consider the Iterative Filtering (IF) approach as an alternative to EMD. We provide

9: sufficient conditions on the filters that ensure the convergence of IF applied to any $L^2$ signal. Then

10: we propose a new technique, the Adaptive Local Iterative Filtering (ALIF) method, which uses

11: the IF strategy together with an adaptive and data driven filter length selection to achieve the decomposition.

12: Furthermore we design smooth filters with compact support from solutions of

13: Fokker--Planck equations (FP filters) that can be used within both IF and ALIF methods.

14: These filters fulfill the derived sufficient conditions for the convergence of the IF algorithm.

15: Numerical examples are given to demonstrate the performance and stability of IF

16: and ALIF techniques with FP filters. In addition, in order to have a complete and truly local

17: analysis toolbox for non--linear and non--stationary signals, we propose new definitions for

18: the instantaneous frequency and phase which depend exclusively on local properties of a signal.}\end{abstract}