Method for filtering signals from nonlinear dynamical systems

• US 6,424,985 B1
• Filed: 11/18/1998
• Issued: 07/23/2002
• Est. Priority Date: 11/18/1997
• Status: Expired due to Fees

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Lee, Chung-Yong
• Primary Examiner(s): NGO, CHUONG D
• Application Number:
  US09/195,263
• Time in Patent Office: 1,343 Days
• Field of Search: 708/300, 708/301, 708/303, 708/314, 708/322, 708/323, 702/191, 702/194, 702/195
• US Class Current: 708/322
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04
• CPC Class Codes: H03H 17/0261 Non linear filters

Abstract

Systems and methods for generalized iterative noise reduction for contaminated chaotic signals which are simple and easily implemented are disclosed. One system and method estimates the deviation of the observed signal from the nearest noise-free signal and uses the result to get a noise-reduced signal. Another system and method enhances the observed signal by iteratively seeking to minimize the signal according to a cost function.

Claims

1. A method for filtering signals from a nonlinear dynamical systems, comprising the steps of:

(a) setting an initial enhanced point to a noisy point obtained from the signals;

(b) estimating a deviation term between the initial enhanced point and the noisy point;

(c) weighting the estimated deviation term to prevent divergence in computing a new enhanced point;

(d) computing a new enhanced point; and

(e) iterating the steps (b) to (d) until the computed enhanced point converges to a true enhanced point which represents a noise reduced signal to provide filtered signals,

wherein the step of computing the new enhanced point includes computing a new enhanced point in accordance with

2. The method as recited in claim 1, wherein the step of estimating includes estimating the deviation term in accordance with the equation:

3. The method as recited in claim 1, wherein the step of weighting includes multiplying the deviation term by a coefficient.

4. The method as recited in claim 3, wherein the coefficient has a value sensitive to the value of the deviation term.

5. The method as recited in claim 1, wherein K₂ is adjusted according to a number of iterating steps.

6. A computer-based system for filtering signals from a nonlinear dynamical systems, comprising:

means for setting an initial enhanced point to a noisy point obtained from the signals;

means for estimating a deviation term between the initial enhanced point and the noisy point;

means for weighting the estimated deviation term to prevent divergence in computing a new enhanced point;

means for computing the new enhanced point wherein the means for computing iterates until the computed enhanced point converges to a true enhanced point which represents a noise reduced signal to provide filtered signals,

wherein the means for computing the new enhanced point includes computing a new enhanced point in accordance with

7. The system as recited in claim 6, wherein the means for estimating includes estimating the deviation term in accordance with the equation:

8. The system as recited in claim 6, wherein the means for weighting includes multiplying the deviation term by a coefficient.

9. The system as recited in claim 8, wherein the coefficient has a value sensitive to the value of the deviation term.

10. The system as recited in claim 6, wherein K₂ is adjusted according to a number of iterating steps.