Does a function exist (prefferably in matlab) that finds a function in noise

Click For Summary

Discussion Overview

The discussion revolves around the existence of a function, preferably in MATLAB, that can identify a function within noisy data. It explores various methods and approaches to noise reduction and function extraction from data series.

Discussion Character

  • Exploratory, Technical explanation, Debate/contested

Main Points Raised

  • One participant suggests that any least-squares fit can be used to find a function in noise.
  • Another participant argues that curve fitting may not be effective and recommends using the FFT of the data series, followed by filtering and IFFT, emphasizing the importance of knowing the type of noise and the appropriate filter type.
  • A similar point is reiterated about the FFT and filtering approach, with an additional note that if the original data series is broadband, nonlinear noise reduction methods, such as the LAZY algorithm, may be necessary.
  • A later reply requests clarification on the LAZY algorithm, indicating a lack of understanding despite searching for information online.

Areas of Agreement / Disagreement

Participants present multiple competing views regarding the effectiveness of different methods for extracting functions from noisy data, and the discussion remains unresolved as to which method is superior.

Contextual Notes

There are limitations regarding the assumptions about the type of noise present in the data and the characteristics of the original data series, which are not fully addressed in the discussion.

j-lee00
Messages
93
Reaction score
0
Does a function exist (prefferably in matlab) that finds a function in noise

Cheers
 
Physics news on Phys.org
Sure, any least-squares fit will do that.
 
Curve fit won't work...try taking the fft of the data series, filter out the noise then ifft. The only thing you need to have an idea about is what type of noise is it and whether to use a high pass or low pass filter.
 
Dr Transport said:
Curve fit won't work...try taking the fft of the data series, filter out the noise then ifft. The only thing you need to have an idea about is what type of noise is it and whether to use a high pass or low pass filter.

This will work only if the original data series is not broadband, otherwise you need to use nonlinear noise reduction (look for the LAZY algorithm).
 
HI, Crosson, can you explain about "LAZY alggorithm", I searched on Google, and still don't understand it's meaning.
Thanks.
 

Similar threads

MATLAB Learn MATLAB Commands: Website for Variable and Function Explanation
  • · Replies 2 ·
Replies
2
Views
2K
MATLAB Recommendations for MATLAB MOOCs on Platforms like Udemy, EdX, and Coursera?
  • · Replies 3 ·
Replies
3
Views
1K
MATLAB Problem with Runge-Kutta script file
  • · Replies 4 ·
Replies
4
Views
2K
MATLAB MATLAB: Fluid Flow - Curl of a Vector Field
  • · Replies 2 ·
Replies
2
Views
4K
MATLAB Find Scilab & MATLAB Help Forums | Converting Code & Programming Questions
  • · Replies 6 ·
Replies
6
Views
4K
MATLAB Eliminate Noise from an Image in MATLAB
  • · Replies 15 ·
Replies
15
Views
2K
MATLAB Help converting MATLAB to Scilab code
  • · Replies 4 ·
Replies
4
Views
4K
MATLAB [~,~,~,~,~,~,~,~,spec] Too many output arguments...
  • · Replies 5 ·
Replies
5
Views
3K
MATLAB Need help defining an anonymous function in Matlab
  • · Replies 18 ·
Replies
18
Views
3K
MATLAB MATLAB IFFT doesn't match the analytical one
  • · Replies 10 ·
Replies
10
Views
3K