Скачать или смотреть How to Achieve Consistent ms-coherence Results in MATLAB and Python

A guide to aligning `ms-coherence` results between MATLAB's `mscohere` and Python's `scipy.signal.coherence` functions for signal analysis.
---
This video is based on the question https://stackoverflow.com/q/66539932/ asked by the user 'HappyPy' ( https://stackoverflow.com/u/2635863/ ) and on the answer https://stackoverflow.com/a/78063320/ provided by the user 'Ed Millerski' ( https://stackoverflow.com/u/5294518/ ) at 'Stack Overflow' website. Thanks to these great users and Stackexchange community for their contributions.

Visit these links for original content and any more details, such as alternate solutions, latest updates/developments on topic, comments, revision history etc. For example, the original title of the Question was: ms-coherence in matlab and python

Also, Content (except music) licensed under CC BY-SA https://meta.stackexchange.com/help/l...
The original Question post is licensed under the 'CC BY-SA 4.0' ( https://creativecommons.org/licenses/... ) license, and the original Answer post is licensed under the 'CC BY-SA 4.0' ( https://creativecommons.org/licenses/... ) license.

If anything seems off to you, please feel free to write me at vlogize [AT] gmail [DOT] com.
---
Understanding ms-coherence in MATLAB and Python

When analyzing the coherence between two signals, achieving consistent results across different programming environments can be challenging. A common scenario arises when comparing MATLAB's mscohere function with Python's scipy.signal.coherence. If you've found discrepancies in the results despite using the same parameters, you're not alone.

In this post, we will explore why these inconsistencies occur and how you can refine your approach to obtain matching results in both MATLAB and Python.

The Problem: Discrepancies in Coherence Calculation

Using the following example signals in both environments:

MATLAB Code Example

[[See Video to Reveal this Text or Code Snippet]]

MATLAB Output

Coherence values, cxy:

0.7489

0.6034

0.2813

0.3319

Frequencies, w:

0

0.1667

0.3333

0.5000

Python Code Example

[[See Video to Reveal this Text or Code Snippet]]

Python Output

Frequencies:

0

0.1667

0.3333

0.5

Coherence values:

0.7676

0.5319

0.3274

0.0776

Here, the coherence values returned by MATLAB and Python differ significantly despite using what appears to be the same set of parameters.

The Solution: Adjusting the Window Function

The root cause of the differing results lies in the implementation of the Hanning window in both functions. The hann function in SciPy generates a periodic Hanning window, which may not yield comparable results for smaller datasets.

Steps to Align Results

Adjust the Hanning Window in MATLAB:
Instead of using the default hann(6), specify the periodic option:

[[See Video to Reveal this Text or Code Snippet]]

Expected Results in MATLAB:
After this adjustment, the coherence values in MATLAB would be:

Coherence values, cxy:

0.7676

0.5319

0.3274

0.0776

Frequencies, w:

0

0.1667

0.3333

0.5000

Verify Results in Python:
Confirm that these values now match with the ones outputted from the Python code.

Conclusion

By making these adjustments to the Hanning window, you can ensure that the ms-coherence results between MATLAB and Python are aligned. This small tweak enhances the accuracy of your signal coherence analysis across different platforms, leading to consistent and reliable results.

Now you can analyze your signals with confidence, knowing that both MATLAB and Python are yielding the same insights!