Hi Florian,
I understand that you want to analyze signals and identify the periodic distance between extrema (maximum and minimum points) in a way that is robust against noise, you can use a combination of signal processing techniques and clustering. MATLAB provides a comprehensive set of tools for this purpose.
Here's a general steps that you might find useful:
- Detrend and Smooth the Signal: This helps in reducing the impact of noise. You can use detrend to remove the linear trend and smoothdata for smoothing the signal.
- Find Extrema: Use findpeaks to find maxima. For minima, you can invert the signal and use findpeaks again.
- Clustering Extrema : If the extrema are clustered in groups (due to noise or signal characteristics), you might want to cluster them to find representative extrema. K-means clustering (kmeans) on the positions of these extrema can be useful, though it's optional and depends on your specific signal characteristics.
You can refer the documentations below:
I hope it helps!
