Filter Design for SHM Feature Engineering

2 views (last 30 days)
Luke Megonigal
Luke Megonigal on 22 Feb 2021
Answered: Hari on 24 Apr 2024
Hello,
My research involves feature engineering signals. I am wondering if anyone can point me in the right direction for using/designing filters to generate features for classification and regression analyses of signals. I found a good source video about using the peaks of filters. However, it is a shot in the dark for me to develop the right filter. Let me know if anyone can point me in the right direction.
Luke

Answers (1)

Hari
Hari on 24 Apr 2024
Hi Luke Megonigal,
To guide you in filter design for feature engineering in signals, consider the following tools/function in MATLAB:
1. Filter Design Tools in MATLAB:
  • "designfilt" Function: MATLAB provides the "designfilt" function, which allows for the design of digital filters based on specifications you provide. This function is versatile and supports various filter types and design methods.
  • Filter Designer App: For a more interactive approach, MATLAB's Filter Designer app provides a graphical interface for designing and analyzing filters. It can be particularly useful if you're exploring different filter designs and their effects on your signal.
2. Feature Extraction Using Filtered Signals:
  • Once you have designed a suitable filter, apply it to your signal using the "filter" function in MATLAB. Analyze the filtered signal to identify features of interest, such as peaks, which can be detected using the "findpeaks" function.
  • Consider extracting various statistical features from the filtered signal, such as mean, standard deviation, skewness, and kurtosis, which might be relevant for your classification or regression models.
Iterative Design and Validation: Filter design for feature engineering is often an iterative process. You may need to experiment with different filter types, parameters, and extracted features to find the combination that provides the most discriminative or predictive power for your application.
References for further exploration:
Hope this helps!

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!