I was successfully able to collect some CSI data using the existing tool(s) on GitHub (https://github.com/StevenMHernandez/ESP32-CSI-Tool). The CSI data is a pair of imaginary and real number which can use to extract amplitude and phase information. However, I have no prior knowledge of digital signal processing and I'm not sure how can to preprocess the CSI data to remove environmental noise like the Butterworth filter and/or PCA. I'm using Python and would like to know if this is an appropriate tool.
I'm trying to recreate steps from the following papers related to human activity detection using Wi-Fi CSI:
- SEARE: A System for Exercise Activity Recognition and Quality Evaluation Based on Green Sensing URL: https://ieeexplore.ieee.org/document/8247258
- WiFi Sensing with Channel State Information: A Survey URL: https://dl.acm.org/doi/10.1145/3310194