How Can I Move from 2D Graphs to Spatial Detection for Posture Classification?

[Asafe33]

Hello everyone, I would like to hear your opinions. I have been working with Nexmon since April of this year and have made some progress in capturing Wi-Fi data, but there are still things I don’t fully understand. For instance, what does amplitude mean, and why is there a separate axis just for phase? Additionally, how can I move from understanding 2D graphs to applying spatial phase for the detection and classification of people's posture?

My setup is as follows: an RPI 4B, a TP-Link AC750 router, and a laptop. I read in a study that a researcher with the same setup had only a 1-meter capture radius, while in other works, using antennas, they managed to extend the radius. I would like to advance my understanding and work on HAR classification for further clarification and potential applications. I am using the codes available from @zeroby0 with the following modifications: I applied other filters such as a moving average, band-pass filter, and Hampel, as mentioned in other studies.

The researchers I previously cited used this code to flood the router and extract data. Is there another way to achieve the same? sudo ping -I wlo1 -f -i 0.023 192.168.0.1

[jlinktu]

If you want to reproduce existing work, best ask the authors directly.

[Asafe33]

Has anyone here done this before? I would greatly appreciate any guidance (: