I think this is pretty hand-wavey and requires a lot assumptions to be true. Also: "We collected training and testing data from 10 volunteers." Not a statistic

I think this is pretty hand-wavey and requires a lot assumptions to be true. Also: "We collected training and testing data from 10 volunteers." Not a statistically useful sample set. Under very controlled environments, measuring signal deltas may be possible- but I would like to see sample data that suggests high success rates before I think this is worthy of concern. Finally- Self tuning antennas are a thing. This is going to get harder over time. https://www.qualcomm.com/videos/qualcomm-rf360-dynamic-antenna-matching-tuner-qfe2550 https://www.qualcomm.com/videos/qualcomm-rf360-dynamic-anten...