To assess whether unusual detections might just be noise, it would help to show such detections
in the context of all other detections on the same receiver in a surrounding period of time, e.g.
one hour or one day.
This would be a receiver status plot, but with these changes:
restrict time axis to relevant period
anonymize tags from other projects; label them by mfg ID only, as this can help disentangle
cases of aliasing. In summary dataset, they can be ID.1, ID.2, ...
To assess whether unusual detections might just be noise, it would help to show such detections in the context of all other detections on the same receiver in a surrounding period of time, e.g. one hour or one day.
This would be a receiver status plot, but with these changes: