Since we've installed more v4.1 sensors the number of spikes (peaks caused by low readings) has massively increased. Readings <22cm are below the minimum distance and should not be transmitted. Instead, consider an error message.
Debugging would also be useful to see
- if all the reads on a particular wakeup period are < 22cm
- the potential causes of this. Why does it fail to read?
Also consider modifying Nominet bridge code to reject or annotate errors.