A question came up in the poster session regarding how reliable the connectivity of sensors was versus the actual measurement. Another metric we should determine for our network is how reliable the sensors are in terms of staying connected to the Internet.
Assuming the UT IoT network remains online (need to check this somehow), we should be able to see if each sensor is reporting values at a somewhat regular interval. A few numbers we should know:
percentage of time the sensors have been online (aggregate)
percentage of time an individual sensor has been online (which PAs are behaving poorly
the average time that sensors go down for i.e. how long until they reconnect
Other things we should know that aren't necessarily numeric:
Do we see sensors going offline more often as they continue running?
Are there any triggers that we can identify that cause our sensors to go offline?
Were there times when all the sensors went offline? (IoT went down, firmware updates, etc.)
Reliability of the BOMP network
A question came up in the poster session regarding how reliable the connectivity of sensors was versus the actual measurement. Another metric we should determine for our network is how reliable the sensors are in terms of staying connected to the Internet.
Assuming the UT IoT network remains online (need to check this somehow), we should be able to see if each sensor is reporting values at a somewhat regular interval. A few numbers we should know:
Other things we should know that aren't necessarily numeric: