About sensor's TIMESTAMP

[Mostima]

As we know we could get the timestamp when we are getting the sensor data.

Is there anyone could tell me what exactly this timestamp means? Are they the time of getting the sensor data? Or they are the timestamp of reaching fitbit studio.

If they are the later, I might need some works to adjust these data.

I will appreciate it if you could provide me an answer!

[Gondwana]

The timestamps are arbitrary: see here. If you need to know an actual time, you'll need to extract that separately, at least once (so you know what the timestamp offset is).

[Mostima]

Thank you for your answer! But sorry I didn't get your point.

What you mean by "arbitrary"? Do you mean that the timestamp generated in each time is a random timestamp rather than a real world timestamp?

Sorry if I didn't make my question clear, I just want to know whether the timestamp will gain a delay while transferring data to fitbit studio?

[Gondwana]

I think the timestamps are okay for comparing with one another (so, eg, you can determine the period between successive readings). But without extra information, you can't know what time of day they correspond to.

[mauro82]

Hi, I have a similar question. In fact, I received two files for each person using the fitbit. For each session, one file contains the data of the accelerometer and the timestamp, while the other file contains data of the heart rate with the timestamp.

Unfortunately data are not sorted because they were sent from the fitbit to a tablet and then from the tablet to a server.

My questions are:

- if I sort the rows of the two files (associated to a session), will I be able to reconstruct the exact sequence?

- will the difference of two consecutive timestamps represent the time in milliseconds between those instances?

- Data of the accelerometer and data of the heart rate do not correspond. In fact, the sampling frequency for the accelerometer is 50 Hz, while the sampling frequency for the heart rate is 1 Hz. So there are more rows containing data from the accelerometer than those containing data associated to the heart rate. I need to do data analytics on these data and merge them into a single table, so my idea is to replicate data of the heart rate for all those data for the accelerometer which timestamps are between two closest timestamps in the data of the heart rate. Is this comparison safe or does the fitbit use two random numbers for the timestamps, i.e., one for the accelerometer and one for the heart rate?

Thank you so much for your help and best regards.