ANT Forum

Hi,

If I have a heart rate monitor, a power meter, and a speed sensor
on my bike, and they're all broadcasting data, how does the head
unit "synchronise" this data, in time? When I download a FIT file
from my device it has decided that heart beat X happened "at the
same time" I was at power wattage Y, travelling at speed Z.

The FIT file I looked at had "records" each with a timestamp, one
second apart, and within that record a value for speed, cadence,
heart rate, power.

Looking at the data messages from HRM there is no timing
information beyond a 1/1024th second time of event that rolls
over. Power data is complex, but likewise contains no absolute
timestamp.

ANT+ extended data does have an RX Timestamp though, which the
receiving device could use to assign an absolute time value to a
received message.

My question is: how then does a head unit/receiver manage to
equate many events on different devices to the same moment in
time? Is it basically maintaining an average for each reading
per-second and then lumping those averages into the record for
that second?

Cheers

Russell      

No one?

Having read the specs for each of the sensors, and the FIT file docs I still haven't the first idea how anyone assigns absolute time values to any sensor event. If anyone can help give me a clue, I'd be very grateful.
     

It's up to the head unit. There are good ways of doing it and not so good. It's not easy to tell what the head unit is doing. As you say an average over the last second is a starting point but often the data is not coming in that fast especially at low cadence and there can also be drops in reception. The underlying ANT+ Protocols allow for you to still maintain good totals and thus averages for things like power, cadence, speed and HR beats over fairly bad reception conditions but what the head unit will present for the second by second data that they typically give you is not obvious in these situations.      

Many thanks for the reply.

But most (all?) headunits will allow you to download a fit file. And the fit file as a common specification that they all share. The fit files I've looked at contain second timestamps with per-second data for sensors (HR, Power, Cadence, Speed, GPS coords etc)

Do different headunits then make different fit files, based on how they average data, handle lost reception, calculate totals etc? Or does the fit file specification have requirements for the inputs?

I'm still trying to figure out how events end up as concurrent. If the headunit starts recording at time T and receives a message from a sensor, does it say that the event that caused the message happened at T? If it gets a two messages, one from Power, one from Speed, can it infer that the power increase caused the speed increase…that the events were simultaneous?

Russell