[ntp:questions] Sub-millisecond NTP synchronization for local network

Jon Kåre Hellan dev at null.com
Mon Dec 8 11:10:44 UTC 2008


leibs at willowgarage.com (Jeremy Leibs) writes:

> Our configuration is 4 machines connected on a local gigabit network located
> on a mobile robotic base.  These machines are subject to frequently being
> powered down or restarted.  In order to use the robot, the clocks on these
> machines must be self-synchronized to less than 1 millisecond.  Ping times
> between machines on this local network vary between 100 microseconds, and
> 1ms depending on saturation of the network by sensor data streams.
>
> The 4 machines are connected to the rest of the world through a wireless
> link.  The delay time on the wireless link is much more variable: in the
> range of 2ms to 300ms depending on the quality of the link and the amount of
> data going over the wire.  We don't care nearly as much about
> synchronization between the robot and the outside world, though it would be
> nice to avoid unbounded drift.  A synchronization in the range of 10's of ms
> would be acceptable.

It should be possible to solve this with any kind of external signals
which all units can detect. It doesn't sound like lab quality time
accuracy is needed, as long as all units agree within the required
tolerance. I've already suggested wired PPS, but it could be a radio
impulse generated within the lab, or even sound!  (Provided it doesn't
drive the workers crazy, and the units don't move around. Propagation
delay would change too much if they did.)  Are there short range radio
reference clocks which operate within unlicensed spectrum, or within
spectrum that could be licensed at reasonable cost? You may also try
staying with NTP and WLAN, but reserve certain channels for
timing. For instance, if you aren't using 5 GHz 802.11a at present,
you could use that for timing.

Jon Kåre Hellan, UNINETT, Trondheim, Norway.




More information about the questions mailing list