[ntp:questions] Extracting ntpq like information programmatically

A C agcarver+ntp at acarver.net
Sat Mar 30 17:39:36 UTC 2013


On 3/30/2013 09:47, David Woolley wrote:
> Claudio Carbone wrote:
>
>>
>> That depends entirely on the performance of the communication
>> framework we are using.
>> Which is what I'm trying to measure.
>
> That's your basic problem:  you are trying to measure with a mechanism
> that is subject to the parameters you are trying to measure!
>
>>
>> I'm open to suggestions as to how to meter these variables in this
>> situation.
>>
>
> As suggested in another thread, your measurement system really ought to
> be an order of magnitude better than your desired measurement errors,
> which would mean using PPS, either directly, or with a timing GPS
> receiver on each robot.

One way to propagate PPS to the robots would be with an optical link. 
Position a transmitter (IR LEDs, possibly modulated with 50-100kHz 
signal to filter out ambient light noise) on the ceiling of your work 
area.  Fit each robot with a top-mounted detector and ultra-wide angle 
lens (use a PLL detector matched to the transmitter's modulation 
frequency to filter noise).  Each meter of optical path length from the 
transmitter will change the PPS reference offset by about 3 ns.

The PLL detector is very inexpensive.  It's used in every commercial 
electronics product that has an IR remote control (typical modulation 
frequency is near 100kHz) so it's an off-the-shelf item from any 
electronics supplier.  The transmitter just needs to be an accurate 
oscillator that is gated by an accurate PPS source (GPS receiver in the 
lab with roof-mounted antenna for example).


More information about the questions mailing list