[Tinyos-devel] Re: [Tinyos Core WG] time sync interfaces

Philip Levis pal at cs.stanford.edu
Tue Dec 18 14:49:16 PST 2007


On Dec 18, 2007, at 2:30 PM, Omprakash Gnawali wrote:

> On Dec 18, 2007 2:09 PM, David Gay <dgay42 at gmail.com> wrote:
>> On Dec 18, 2007 2:03 PM, John Griessen <john at ecosensory.com> wrote:
>>> David Gay wrote:
>>>
>>>> Surely that depends on what resolution timesync you think is
>>>> worthwhile? My assumption is that different apps have different
>>>> precision requirements, from second-level timesync down to
>>>> microsecond-level timesync. I'm fairly sure that second-level  
>>>> timesync
>>>> doesn't require an adjustable PLL oscillator ;-)
>>>
>>> OK.  Yes, a .1 second resolution HIL module would be a go...
>>> and yes that is more useful than systems out of sync by half a  
>>> second...
>>>
>>> Any chance we have of using the most deterministic delays coming  
>>> from the hardware
>>> we should take it is all I meant.
>>>
>>> What time accuracy do you call a useful timesync on our current  
>>> platforms?
>>
>> Well I was asking that question too, actually ;-) I'm not an
>> application developper/expert, so I can't really answer.
>>
>> David
>> PS: I suspect with reasonable evidence that millisecond level  
>> timesync
>> is practical.
>>
>
> Or you might need more if you look at the applications that use FTSP:
> seismic monitoring by various groups and certain flavor of
> localization.
>

True, but I think the question is what's the 90%. Any HIL you come up  
with will have an app that needs something more or different. The  
hope is the HIL can cover the 90%.

Phil


More information about the Tinyos-devel mailing list