[Tinyos-devel] timestamps in seriallisten/sflisten etc.

[Omprakash Gnawali]

On Thu, Jun 26, 2008 at 8:00 AM, Philip Levis <pal at cs.stanford.edu> wrote:
>
> On Jun 26, 2008, at 7:34 AM, Razvan Musaloiu-E. wrote:
>
> >>
> >>
> >> Does each packet source add a timestamp? If not, where does the
> >> timestamp originate? Say I have a chain of
> >>
> >> mote -> SF-a -> SF-b -> SF-c -> client
> >>
> >> What timestamp will the client see? An SF can use any packet source,
> >> after all. Or is an SF packet source a special case where you do not
> >> add a timestamp?
> >
> > I think the timestamp the client is interested in is the one from
> > SF-a, the SF that is the closest to the mote. That's what he could
> > have computed himself if he was directly connected to the mote.
>
> So how does this work?
>
> A packet will need to have a timestamp field. Does it have a chain of
> timestamps from each hop, or a single one? Each packet source would
> need to be able to overwrite(single)/add(chain) a timestamp. If the
> former, probably the best way would be to have an option on packet
> sources as to whether they overwrite the timestamp. I.e., in the case
> above, SF-a's packet source of serial@ has the overwrite option set.
>
> This would involve changing the serial packet formats and logic in
> all of the current support codes (C, C++, Python, Java) and in
> TinyOS, and incrementing the protocol identifiers to detect
> mismatches. Any user code which has been written to read serial
> packets will need to be updated.
>

Are timestamps hard to implement or not necessary? My observation is
almost everyone has had to devise some kind of timestamping scheme. So
the hope was to have this discussion to come up with a general enough
solution that worked for most people.

- om_p