[Tinyos-help] Delta application -help required
Michael Schippling
schip at santafe.edu
Wed Aug 30 12:56:02 PDT 2006
As I said, based on my (small) experience with micaz I believe
that 100 messages per second is about the max you can expect.
And that 100 is for the total system, not per mote (unless of
course you have only one mote transmitting). If you are doing
multihop then you need to divide the total messages by the
number of hops (each hop being it's own message) to get even
a best case estimate -- ignoring hidden nodes and any other
actual practical considerations.
So the only way I see to get your higher sample rate is to
aggregate samples, say 10 per message. You would need to
do some testing in various configurations to see how many
messages you can get...let's say, the source and two hops gives
you best case 30 msg/sec, with 10 samples/msg, is 300 samples/sec.
Nobody said that motes were a High Speed sensor network...
MS
siva prasad wrote:
> thanks for ur reply michael,
>
> if i sample the signal at 1ms(suppose for 50hz signal) then it will be
> better ...that is required for my application ...one more thing is that,
> the system tht i want to monitor is highly dynamic in nature..so the
> data should be send for every 1ms...i think 1ms is sufficient for just
> measuring the external ADC value..and sending this value using
> multihop(must) technique...
> But the problem is..... i am losing lot of packets abt 50% even for the
> sampling period at 10 ms....do u hav any different idea tht ...how can i
> overcome this problem....
> any help should be highly appreciated...
>
> if we cant sample and send(using multihop) the data in 1ms ...pls give
> the reason...
>
> still i did nt clear my old doubts...
>
>
> On 8/30/06, *Michael Schippling* <schip at santafe.edu
> <mailto:schip at santafe.edu>> wrote:
>
> I don't know the message rate you can expect from moteiv motes
> but if you are trying to send one sample every 10ms you are
> probably pretty close to the limit (somewhere on the order of
> 100 msg/sec is the max for micaz's). You'll probably need to
> collect multiple samples in one message to get anywhere at 30Hz...
>
> Let me know if you need clarification one answers to your other
> questions.
> MS
>
>
> siva prasad wrote:
> >
> >
> > From: *siva prasad* <sivakatru at gmail.com
> <mailto:sivakatru at gmail.com> <mailto:sivakatru at gmail.com
> <mailto:sivakatru at gmail.com>>>
> > Date: Aug 28, 2006 12:52 AM
> > Subject: Delta application -help required
> > To: tinyos-help at millennium.berkeley.edu
> <mailto:tinyos-help at millennium.berkeley.edu>
> > <mailto: tinyos-help at millennium.berkeley.edu
> <mailto:tinyos-help at millennium.berkeley.edu>>
> >
> > hi all...
> >
> > i am working on sampling the ADC value instead of
> DEmosensorc...in the
> > delta application provided by moteiv.....
> >
> > i am glad to say tht ..i hav successfully compiled it ...
> > and installed..it worked...
> > the problem is when i tried to undrstand the program..i am facing
> some
> > difficulties....
> >
> > -> i could nt understand the task..named...sendDone( )
> >
> > -> while i tried to see the raw data using listen tool...it is
> showing
> > 42 bytes...but i could nt figured out the data format..
> >
> > ->one more problem is...the signal tht i want to sample is a low
> > frequency signal..abt 30Hz...
> > if i increase the sampling time to 10ms ..i am losing lots of
> packets...
> >
> > can any body pls help me...
> >
> > thanks
> >
> > siva katru
> >
> >
> >
> ------------------------------------------------------------------------
> >
> > _______________________________________________
> > Tinyos-help mailing list
> > Tinyos-help at Millennium.Berkeley.EDU
> <mailto:Tinyos-help at Millennium.Berkeley.EDU>
> >
> https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
>
>
More information about the Tinyos-help
mailing list