Hi David,
I did 2 simple experiments today about that..

1st
I deployed 2 synchronized senders (using STXON without the CCA control, and
exactly the same packet) sending a packet periodically. A receiver in the
nearby was supposed to blink the led on or off when an SFD was detected. The
result was that the receiver was blinking periodically with the same rate of
the senders: this is what I expected. The time difference is too small that
a receiver can still decode the message because of the "gain" given by
superposition; it is a known fact in modulation theory, for very small
delays (assuming they are transmitting the same packet, so same waveform),
otherwise you get a collision.

In order to check the 2 senders were actually sending, and not one of them,
I did the 2nd experiment

2nd
I put the three nodes as follow

receiver <-----------> sender1 <---------------> sender2

where <---------> indicates a communication link. sender1 and sender2 were
synchronized before starting the experiment, and were supposed to send the
same packet periodically. What I saw is that receiver was able to receive
from sender1 (not from sender2 since I was sure  there was no communication,
I deployed the node to get this situation), while it didnt blink if sender1
was turned off.

So I can conclude that if you issue a STXON command the radio switches off
the RX part and turns to the TX mode. Obviously there is some delay before
delivering the packet, due to calibration, oscillator and other hardware
stuff, but my guess is that whenever you issue a STXON strobe command, the
RX part is basically off, and every incoming packet is lost.

Roberto




On Dec 21, 2007 9:50 AM, David Moss <[EMAIL PROTECTED]> wrote:

>  Roberto –
>
>
>
> Just wanted to let you know I'm sorry to say I don't know the complete
> answer to your question.   In the 2.x CC2420 radio stack, I don't think we
> even tried to juggle RX and TX – when we want to transmit, we just tell the
> radio to transmit on CCA.   If there's another transmitter nearby, then the
> transmission isn't attempted (as would be the case when there's a
> transmitter nearby sending to the wrong address). It always seemed to me
> that when you start transmitting without CCA, the radio basically ignores
> any reception that's going on and begins transmitting the packet you're
> telling it to transmit.
>
>
>
> -David
>
>
>
>
>  ------------------------------
>
> *From:* roberto pagliari [mailto:[EMAIL PROTECTED]
> *Sent:* Thursday, December 20, 2007 6:18 PM
> *To:* David Moss; TinyOS ML
> *Subject:* Re: [Tinyos-help] tinyos 2.x radio
>
>
>
> Hi All,
> I've been using tinyos 1.x stack on micaz motes. From CC2420 datasheet,
> when address recognition is enabled, the SFD goes low if address recognition
> fails. In this case, how much time does the radio need  to wait before
> transmitting a packet? I tried something like
>
>       while (TOSH_READ_CC_SFD_PIN()){};  // wait until SFD pin goes low
>       call Leds.yellowToggle();
> //      TOSH_uwait(1000);           // ~ 400 usec delay until SFD goes
> high!!!
>       // here send a message
>
> I inserted a wait but the procedure still does not work. Even assuming the
> radio continues decoding the rest of the message, 1000 should be enough.
> Does anyone have an idea on what's going on?
>
_______________________________________________
Tinyos-help mailing list
[email protected]
https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help

Reply via email to