[Tinyos-help] Hardware Address Recognition

[André Miguel de Almeida Marrão Rodrigues]

Hi

>The base station does not automatically retransmit failed packets.
It means that I should change the code of the standard basestation that 
comes with T2.02?
Includind adding buffer to allow retransmission of packets?

 > interference - it's possible from wifi
> collisions - in this configuration I think they should be very few
> range - that's not an issue in this case
> mismatched duty cycling parameters, etc.  Did you modify your base station
application to configure the outbound packet for delivery to a duty cycling
receiver - yes we did

So if the major problem is interference the acks shoudn't have resolved it?

Thanks,

André Rodrigues


-----Original Message-----
From: André Miguel de Almeida Marrão Rodrigues
[mailto:andremiguelrodrigues at gmail.com]
Sent: Thursday, December 13, 2007 11:20 AM
To: David Moss; 'André Cunha'; 'TinyOS ML'
Subject: Re: [Tinyos-help] Hardware Address Recognition

Hi

I would like to know why in a simple configuration like

BS <---> Node

I have packet loss when the acks are enabled and the code in the node
retransmits
the packed if it get lost. We are using 2% low power listening mode in T2 /
TMote.

Any ideias would be very apreciated.

Regards,

André Rodrigues

PS: If a packet get lost in the BS -> node direction, the 2.02 basestation
code does
the retransmission?


----- Original Message ----- 
From: "David Moss" <dmm at rincon.com>
To: "'André Cunha'" <andreribeirocunha at gmail.com>; "'TinyOS ML'"
<tinyos-help at millennium.berkeley.edu>
Sent: Thursday, December 13, 2007 5:16 PM
Subject: RE: [Tinyos-help] Hardware Address Recognition


> Address recognition is in both hardware and software.  You can disable
> address recognition in hardware to sniff packets, but with software
> acknowledgements enabled you don't want to send an ack back unless you've
> received a packet to your local address or to the broadcast address.
> That's
> why there's a redundant check.
>
> Hardware address recognition is enabled by default, and the redundant
> software address check performed for acknowledgements doesn't change that.
>
> Note at the top of CC2420ControlP the compile time options you have
> available (you should be able to configure these at runtime as well)
>
> #if defined(CC2420_NO_ACKNOWLEDGEMENTS)
>    autoAckEnabled = FALSE;
> #else
>    autoAckEnabled = TRUE;
> #endif
>
> #if defined(CC2420_HW_ACKNOWLEDGEMENTS)
>    hwAutoAckDefault = TRUE;
> #else
>    hwAutoAckDefault = FALSE;
> #endif
>
> // This changes the hardware address recognition.
> #if defined(CC2420_NO_ADDRESS_RECOGNITION)
>    addressRecognition = FALSE;
> #else
>    addressRecognition = TRUE;
> #endif
>
> -David
>
>
>
> -----Original Message-----
> From: tinyos-help-bounces at millennium.berkeley.edu
> [mailto:tinyos-help-bounces at millennium.berkeley.edu] On Behalf Of André
> Cunha
> Sent: Thursday, December 13, 2007 9:21 AM
> To: TinyOS ML
> Subject: [Tinyos-help] Hardware Address Recognition
>
> Hello all,
>    I was looking at the hardware recognition in the CC2420 in the file
> CC2420ControlP and, if fact, the CC2420 register (MDMCTRL0->ADDR_DECODE)
> is enabled. Looking at the CC2420ReceiveP in the RXFIFO.readDone I
> realized that the address verification is by software. Can anyone
> clarify this? I have changed the code of RXFIFO.readDone to only signal
> data reception but I would like to activate the hardware address
> recognition.
>
> Best Regards,
>
> André
> _______________________________________________
> Tinyos-help mailing list
> Tinyos-help at millennium.berkeley.edu
> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
>
>
>
> _______________________________________________
> Tinyos-help mailing list
> Tinyos-help at millennium.berkeley.edu
> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help