[Tinyos-help] Corr value

[Rodolfo de Paz Alberola]

On Fri, 2008-03-28 at 14:14 -0700, Philip Levis wrote:
> On Mar 26, 2008, at 7:27 AM, Rodolfo de Paz Alberola wrote:
> 
> > Hi all,
> >
> > I am emulating cc2420 radio chip in avrora and testing it with
> > tinyos-2.x. Right now, I am working with LQI and RSSI values.
> >
> > However, I don't have idea of how to simulate the correlation value.  
> > In
> > the data sheet says that this value can be seen like a "chip error
> > rate". So, I was thinking about using the S/N and then the BER to
> > calculate that, but, I don't understand why it ranges between 50 and
> > 110.
> >
> > Do you have any idea of how to model that?
> >
> > Any suggestion will be really appreciated.
> 
> There are some posts on this topic, which include some comments from  
> TI (was ChipCon then). Contacting them directly might be the best way  
> to figure this out.
> 
> Phil

Thanks for your reply, I have found a post where there is a explanation
from Oyvind Janbu. However, I have still some questions:

My understanding is that the correlation value in the rxfifo is the sum
of the 8 first correlation values for the symbols following SFD. But how
are these values calculated? Each of this correlation values should
range between 0-16 to have as a result a 7 bit uint value (16*8 = 128
values), so I still don't understand that. In the explanation is also
said that the values don't go to 127 due to bandwidth limitations.

I tried to contact TI by the webpage and I didn't have any reply.

Could it be possible to get any TI-cc2420 contact mail? 

Thanks for your help,

-------
Rodolfo