Re: [bluetooth] Re: Clock Recovery in BT

["Ling Su" <lingsu@palmmicro.com>]

> Ling Su wrote:
>
> > The threshold should be 6 if we set it statically, since the encoding
method
> > of Access Code give 6 coding distance. Of course, I didn't see some
paper
> > talking about dynamic change the threshold.
> >
>
> >From where did you get this number? it is not mentioned in the BT spec.
> Do you mean that when 6 bits are matched the correlator should trigger hte
start of frame?

No, if the code distance is 6, then when (68-6) bits matched, it means a
successful correlation.

>
> >
> >  Usually a wireless system has a preample
> > like 10 or more bits, which gives more time to do clock recovery, but in
> > bluetooth, there is only 4 bits, so it will be more chanllenging.
> >
>
> Should we operate teh correlator at double the BT data rate to solve such
problem?
>

No enough, we need operate at more than 11Mhz to recovery the clock in the
receiver end. I am not sure if you understand what I mean, the correlator is
working in Frame Level, and the clock recovery working at symbol level, the
correlate will using the clock from "clock recovery" module as its operation
clock to do the correlation to make frame level synchronization.

About current implementation of correlate, it is to slow if we do sequential
summation, we need impalement a parallel adder to make it faster. At least
can be finish within 6 system clock.

Regards,
-Ling

--
To unsubscribe from bluetooth mailing list please visit http://www.opencores.org/mailinglists.shtml