question on preamble

philgo@gmail.com wrote:
> normally, why do we use differential BPSK spread by Barker sequence as
> preambles?
> what do we need to take into account when designing preambles?
> thanks a lot

Hello Phil,

In general preambles allow for either bit or byte or maybe both
synchronization. A Barker sequence has the property of having a
maximally peaked autocorrellation function given it is made up of just
ones and minus ones. A common use of Barker code for spreading is in
80.211.b where the spread data allows one to not need equalization. The
higher data rate versions of 80.211 use complementary code keying which
is a two dimensional form of Barker code. Do you have a particular
application in mind or is your question general?

IHTH,

Clay S. Turner

I am thinking about some non-CDMA systems. SUppose we are using OFDM as
the basic PHY structure.

Can we still use Differential BPSK spreaded by Barker to do AGC, frame
sync, and freq offset estimation?

philgo

Clay wrote:
> philgo@gmail.com wrote:
> > normally, why do we use differential BPSK spread by Barker sequence as
> > preambles?
> > what do we need to take into account when designing preambles?
> > thanks a lot
>
> Hello Phil,
>
> In general preambles allow for either bit or byte or maybe both
> synchronization. A Barker sequence has the property of having a
> maximally peaked autocorrellation function given it is made up of just
> ones and minus ones. A common use of Barker code for spreading is in
> 80.211.b where the spread data allows one to not need equalization. The
> higher data rate versions of 80.211 use complementary code keying which
> is a two dimensional form of Barker code. Do you have a particular
> application in mind or is your question general?
> 
> IHTH,
> 
> Clay S. Turner

philgo wrote:
> I am thinking about some non-CDMA systems. SUppose we are using OFDM as
> the basic PHY structure.
>
> Can we still use Differential BPSK spreaded by Barker to do AGC, frame
> sync, and freq offset estimation?
>
> philgo

Hi Phil,

One characteristic that is nice in a preamble is that the energy across
the signal bandwith does not vary.  This help with channel estimation.
The basis of the questions that you are asking are pretty deep (at
least in my opinion).  Although Barker can be used as a preamble for
OFDM ( I think that this was one idea for IEEE 802.11g for backwards
compatibility) it is probably not the best way of designing the
preamble.  It will really come done to the application and the channel.

I hope that his is useful.

different Phil

philgo@gmail.com wrote:

> normally, why do we use differential BPSK spread by Barker sequence as
> preambles?

Ethernet uses the preamble to sync up the PLL in the receiver.

> what do we need to take into account when designing preambles?

That it be easy for the phase comparator to detect phase differences.

-- glen