|
|
 | | From: | porterboy | | Subject: | A small doubt about signal constellation | | Date: | 21 Jan 2005 05:49:08 -0800 |
|
|
 | I think MLT-3, a basedband line code used in Ethernet, has a zero
constellation point. But generally, a constellation point on the
origin would have zero amplitude and ambiguous phase, which would
reduce the receiver's ability to track smapling/carrier
phase/frequency (which is usually driven by amplitude statistics or
zero crossings). In fact some modulations are specifically designed
not to cross the origin at all, ever, not just at the sampling
instant. Offset-QPSK is one such... the I and Q modulations are
staggered by half a sampling interval, so that the eye-diagram is
always open...
O____O O____O
| | |\ /|
| | | \/ |
| | | /\ |
| | |/ \|
O----O O----O
Ofset-QPSK QPSK
|
|
| | From: | Randy Yates | | Subject: | Re: A small doubt about signal constellation | | Date: | 21 Jan 2005 10:04:17 -0500 |
|
|
| porterboy76@yahoo.com (porterboy) writes:
> I think MLT-3, a basedband line code used in Ethernet, has a zero
> constellation point. But generally, a constellation point on the
> origin would have zero amplitude and ambiguous phase, which would
> reduce the receiver's ability to track smapling/carrier
> phase/frequency (which is usually driven by amplitude statistics or
> zero crossings). In fact some modulations are specifically designed
> not to cross the origin at all, ever, not just at the sampling
> instant. Offset-QPSK is one such... the I and Q modulations are
> staggered by half a sampling interval, so that the eye-diagram is
> always open...
>
> O____O O____O
> | | |\ /|
> | | | \/ |
> | | | /\ |
> | | |/ \|
> O----O O----O
>
> Ofset-QPSK QPSK
Nice ascii diagrams!
The only reason I know of to not allow a transition through the origin
is directly related to practical issue of RF power amplifier (PA)
design. A power amp can be made more efficiently if it can be
non-linear, and constellations which require a transition through the
origin require a more linear PA.
--
Randy Yates
Sony Ericsson Mobile Communications
Research Triangle Park, NC, USA
randy.yates@sonyericsson.com, 919-472-1124
|
|
|
