> -----Original Message-----
> From: Dave Tweed [SMTP:pic@DTWEED.COM]
> Sent: Wednesday, July 10, 2002 1:12 AM
> To:   PICLIST@MITVMA.MIT.EDU
> Subject:      Re: [OT]: About smart dust and more
>
> On Tue, 9 Jul 2002, Peter L. Peres wrote:
> > In theory, about c/(2*lambda) Hz (== Bauds in this case). In practice,
> as
> > much as you can drive the device to. Diodes for high data rates have
> > special design (like controlled impedance feeds and such). I once had a
> > book that showed how to interface a laser diode to a ham RF transmitter
> > for 432MHz. It consisted of an impedance matching network and a fast
> diode
> > or two, plus a very simple bias network that used a D cell I think. The
> Tx
> > ran 10-20W into the air cooled diode.
>
> and later added:
> > The actual limit is lower than I implied, the modulation is deteriorated
> > by the decay time of the intermediary laser levels (the non-lasing
> > transitions). However frequency-modulated lasers apparently do not have
> > this limit ?
>
> I thought really high data rates used external (electro-acoustic?)
> modulators.
>
They do mostly.  You can achive 10Gbps through a direct modulated laser, but
chirp really limits your link length due to dispersion.  For lower
dispersion, EA (Electro-Absorbtion) and MZ (Mach-Zender) modulators are used
with a CW laser source.  At 40Gbps it's pretty much mandatory have have an
external modulator.  Interestingly both of these external modulators can be
configured to give negative chirp, so that performance over a long length of
fibre can actualy be better than a short length.

Frequency modulated lasers have been under a good deal of research.  There
are several benefits, but one big negative which is the increased complexity
of the receiver.

Regards

Mike

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu