| From: | Jeffrey Mach <mach@****.CALTECH.EDU> |
|---|---|
| Subject: | Re: The Crash |
| Date: | Wed, 4 Mar 1998 15:02:14 -0800 |
On Wed, 4 Mar 1998, Mark L. Neidengard wrote:
> >Not quite. It was a technology to tap into fibre optic lines from
> >outside and read them that was rediscovered, not FO itself.
>
> Okay, that's a _bit_ more like it. Of course, the "technology" to tap an
> ordinary fiber optic line _not_ the sort of thing that can't be rederived
> easily. Make a bend in the wire of the right radius to get Brewster's Angle
> effects on the outer surface of the waveguide, sand off the insulation and
> align with the right optics, and *presto*. Of course, that's the way standard
> fiber works; there have been schemes proposed that use quantum effects that
> would make it _impossible_ to eavesdrop on conversations due to quantum
> effects (the bad guys look at the transmission and the connection simply gets
> broken)...
Mark...you should know better to say "cladding" than "insulation." I
thought they covered that in Phys 2.
Since I am Mr. Science today...for those that don't know how fiber optics
work (possibly including the people at FASA), it is simply a matter of
index of refraction (the speed of light in the medium, which is also a
reflection of how much light bends as it passes into or out of the
medium). The core of the fiber has a higher index of refraction, while
the "cladding" around the core has a lower index of refraction. If you
have ever played with a stick of lucite (plexiglass) you were holding a
HUGE fiber with only air as cladding. When light passes down the core, if
it is not travelling straight down the center, it will eventually reach
the core/cladding boarder, at which point it will either go out of the
fiber (but at a closer angle to the surface than it approached), or it
will bounce back in, just like how looking up from the underneath the
surface of a pool (water's index is greater than air), if you look
straight up, you see out, but at a shallower angle, the water surface
looks like a mirror. And so light nearly parallel to the core travels on
its merry way, always being nudged back into the core by the cladding.
Usually another layer of insulation is added to the cladding to 1) protect
the fiber from damage, 2) make it big enough to actually work with (see
below).
This has certain problems. The big one is that if you send a single pulse
of light in, not all of the photons are aligned with the core, so some
photons travel a straighter path than the others who bounce back and forth
more and therefore end up traveling a longer path. One answer is to have
a cladding gradient, which means that since the light travels faster in a
lower index material, you gently change the index of refraction so that
the beams that travel farther travel faster in the cladding and catch up,
as well as bend straighter and straighter to the core as they pass along
the fiber. Great idea, huh? Problem is, you are dealing with a fiber
less than the width of a human hair, and you want to "gradually" change
the index of refraction? You can buy this kind of fiber for research
purposes, for when you need to fiber optic a lot of light in/out of a
place that it is hard to align optics with (like to look into a running
engine) but it is _really_, _really_ expensive. Next choice is to use a
fiber whose core is on the order of the size of the wavelength of light so
you basically force all the photons to travel only along the line of the
fiber. So called "single mode" fiber has very little loss, and has much
better coherence (little blurring even over transcontinentaly distances),
which is why it is the main telecommunications fiber.
But if you read the last post, you might remember me mentioning a figure
as to the laser wavelenghts being on the order of a 1000nm or one _micron_
(a.k.a. 1000 times smaller than a millimeter). You can imagine how
splicing into a single mode fiber is _not_ something you can do with your
bare hands, and laugh quite loudly if you ever catch an author or movie
trying to tell you so. Nanites, maybe. Their main drawback is that it is
difficult to focus light onto so small of a core. By diffraction alone,
you are limited by Physics as to how small you can focus down your beam,
even if you had "perfect" optics, so you therefore can't get much light
power into such a fiber and degredation due to imperfections in the core
is a worry. Splicing onto one of these things such that you don't
actually damage the core _and_ don't degrade the signal beyond recognition
would be a challenge. Currenly it is a pain in the ass just to get two
single-mode fibers aligned with each other to splice into one fiber, let
alone generate some sort of Y junction, so yes, it is more believable that
this technology could be lost for a time. But probably not on the order
of decades.
By comparison electrical wiring is trivial to tap/monitor, which is why
fiber would seem to be such a nice thing for secure systems.
--Ask Dr. Science
Jeff
