181 lines
8.9 KiB
Plaintext
181 lines
8.9 KiB
Plaintext
______________________________________________________________________________
|
|
| File Name : TIMEREV1.ASC | Online Date : 11/20/94 |
|
|
| Contributed by : Jerry Decker | Dir Category : KEELY |
|
|
| From : KeelyNet BBS | DataLine : (214) 324-3501 |
|
|
| KeelyNet * PO BOX 870716 * Mesquite, Texas * USA * 75187 |
|
|
| A FREE Alternative Sciences BBS sponsored by Vanguard Sciences |
|
|
|----------------------------------------------------------------------------|
|
|
The following is from the July 22, 1994 issue of Science, Volume 265.
|
|
------------------------------------------------------------------------------
|
|
Time-Reversed Sound Waves Resonate among Physicists
|
|
by James Glanz
|
|
|
|
Carelessly chosen words often prompt people to wish they could take back what
|
|
they have just said. Now, thanks to a team of French acoustics researchers,
|
|
that may no longer be an empty wish: They have developed a device that can
|
|
take a sound, flip it around, and send it back along its original path in
|
|
time-reversed form - almost as if time were going backwards.
|
|
|
|
Team leader Mathias Fink, director of the Waves and Acoustics Laboratory at
|
|
the University of Paris, faces the device and says "bonjour" to it. An
|
|
instant later, the phonemic reversal "ruojnob" arrives back at his mouth
|
|
INAUDIBLE to anybody else - the reconstruction is crisp, despite strong echoes
|
|
in the room.
|
|
|
|
But the experiment is not as frivolous as it sounds. The French team has
|
|
already applied it to medicine and materials analysis. They have "made heavy
|
|
inroads in medical ultrasonic imaging," says John Gilmore, a physicist at
|
|
General Electric's Research Center in Schenectady, New York, who has seen
|
|
Fink's setup. Other researchers are now beginning to apply it to undersea
|
|
communications.
|
|
|
|
The key to this trick is the acoustic wave equation, which is no respecter of
|
|
time. The equation looks the same whether time is moving backward or forward,
|
|
and this means that for every burst of sound or ultrasound DIVERGING FROM a
|
|
source - and possibly being reflected, and fragmented by multiple barriers and
|
|
propagation media - there exists in theory a set of waves that PRECISELY
|
|
RETRACES all of these complex paths and CONVERGES, in SYNCHRONY, at the
|
|
original source, as if time were going backwards.
|
|
|
|
It is not a simple task, however, to take a sound wave and produce its time-
|
|
reversed twin. Following similar work with microwaves and light, acousticians
|
|
have succeeded in solving this problem in limited cases using phase-conjugate
|
|
mirrors (PCMs), materials that can spontaneously reverse an approximately
|
|
continous and monotone signal of any spatial shape. But acoustic systems
|
|
usually have to deal with pulsed mixtures of many tones, and in such cases,
|
|
says Fink, a simple PCM "isn't sufficient to bring the waves back to a good
|
|
focus." Instead, he says, "you have to completely time-reverse the signal."
|
|
|
|
Fink realized in the late 1970's that a computer and an array of transducers
|
|
might be able to work fast enough to store digitized versions of the peaks and
|
|
troughs of incoming acoustic waves, process them, and then, shoot them out
|
|
backward. Unlike a passive reflector, such a device should also be able to
|
|
convert a diverging wave into a converging one. At the time, however, the
|
|
cost of the necessary memory and analog-to-digital converters was "crazy,"
|
|
Fink recalls.
|
|
|
|
But by 1990, prices had dropped sufficiently and computers had gotten faster.
|
|
Fink and his collaborators in Paris have now built several devices, the first
|
|
time-reversal mirrors (TRMs) for acoustics, which rely on fast computer
|
|
processing and a novel class of materials called piezocomposites arranged in
|
|
an array. The piezocomposite transducers turned out to be a key element. A
|
|
normal piezoelectric material will produce an electrical signal when an
|
|
acoustic wave passes through it, and conversely will emit sound if excited
|
|
with an electrical signal.
|
|
|
|
"On paper, it isn't difficult to make one channel," says Fink. The problems
|
|
start when you put the crystals in an array, because the signals coming and
|
|
going from one transducer will interfere with its neighbors. Fink's team got
|
|
around this difficulty by using piezocomposites: rodlike piezoelectric
|
|
elements embedded in a polymer matrix. These are engineered to respond in one
|
|
direction only, along the axis of the rods, and so do not affect nearby
|
|
transducers.
|
|
|
|
Once these technical problems were overcome, Fink and his team were quick to
|
|
show the practical applications of their device. Two years ago they began to
|
|
tackle the problem of medical imaging, scattering ultrasonic pulses off a
|
|
kidney stone in order to track it - through layers of fat and connective
|
|
tissue - in real time while a patient breathes.
|
|
|
|
The team starts by sending in an ultrasonic seed pulse, some of which scatters
|
|
off the stone and some off small, random inhomogeneities in its surroundings.
|
|
The TRM initially sees a reflected signal from the stone buried in the noise.
|
|
The entire scattered signal is reversed by the TRM, rerouted back through the
|
|
body, scattered again, gets reversed again, and so on.
|
|
|
|
During these iterations, a pulse ping-pongs steadily between the stone and the
|
|
mirror - but signals from fine-scale features of the tissue are occasionally
|
|
missed by the TRM on its return shots, and these scatterers gradually lose all
|
|
their "ping-pong balls" of ultrasound, leaving only the stone's clear signal.
|
|
|
|
Once the stone has been reliably located, intermittent amplified pulses can be
|
|
applied to shatter it. As the stone moves, the process is repeated to locate
|
|
it again and again. This location method, which has been tried successfully
|
|
on two patients in France without the final treatment, should improve on the
|
|
current method of steering the ultrasound using x-ray measurements.
|
|
|
|
Similar location methods underlie Fink's approach to an important engineering
|
|
task - finding defects in titanium alloys, whose randomly oriented,
|
|
crystalline grains also create noise. Such a defect was blamed for the crash
|
|
of a DC-10 airliner in Sioux City, Iowa, in 1989. The TRM could find defects
|
|
"as small as half the size of what we can currently detect," a critical
|
|
difference for safety, says Gilmore.
|
|
|
|
Although Jan Achenbach, director of the Center for Quality Engineering and
|
|
Failure Prevention at Northwestern University in Evanston, Illinois, thinks
|
|
that TRMs may still be too expensive for this application - he has his own
|
|
array to steer the beam without actually time-reversing it - he agrees there
|
|
is a "tremendous simplification in doing the scanning ELECTRONICALLY" as
|
|
opposed to the current, mechanical methods.
|
|
|
|
Potential uses of TRM technology are not just up in the air. Independently,
|
|
researchers at the University of Washington in Seattle have proposed building
|
|
an underwater TRM, consisting of an array of hydrophones, to overcome the
|
|
problem of "multipath distortion" in underwater communications. The problem
|
|
arises because acoustic transmissions bounce off the ocean's surface and
|
|
floor, and bend in temperature and pressure gradients, so that "the pulse
|
|
comes staggering in several times," says Darrell Jackson of the University of
|
|
Washington.
|
|
|
|
Along with colleague James Ritcey, Jackson proposes a solution based on time-
|
|
reversal with a twist: One submarine sends out a "probe" pulse, and multiple
|
|
copies of it arrive at another sub's TRM. This turns all the signals around
|
|
and sends them back, but this time with digital information encoded into them.
|
|
All of these encoded signals should then arrive back at the original sub
|
|
simultaneously, so that one path can't confuse the data stream arriving along
|
|
another.
|
|
|
|
For all the practical uses of TRMs, Fink is not neglecting basic science. He
|
|
has recently put together experiments to test the limits of acoustic time-
|
|
reversal. He is totally scrambling sound pulses by passing them through a
|
|
"forest" of thousands of scatterers, then reassembling the pulses with a TRM
|
|
and sending them back through the forest to see how well the original signal
|
|
survives. So far, none of his pulses has lost its reversibility in the
|
|
forest. But Fink has still more severe tests planned for his acoustic
|
|
pilgrims. If those pan out, growing numbers of physicists are likely to be
|
|
saying "bonjour" to this technology.
|
|
------------------------------------------------------------------------------
|
|
Vanguard Note
|
|
|
|
There are consumer devices available from two sources for personal noise
|
|
cancellation devices. I found them in an airline magazine but don't have
|
|
the information at hand for this file. Next time you fly, check out the
|
|
advertisements.
|
|
|
|
Also, the seed technology companies for this in the US are listed in the
|
|
file NOISECNC.ASC on KeelyNet.
|
|
------------------------------------------------------------------------------
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|