[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
update.293 (fwd)
Forwarded message:
>From [email protected] Wed Oct 30 17:36:20 1996
Date: Wed, 30 Oct 96 13:58:39 EST
From: [email protected] (AIP listserver)
Message-Id: <[email protected]>
To: [email protected]
Subject: update.293
PHYSICS NEWS UPDATE
The American Institute of Physics Bulletin of Physics News
Number 293 October 30, 1996
by Phillip F. Schewe and Ben Stein
SIGNAL TRANSMISSION THROUGH A MAMMALIAN
NERVE-CELL NETWORK can be enhanced with the help of
electrical noise, a new experiment has shown. First proposed to
explain the periodicity of ancient ice ages (in which the random
"noise" of climate variations may have augmented the effects of
predictable Earth-Sun distance variations from year to year) and first
experimentally demonstrated in lasers (in which the direction of laser
light traveling around a loop was switched from clockwise to
counterclockwise by adding acoustical noise to the crystal from
which the light emerged), the phenomenon of "stochastic resonance"
(SR) describes how introducing a certain amount of noise into a
system can actually enhance the transmission or detection of a weak
signal so as to maximize the ratio of signal to noise. In the first
demonstration of SR in mammalian tissue, researchers (Mark Spano,
Naval Surface Warfare Center, 301-227-4466) apply a weak electric
signal (containing both signal and noise) to a slice of rat
hippocampus, a brain region essential for memory and other tasks.
With the slice parallel to the plane in which nerve cells convert
incoming signals into electrical nerve impulses, the researchers
could transmit a weak signal to all nerve cells in the network. At an
optimal noise intensity a maximum in the signal-to-noise ratio was
reached--a hallmark of SR. This experiment offers the intriguing
possibility that SR may potentially be exploited to aid transmission,
detection and processing of signals in neuronal networks. (B.J.
Gluckman et al., Physical Review Letters, 4 November 1996.)
A QUANTUM COMPUTER COULD TOLERATE ERRORS while
carrying out calculations, researchers at Los Alamos have now
shown. Computers that operated according to the rules of quantum
mechanics have the potential to perform powerful tasks (such as
factoring huge numbers) because of their radically different
approach to logic: unlike a conventional computer's bits, which exist
either as a 0 or a 1, a quantum bit (or "qubit") could not only exist
simultaneously as a 0 and a 1 but could interact with other qubits so
that its properties became "entangled" with those of the other qubits.
Yet some physicists argue quantum computers may be impossible to
achieve on a practical level because the slightest amount of noise
would destroy the entanglement and thus corrupt the state of the
qubits. Up to now, proposed "quantum error correction" schemes
have shown merely how to preserve the state of qubits. Now, Los
Alamos researchers (Raymond Laflamme, 505-665-3394) have
developed an algorithm for carrying out reliable calculations on a
"qubyte" made of 7 entangled qubits while accounting for the
possibility that one of the qubits is corrupt (upcoming paper in Phys
Rev Lett). Experimentally, quantum computing is regarded as a
long-term possibility: although quantum versions of logic gates have
been constructed (see Update #250), researchers are still working to
entangle more than two quantum systems at a time.
A PHOTOELECTROCHROMIC (PEC) cell harnesses a
photochromic layer, which changes color by absorbing light, with an
electrochromic layer, which changes color under the influence of an
electric field, to make a self-powered smart window. On a sunny
day the one layer supplies the photovoltage needed by the other layer
to darken the window, letting in less light and thus lowering air-
conditioning costs. This photoelectrochromic process, developed by
scientists at the National Renewable Energy Lab in Golden,
Colorado, might also be useful in display applications. (C.
Bechinger et al., Nature, 17 October 1996.)