Information Technology Reference
In-Depth Information
Vcc
a
b
6
5
4
3
2
1
0
-1
Vin
Vout
0
10
20
30
40
50
time, usec
6
c
4
2
0
-2
0
10
20
30
40
50
time, usec
Fig. 7.1 Scheme of a semiconductor signal inverter (a), dependence of the voltage at its output on
the voltage at its input (b), and scheme of the formation of an inverter using amorphous computing
methods (c)
shown in Fig.
7.1
. The signal at the input of such a device removes the signal at its
output, and conversely the absence of the input signal causes a signal at the output
to appear (Fig.
7.1
). Consider the formation of such a device in an amorphous
system.
The programming language of amorphous systems was developed by Daniel
Coore and was named GPL—Growing Point Language. This language is applicable
to particles to which certain properties are attributed. It is assumed that each particle
possesses relatively limited logical capabilities and a small memory. Particles
operate asynchronously. But at the same time their information capacity is approx-
imately the same as they are created using the same technology. All particles are
programmed the same way. But each of them remembers its state and can generate a
sequence of random numbers. In general, particles do not memorize their location
and orientation. Each particle can pass information to a certain number of its
neighbors; this can be accomplished by radio signals or chemical factors. It is
assumed that there is some communication radius r, which is considerably larger
than the particle size and, at the same time, much smaller than the spatial dimen-
sions of the amorphous medium.
The main concept of GPL is growing points. Suppose there are a large number of
particles with identical properties. Let us introduce into the population of particles,
at random or according to some rule, a particle with properties different from the
environment. This particle—a growing point—initiates a wave of switching states
