Cryptography Reference
In-Depth Information
if the radio terminology is translated into digital speak. This is not as
complicated as it looks. Plus, using digital metaphors can offer a few
additional benefits which may be why most spread-spectrum radios
today are entirely digital.
The basic approach in radio lingo is to “spread the energy out
across the spectrum”. That is, place the signal in a number of dif-
ferent frequencies. In some cases, the radios hop from frequency to
frequency very quickly in a system called time sequence. This fre-
quency hopping is very similar to the technique of using a random
number generator to choose the locations where the bits should be
hidden in a camouflaging file. In some cases, the same randomnum-
ber generators developed to help spread spectrum radios hop from
frequency to frequency are used to choose pixels or moments in an
audio file.
Sometimes the systems use different frequencies at the same
time, an approach known as direct sequence. This spreads out the
information over the spectrum by broadcasting some amount of in-
formation at one frequency, some at another, etc. The entiremessage
is reassembled by combining all of the information.
The way the energy is parceled out is usually pretty basic. At
each instance, the energy being broadcast at all of the frequencies
is added up to create the entire signal. This is usually represented
mathematically by an integral. Figure 14.1 shows two hypothetical
distributions. The top integrates to a positive value, say 100
.
03 ,and
the bottom integrates to 80
2 . Both functions look quite similar. They
have the same number of bumps and the same zero values along the
x-axis. The top version, however, has a bit more “energy” above the x-
axis than the other one. When all of this is added up, the top function
is sending a message of “ 100
.
03
spread-spectrum radio signals like this are said to be resistant
to noise and other interference caused by radio jammers. Random
noise along the different frequencies may distort the signal, but the
changes are likely to cancel out. Radio engineers have sophisticated
models of the type of noise corrupting the radio spectrum and they
use the models to tune the spread-spectrum algorithms. Noise may
increase the signal at one frequency, but it is likely to decrease it
somewhere else. When everything is added together in the integral
the same result comes out. Figure 14.2 shows the same signals from
Figure 14.1 after a bit of noise corrupts them.
A radio jammer trying to block the signal faces a difficult chal-
lenge. Pumping out random noise at one frequency may disrupt a
signal concentrated at that single frequency, but it only obscures one
small part of the signal spread out over a number of frequencies. The
.
Search WWH ::




Custom Search