Graphics Programs Reference
In-Depth Information
0x712
One-Time Pads
One example of an unconditionally secure cryptosystem is the one-time pad .
A one-time pad is a very simple cryptosystem that uses blocks of random data
called pads . The pad must be at least as long as the plaintext message that is
to be encoded, and the random data on the pad must be truly random, in
the most literal sense of the word. Two identical pads are made: one for the
recipient and one for the sender. To encode a message, the sender simply
XORs each bit of the plaintext message with the corresponding bit of the
pad. After the message is encoded, the pad is destroyed to ensure that it is
only used once. Then the encrypted message can be sent to the recipient with-
out fear of cryptanalysis, since the encrypted message cannot be broken
without the pad. When the recipient receives the encrypted message, he also
XORs each bit of the encrypted message with the corresponding bit of his
pad to produce the original plaintext message.
While the one-time pad is theoretically impossible to break, in reality it's
not really all that practical to use. The security of the one-time pad hinges
on the security of the pads. When the pads are distributed to the recipient
and the sender, it is assumed that the pad transmission channel is secure.
To be truly secure, this could involve a face-to-face meeting and exchange,
but for convenience, the pad transmission may be facilitated via yet another
cipher. The price of this convenience is that the entire system is now only
as strong as the weakest link, which would be the cipher used to transmit
the pads. Since the pad consists of random data of the same length as the
plaintext message, and since the security of the whole system is only as
good as the security of pad transmission, it usually makes more sense to just
send the plaintext message encoded using the same cipher that would have
been used to transmit the pad.
0x713
Quantum Key Distribution
The advent of quantum computation brings many interesting things to the
field of cryptology. One of these is a practical implementation of the one-
time pad, made possible by quantum key distribution. The mystery of quantum
entanglement can provide a reliable and secret method of sending a random
string of bits that can be used as a key. This is done using nonorthogonal
quantum states in photons.
Without going into too much detail, the polarization of a photon is the
oscillation direction of its electric field, which in this case can be along the
horizontal, vertical, or one of the two diagonals. Nonorthogonal simply means
the states are separated by an angle that isn't 90 degrees. Curiously enough,
it's impossible to determine with certainty which of these four polarizations a
single photon has. The rectilinear basis of the horizontal and vertical polariza-
tions is incompatible with the diagonal basis of the two diagonal polarizations,
so, due to the Heisenberg uncertainty principle, these two sets of polarizations
cannot both be measured. Filters can be used to measure the polarizations—
one for the rectilinear basis and one for the diagonal basis. When a photon
passes through the correct filter, its polarization won't change, but if it passes
Search WWH ::




Custom Search