Karen Hunter

SCI 510: Quantum

Quantum Cryptography

Spies, communication, and secret codes! Cryptography is the art of encoding and
decoding messages and has existed as long as people have distrusted each other and sought
forms of secure communication. The purpose of cryptography is to transmit information such
that only the intended recipient receives it. Although the field of cryptography is ancient, it is
not static. Cryptographic techniques have evolved over the centuries, with the code-makers
working to stay ahead of the code-breakers. The next major step in this evolutionary process
may be at hand. Today’s most common encryption methods are threatened by the potential
creation of the quantum computer. But already quantum cryptography has been developed
which promises more secure communication than any existing technique and cannot be
compromised by quantum computers.
Quantum cryptography takes advantage of the unique and unusual behavior of
microscopic objects to enable users to securely develop secret keys as well as to detect
eavesdropping. Although work on quantum cryptography was begun by Stephen J. Wiesner in
the late 1960’s, the first protocol for sending a private key using quantum techniques was not
published until 1984 by Bennett and Brassard.
The development of quantum cryptography was motivated by the short-comings of
classical cryptographic methods, which can be classified as either “public-key” or “secret-key”
methods.
Public-key encryption is based on the idea of a safe with two keys: a public key to lock
the safe and a private key to open it (Ford, 2002; Ekert, 1995). Using this method, anyone can
send a message since the public key is used to encrypt messages, but only someone with the
private key can decrypt the messages. Since the encrypting and decrypting keys are different, it
is not necessary to securely distribute a key. The security of public-key encryption depends on

-1-
the assumed difficulty of certain mathematical operations, such as factoring extremely large
prime numbers. (Ford, 2002).
There are two problems with basing security on the assumed difficulty of mathematical
problems. The first problem is that the difficulty of the mathematical problems is assumed, not
proven. All security will vanish if efficient factoring algorithms are discovered. The second
problem is the threat of quantum computers. The theoretical ability of quantum computers to
essentially process large amounts of information in parallel would remove the time barrier to
factoring large numbers. Thus, public-key encryption, though secure at the moment, faces a
serious threat as quantum computing comes closer to reality. Currently, however, this method is
still widely used, especially for the encryption of financial information sent over the internet.
Secret-key encryption requires that two users first develop and securely share a secret
key, which is a long string of randomly-chosen bits. (Ekert, 1995). The users then use the secret
key along with public algorithms to encrypt and decrypt messages. The algorithms are very
complex, and can be designed such that every bit of output is dependent on every bit of input
(Ford, 2002). Suppose that a key of 128 bits is used. “Assuming that brute force, along with
some parallelism, is employed, the encrypted message should be safe: a billion computers doing
a billion operations per second would require a trillion years to decrypt it” (Ford, 2002, p.2).
There are two main problems with secret-key encryption. The first problem is that by
analyzing the publicly-known encrypting algorithm, it sometimes becomes easier to decrypt the
message. This problem can be somewhat offset by increasing the length of the key (Ford, 2002).
The second problem is securely distributing the secret key in the first place. This is the well-
known “key-distribution problem”. Users must either agree on the secret key when they are
together in the same location or when they are in different locations. The drawbacks to
developing the key when they are in the same location are that it is not always practical for the
users to meet, a large database would be needed to store the pre-determined keys, and such
storage is not secure. The drawback to developing a key when the users are in different locations
is that all classical methods of transmitting the key are subject to eavesdropping that cannot be
detected by the users.
Quantum cryptography solves the problems of secret-key cryptography by providing a
way for two users who are in different locations to securely establish a secret key and to detect if
eavesdropping has occurred. In addition, since quantum cryptography does not depend on

-2-
difficult mathematical problems for its security, it is not threatened by the development of
quantum computers. Quantum cryptography accomplishes these remarkable feats by exploiting
the properties of microscopic objects such as photons.
Since photon polarization measurements form the foundation for the most common
quantum cryptographic techniques, it is important to first understand their properties. The three
chosen bases of polarization and the possible results of a measurement according to the bases
are: rectilinear (horizontal or vertical), circular (left-circular or right-circular), and diagonal (45°
or 135°) (Ford, 2002). Although there are three bases, only two bases are used in any given
protocol for quantum cryptography.
Photons can be measured to determine their orientation relative to one of these bases of
polarization at a time. Classically, one would expect the photon to have a certain polarization,
which can be measured but which is not changed by the measurement. Photons, however, are
quantum objects, and in the quantum world an object can be considered to have a property only
after you have measured it, and the type of measurement impacts the property that you find the
object to have. This implies that a photon can only be considered to have a particular
polarization after you measure it, and that the basis you choose for the measurement will have an
impact on the polarization that you find the photon to have.
For example, if you send a photon through an apparatus to measure its orientation relative
to a rectilinear coordinate system, you are asking the question, “How is the photon oriented
relative to a rectilinear coordinate system?” You will find the photon is either vertically
polarized or horizontally polarized -- there are only two possibilities. Suppose you measure this
photon as horizontally polarized. Next you send this same photon through an apparatus to
measure its orientation relative to a diagonal coordinate system. Now you are asking the
question, “How is the photon oriented relative to a diagonal coordinate system?”, and you will
find that the photon is either 45° polarized or 135° polarized – there are only two possibilities.
The type of measurement does indeed have an impact on what property you find. This is in
surprising contrast to the classical situation where something that is horizontally oriented would
be expected to have a component in the diagonal direction . The fact that a horizontally-oriented
photon may subsequently be measured to have a 45° polarization occurs because the state of
horizontal polarization is actually a superposition of the two diagonal polarization states. All
polarization states are actually superpositions of other polarization states.

-3-
 It is important to note that once the diagonal measurement was made, all information
about the previous “property” of horizontal polarization of the photon vanished. As a result it is
impossible to determine a photon’s rectilinear and diagonal polarizations at the same time. This
is analogous to the impossibility of specifying a particle’s position and momentum at the same
time. More information about one results in less information about the other.
The behavior of photons sent through a series of polarizers is illustrated below:
LEGEND
+ = an apparatus that measures rectilinear polarization
V = vertical polarization
H = horizontal polarization
o = an apparatus that measures circular polarization
L = left-circular polarization
R = right-circular polarization

1. A photon is sent through a rectilinear (+) measurement apparatus. The photon has equal
probability of being vertically or horizontally oriented.
photon 1‡ + ‡ V
photon 2‡ + ‡ H
2. A photon that is repeatedly sent through the same measurement apparatus will always give the
same answer.
photon 1‡ + ‡ V ‡ + ‡ V ‡ + ‡ V
photon 2‡ + ‡ H ‡ + ‡ H ‡ + ‡ H
2. A photon that was measured to be vertically polarized is sent through an apparatus to measure
its circular polarization The photon will come out either left-circular polarized OR right-circular
polarized, with equal probability.
V‡o‡L OR V‡o‡R
3. Analogous results would occur if a circularly-polarized photon was sent through a rectilinear
measurement apparatus.

-4-
 In 1984, Bennett and Brassard suggested the first protocol, called “BB84,” for establishing a
secret key using quantum transmissions (Ford, 2002; Ekert, 1995). This protocol uses the
rectilinear and circular polarization bases for photons. The steps of the protocol are explained
below, using the standard convention that Alice is the sender, Bob is the receiver, and Eve is the
eavesdropper.
1. Alice prepares photons randomly with either rectilinear or circular polarizations.

2. Alice records the polarization of each photon and then sends it to Bob.

3. Bob receives each photon and randomly measures its polarization according to the
rectilinear or circular basis. He records the measurement type (basis used) and the
resulting polarization measured. (It is important to remember that the polarization sent by
Alice may not be the same polarization Bob finds if he does not use the same basis as
Alice.

4. Bob publicly tells Alice what the measurement types were, but not the results of his
measurements.

5. Alice publicly tells Bob which measurements were of the correct type. A correct
measurement is the correct type of Bob used the same basis for measurement as Alice did
for preparation.

6. Alice and Bob each throw out the data from measurements that were not of the correct
type, and convert the remaining data to a string of bits using a convention such as:
left-circular = 0, right-circular = 1
horizontal = 0, vertical = 1

Using an online demonstration program (Henle, 2002), the following example data was
generated assuming that Alice sends 12 photons and the detector never fails.
Step Description 1 2 3 4 5 6 7 8 9 10 11 12
1 Filters used by Alice to prepare + + o + o o o + + + o o
photons
2 Polarizations of photons sent by V H L H R L R V H H R L
Alice
3a Measurement types made by Bob + + + + o o o o + o o +
3b Results of Bob’s measurements V H H H R L R L H L R H
4 Bob publicly tells Alice which type + + + + o o o o + o o +
of measurement he made on each
photon
5 Alice publicly tells Bob which yes yes no yes yes yes yes no yes no yes no
measurements were the correct type
6 Alice and Bob each keep the data 1 0 0 1 0 1 0 1
from correct measurements and
convert to binary

-5-
The string of bits now owned by Alice and Bob is: 1 0 0 1 0 1 0 1. This string of bits forms the
secret key. In practice, the number of photons sent and the resulting length of the string of bits
would be much greater.
An essentially equivalent protocol that utilizes EPR correlations has been worked on by
Artur Ekert and David Mermin (Collins, 1992). To take advantage of EPR correlations, particles
are prepared in such a way that they are “entangled”. This means that although they may be
separated by large distances in space, they are not independent of each other. Suppose the
entangled particles are photons. If one of the particles is measured according to the rectilinear
basis and found to have a vertical polarization, then the other particle will also be found to have a
vertical polarization if it is measured according to the rectilinear basis. If however, the second
particle is measured according to the circular basis, it may be found to have either left-circular or
right-circular polarization. The steps of the protocol for developing a secret key using EPR
correlations of entangled photons are explained below:
1. Alice creates EPR pairs of polarized photons, keeping one particle for herself and
sending the other particle of each pair to Bob.

2. Alice randomly measures the polarization of each particle she kept according to the
rectilinear or circular basis. She records each measurement type and the polarization
measured.

3. Bob randomly measures each particle he received according to the rectilinear or

4. Alice and Bob tell each other which measurement types were used, and they keep the
data from all particle pairs where they both chose the same measurement type.

5. They convert the remaining data to a string of bits using a convention such as:
left-circular = 0, right-circular = 1
horizontal = 0, vertical = 1

One important difference between the BB84 and the EPR methods is that with BB84, the key
created by Alice and Bob must be stored classically until it is used. Therefore, although the key
was completely secure when it was created, its continued security over time is only as great as
the security of its storage. Using the EPR method, Alice and Bob could potentially store the
prepared entangled particles and then measure them and create the key just before they were
going to use it, eliminating the problem of insecure storage.

-6-
 But what if Eve has been eavesdropping on Alice and Bob’s transmissions? To
understand why eavesdropping does not compromise the security of keys developed using
quantum cryptography, it is important to note that transmissions between Alice and Bob take
place on two different types of channels. The photon transmissions are quantum in nature and
occur on what will be called a quantum channel, such as optical fiber. The discussions between
Alice and Bob about the types of measurements made occur on a classical channel, such as
telephone or email. The properties of these two channels are very different.
Eve can intercept information transmitted on the classical channel without being detected.
Although this was a serious problem in previous modern cryptographic methods, it is not a
problem in quantum cryptography. This is because in quantum cryptography protocols the only
information Bob and Alice exchange on a classical channel is the type of measurements they
made, which tells Eve nothing about the results of the measurement, and therefore nothing about
the key that was developed.
Suppose Eve listens on the quantum channel. One way she may do this is by skimming
some photons from the burst sent from Alice to Bob. (Although the explanation of the protocol
implied that single photons were sent, in practice it is easier to send bursts of photons.) Now Eve
has photons that are identical in polarization to those received by Bob, but she has to randomly
choose her own measurement type since she doesn’t know what measurement type Bob is going
to use. Therefore, about half the time she will choose a different basis than Bob for
measurement. For example, suppose Alice sends a burst of circularly polarized photons, some of
which are received by Eve and Bob. Eve decided to use the rectilinear basis for measurement,
and Bob decides to use the circular basis. Alice and Bob keep the resulting data since they both
used the same basis, but since Eve used the wrong basis, she doesn’t know what their result was.
Now suppose instead that Eve had decided to measure it according to the circular basis, but Bob
decided to measure it according to the rectilinear basis. Here Eve would know the polarization
that Alice sent, but since Bob did not choose the correct basis, Alice and Bob would throw the
results out. As you can see, Eve will not end up with anything resembling the string of bits that
Alice and Bob create.
Another method Eve could use to eavesdrop on the quantum channel is to intercept the
photons, measure them, and then send them on to Bob. When she chooses a different basis for
measurement than Alice had used for preparation, she will change the photon’s polarization

-7-
through the act of measurement, causing Bob to receive a photon that does not have the same
polarization as that sent by Alice. For example, consider the following scenario. Alice sends a
right-circularly polarized photon, which is intercepted by Eve. Eve measures it according to the
rectilinear basis, finds it to be vertically polarized, and sends it on to Bob. Bob measures it
according to the circular basis, and has an equal probability of finding it to be right-circularly
polarized and left-circularly polarized. It is clear that this will introduce errors into Bob’s final
string of bits. Alice and Bob can detect these errors when they run a “key validity check”, as
described below. Thus, although Eve could obtain some correct bits of the final key by
intercepting photons, Alice and Bob will know that the security of their key has been
compromised.
In the example data generated using the online demonstration program, it was assumed
that the detector never fails. In reality, detectors fail some of the time, there is noise in the
channel, and Eve could be eavesdropping. All of these factors can cause Bob to end up with a
different string of bits than Alice. Thus it is important for Alice and Bob to perform a “key
validity check” to confirm that they have the same string of bits. They cannot compare the whole
string of bits over a classical channel since this would compromise the security of their key. One
approach is to compare a large random subset of their string of bits, assuming that if these match
up, then the others that they are not comparing also match up and can be used as the key
(Collins, 1992). The bits that they have compared are discarded, since this information was
shared over a public channel and could have been intercepted. This technique is useful for
detecting eavesdropping, since any activity by Eve would introduce a large number of errors.
Another technique was developed by Bennett, Brassard and Jean-Marc Robert of the University
of Quebec at Rimouski (Collins, 1992). Using this technique, Alice and Bob agree on certain
blocks of bits of their key and calculate and compare the parity of each block. If an odd-number
of errors exist, the parity calculated by Alice and Bob will be different. Alice and Bob check
many overlapping blocks, and make smaller and smaller block sizes to find the errors. Each time
a comparison is performed, Alice and Bob discard the last bit in the block. In this way, errors an
be eliminated, and Alice and Bob can be sure that they have the same string of bits for use as the
key.
The practicalities of performing transmissions on a quantum channel are not as simple as
the theory. The light source is usually an LED or laser, and the sender produces a low-intensity

-8-
polarized beam that is emitted in short bursts. The polarization of each burst is randomly
modulated to either horizontal ,vertical, left-circular, or right-circular before being sent on to the
receiver (Ford, 2002). Recently, researchers at the University of California at Santa Barbara
have reported that they have found a way to emit single photons. Single photon emission would
prevent Eve from skimming off part of a photon burst, making it possible to produce a key that is
“secure from the most advanced attacks.” (Savani, 2000).
The quantum information in the form of polarized photons may then be sent over optical
fiber or through free space, also called free space optics (FSO) (Bains, 2002). The difficulty
with sending quantum information over optical fiber is that polarizations are not retained over
long distances (Collins, 1992). Improvements in optical fiber may help extend the distance over
which such information can be sent. Another possible way to extend the distances is to use
interferometry, looking at differences in phase instead of polarization (Collins, 1992). Using
fiber optic cables, photon bits have successfully been transmitted over distances up to 60 km,
which is about 37 miles (HP, 2001). Transmitting through the air eliminates the problems of
impurities in optical fiber, but so far, successful transmission has been over shorter distances and
the weather conditions must be ideal. In early 2002, researchers exchanged a key at night from
the mountaintops at Zugspitze and Karwendelspitze in Germany. This transmission over 23.4 km
was a record (Bains, 2002). Los Alamos National laboratory has reported an exchange over 1.6
km during daylight (Bains, 2002). Such transmissions could be useful in military applications,
where the key is exchanged from one ground station to a satellite and then to another ground
station (Bains, 2002).
Whether quantum cryptography will replace classical cryptography techniques will
depend on many factors including transmission distance, expense, and ease of use. It is suspected
that quantum cryptography is already being used between the White House and the Pentagon
(HP,2001; Metz, 2002). There may also be connections between certain military cites, large
defense contractors, and research laboratories that are not very far apart (HP, 2001). A possible
commercial application that could utilize quantum cryptography is “two-party secure
computation” in which two parties compare results of a computation without revealing the data
used by each party to complete the computation (Collins, 1992). Since the two parties could be
sitting at the same table, distance is not a problem.

-9-
 In conclusion, the ultimate security of quantum cryptography makes these techniques a
valuable resource for this dangerous age when secure transmission of sensitive information is
more important than ever.

References

Bains, S. 2002. It’s free-space optics’ turn. Electronic Engineering Times, March 18, 2002,
www.eet.com.

Collins, G. 1992. Quantum cryptography defies eavesdropping. Physics Today, November 1992,
pp. 21-23.

Ekert, A. 1995. What is quantum cryptography. http://www.qubit.org/index.html Accessed

Ford, J. Quantum cryptography tutorial. http://www.cs.dartmouth.edu/~jford/crypto.html.

Henle, F. 2002. BB84 Demo. http://www.cs.dartmouth.edu/~henle/Quantum/cgi-bin/Q2.cgi.

Hewlett Packard, 2001. http://searchhp.techtarget.com/sDefinition/0,,sid6_gci284012,00.html.

Metz, C. 2002, Total Lockdown. PC Magazine, September 17, 2002. www.pcmag.com.

Savani, J. 2000. Eavesdroppers beware: Single photon emission prepares way for quantum
cryptography. UCSB College of Engineering Press Release.
http://www.engineering.ucsb.edu/Announce/quantum_cryptography.html.

- 10 -