Performance Evaluation of Secure Key Distribution Based on the B92 Protocol

International Journal of Advanced Engineering, Management and Science (IJAEMS) [Vol-4, Issue-6, Jun- 2018]
https://dx.doi.org/10.22161/ijaems.4.6.6 ISSN: 2454-1311
www.ijaems.com Page | 466
Performance Evaluation of Secure Key
Distribution Based on the B92 Protocol
Brahim Ouchao1
, Abdeslam Jakimi2
1
Department of Computer Science Faculty of Sciencs and Technics Errachidia, RO&I Team, Moualy Ismail University
ouchaob@gmail.com
2
Department of Computer Science, GL&ISI Team, Faculty of Sciencs and Technics Errachidia, Moualy Ismail University
ajakimi@yahoo.fr
Abstract— Quantum Cryptography (QKD) uses the laws
of Quantum Mechanics to create new cryptographic
primitives. Precisely, this technique is based on two laws
of quantum mechanics, namely uncertainty principle and
no-cloning theorem. The best known QKD is the BB84
protocol published by Bennett and Brassard in 1984. Since
then, various QKD protocols have been developed for
example a B92 protocol. This latter is similar to that of
BB84but this time using only two non-orthogonal states of
the four states of BB84. In this work, we develop the java
simulation B92 protocol and we give some details of the
implementation. Next, we show by an example how Alice
(Initiator) can configure the protocol parameters before
execution. Finally, we give the statistical results allowing
to compare this protocol with that of BB84.
Keywords— QKD, Qubit, BB84, B92.
I. INTRODUCTION
The basic idea of the security of quantum transmissions
was the instability of the quantum elements during the
measurement. The B92 protocol is a simpler version of
BB84 [1,2]. It is a two-state protocol (it uses two non-
orthogonal quantum states) invented by Charles Bennett in
1992. It is based on the fact that two non-orthogonal
quantum states are sufficient to guarantee the detection of
an eavesdropper. Since no measurement can distinguish
two non-orthogonal quantum qubit [3], it is impossible to
identify the qubit with certainty. In addition, any attempt
to learn the qubit will alter the qubit in a notable way. The
coding scheme B92 is sometimes easier to implement. The
coding scheme B92 uses a one-to-one correspondence
between the conventional qubits and the quantum states.
To send the qubit b, Alice prepares a photon in the
following quantum state: 0° Horizontal polarization codes
0 and +45° Diagonal polarization for coding 1.
The rest of the paper is organized as follows: Section 2
presents and introduces our approach. Section 3 gives a
predict the ear of a spy. Section 4 illustrates and presents
the comparison of BB84 and B92. Section 5 concludes the
paper and provides an outlook into future work.
II. CONCEPTION
Fig.1 shows the interface of our application that provides
an implementation explaining a quantum transmission
simulation based on the B92 protocol. Although it
provides only a secret key distribution, the architecture can
exploit to ensure a more secure transmission. Our
application consists of designing a protocol running on a
single computer or on two computers (servers) connected
remotely via a socket connection [6].
Fig. 1: Simulation of the B92 protocol without spy.
These servers can be used either to simulate a quantum
channel, or to interact the two ends of a real quantum
channel. The output generated by a protocol execution is a
collection of HTML [7] files providing feedback for all
relevant phases of the protocol. The software architecture
consists of 4 main steps:
 Alice randomly prepares each photon with ether 0°
Horizontal polarization assigned a qubit of value
(|0>) or |+45> Diagonal polarization assigned a
qubit of value (|1>) [4, 5].
 Alice sends the polarized photon to Bob. This one
is equipped with a polarization analyzer that
measures randomly along the two basis. that is to
say according to the following four directions (0°,
+45°, 90° or -45°)

 Bob informs Alice via an authenticated public
channel the basis used, but not the polarization of
the photon.
 Alice and Bob keep the polarizations of the photons
(qubits) sent and received in directions 45° and 90°
and reject those received in directions 0° and 45°.
1.1. Example
For example, if Bob detects a photon by measuring with a
polarized filter on the 90° direction, he knows that Alice
has sent a polarization photon 45° (it cannot be 0°
polarization) and therefore it encodes this value by the bit
1. Bob can therefore assign the detections 90° the binary
values 1, and the detections 45° the value 0. If Bob does
not detect the photon, he cannot be certain which state
Alice sent. Thus, Alice and Bob keep only those
measurements where Bob detected a photon. This
sequence of 0 and 1 bits values forms the key.
1.2. Algorithm of B92
The algorithm of the B92 protocol is:
III. PREDICT THE EAR OF A SPY
1.3. Case: without spy
Alice and Bob communicate in an authenticated public
channel to determine which photons have been detected.
To do this, they exchange a small number of their bit
values which are then rejected (because they are no longer
secure) to check for errors. Assuming no spy has
intervened, Alice and Bob will have the same qubit
sequences. Unlike the BB84 protocol which extracts a
sample from the shared key, the B92 protocol even
exploits the rejected qubits to calculate the probability of
transmission error. It leaves the key intact. After publicly
communicating, Alice and Bob find half of these qubits
rejected common. Which shows the absence of espionage
listening. This is the ideal case for negligence and errors
due to the impurity of the transmission channel. Fig 2.
illustrates the results obtained
Fig. 2: Average number of qubits received correctly in the
absence of the Spy.
1.4. Case: with spy
Figure3 illustrates the interface of the application in the
presence of a spy. The analysis of the results obtained
shows that the increase in the measurement rate of the spy
disrupts the transmission. Indeed, if the probability of error
is better than 0:25, then the transmission is no longer
secure.

Fig. 3: Simulation of B92 protocol with spy.
1.5. Interpretation
Fig. 4 illustrates the results obtained. It can be seen that the
average number of qubits measured between Alice and
Bob decreases compared to the previous case (case without
spy). Certainly, as long as the error rate is below a
threshold (Perror < 0:25) [4] the spy remains deaf.
IV. COMPARISON OF BB84 AND B92
1.6. View on BB84 and B92
With some simple modifications in BB84, one can adopt it
in model B92. In this last protocol Alice does not encode
the qubit randomly. If the generated bit is zero then it
sends the polarized photon 0° if the bit is equal to one then
the polarized photon 45° is exchanged. With some
modifications in the code of BB84, we can adopt it in
model B92.
1.7. Attack "Intercept Resend"
In terms of indiscreet detection, B92 is better than BB84.
This is well explained in figure5 and figure6.
1.8. Analysis and interpretation
The probability of detection of spy converges faster to 1
when the number of quits exchanged is large. Thus the
number of correct measurements between Alice and Bob is
reduced; this is due to the listening of the spy's indiscreet
ear.
Fig. 4 :Average number of qubits received in the same
databases in thepresence of a only spy.
In addition, the simplicity of implementing the B92
protocol and the Figure 5: BB84 and B92 comparison for
Intercept Resend correct measurements by eavesdropper.
Number of de-mined exchange bit makes this protocol
more efficient compared to its BB84 ancestor.
Fig. 5: BB84 and B92 Comparison for Intercept Resend
eavesdropper detection.

V. CONCLUSION
In this work we have developed a java simulation of
protocol B92 (BB84 version) for key quantum exchange.
The analysis of the results obtained allowed us to define a
security zone where the presence of the spy does not
influence the transmission. This study also confirmed to
us, after a statistical comparison, the importance of the
B92 protocol in relation to its BB84 ancestor and this,
even if their main steps are similar.
As future works, we will evaluate our approach on more
complex systems as quantum teleportation. We will study
the possibility of developing the java simulation,
comparing and giving more details of the implementation.
REFERENCES
[1] Ch.H. Bennett, G. Brassard and A. Ekert, Quantum
cryptography,Scienti_c Am. 267,pp. 26-33, 1992 (int.
ed.).
[2] Ch.H. Bennett, G. Brassard, Quantum cryptography:
Public key distribution and coin tossinh. In
Proceedings of the IEEE International Conference on
Computers, Systems, and Signal Processing,
Bangalore, India, IEEE, New York, pp. 175-179,
1984.
[3] B. Ouchao and E. H. El Kinani, Statistical Analysis of
Common Qubits between Alice and Bob in BB84
Protocol, Contemporary Engineering Sciences, 4, 8,
363-370, 2011.
[4] B. Ouchao and E. H. El Kinani, Analysis of the
Inuence of the Rate of Spies'Measure on the Quantum
Transmission, Science Journal of Mathematics and
Statistics, 2012, 7
[5] B. Ouchao and E. H. El Kinani, Etude statistique de
l'inuence de pr_esence des espions: Simulation du
protocole BB84 en langage Java, African Journal of
Mathematical Physics, 11, 41-46 2012.
[6] V. S. Bagad and I. A. Dhotre, Computer Networks
(5th revised edition, 2010 ed.), Technical Publications
Pune, p. 52, 2008.
[7] Matthew MacDonald, HTML 5 - The Missing
Manual", 2nd ed, OReilly. 2014.8

Performance Evaluation of Secure Key Distribution Based on the B92 Protocol

Recommended

More Related Content

Similar to Performance Evaluation of Secure Key Distribution Based on the B92 Protocol (20)

Recently uploaded (20)

Performance Evaluation of Secure Key Distribution Based on the B92 Protocol