Optics and Laser Technology 101 (2018) 312–318

Contents lists available at ScienceDirect

Optics and Laser Technology

journal homepage: www.elsevier.com/locate/optlastec

Full length article

Development of authentication code for multi-access optical code

division multiplexing based quantum key distribution
Ambali Taiwo a,⇑, Ghusoon Alnassar a, M.H. Abu Bakar a, M.F. Abdul Khir b, Mohd Adzir Mahdi a,
M. Mokhtar a
a
Computer and Communication Sys. Engineering/Center of Excellence for Wireless and Photonic Network (WiPNET), Faculty of Engineering, University Putra Malaysia, Malaysia
b
Information Security and Assurance Programme, Faculty of Science and Technology, Universiti Sains Islam Malaysia, Bandar Baru Nilai 71800 Nilai, Negeri Sembilan, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history: One-weight authentication code for multi-user quantum key distribution (QKD) is proposed. The code is
Received 11 January 2017 developed for Optical Code Division Multiplexing (OCDMA) based QKD network. A unique address
Received in revised form 4 October 2017 assigned to individual user, coupled with degrading probability of predicting the source of the qubit
Accepted 20 November 2017
transmitted in the channel offer excellent secure mechanism against any form of channel attack on
OCDMA based QKD network. Flexibility in design as well as ease of modifying the number of users are
equally exceptional quality presented by the code in contrast to Optical Orthogonal Code (OOC) earlier
Keywords:
implemented for the same purpose. The code was successfully applied to eight simultaneous users at
Authentications
Key distribution
effective key rate of 32 bps over 27 km transmission distance.
Information security Ó 2017 Elsevier Ltd. All rights reserved.
Quantum communication
Cryptography

1. Introduction eventually lessens the impending challenges in practical QKD that

are based on weak pulse laser source.
Ever since its discovery three decades ago [1], quantum cryp- Subsequent works focused on the transmission distance and
tography has maintained its exceptionality in providing antici- secured key rate. The first practical implementation of QKD was
pated secure mechanism between two communicating ends. Its only achieved over 30 cm distance. The latest development have
security establishment has remained a promising solution to the reported between tens and hundreds of kilometers [10,11 which
challenges posed by the progress towards quantum computing, a have been achieved both theoretically and in practical. A number
rare opportunity that might be explored to mount attack on the of works on the other hand, focus on improving the secret key rate
existing security system that relied on mathematical complexity. while sacrificing the supported transmissions distance. A key rate
Quantum cryptography often referred to as quantum key of 1 Mb/s was recently reported in [12] over a distance of 20 km
distribution (QKD) proffers its security measure by exchange of and 50 km [13] respectively. One specific thing about all the
set of random bits between two users [2–4], which upon series above-described works is that they are only addressing a point-
of post-processing, are used in the generation of secured key for to-point communication between two end users.
information exchange. Its security provision, guaranteed by funda- The latest trends in QKD development are unequivocally
mental law of physics [2,4–6] has remained unconditional in the exploring the multi-user ability of the system. Imagine an estab-
presence of eavesdroppers. lishment with several sections within a certain location. Each user
Over the years, tremendous progress have been recorded in the requires a distinctive way of identification with minimal resources.
area of the security enhancement against all emerging loopholes in Subcarrier QKD multiplexing was proposed in [14] and further
various QKD protocols [4, 7–9] One of the most appealing among developed in [15,16]. The technique, which saw bands of diverse
them is the decoy state system [2], which provides security against frequency used in modulating the weak pulse laser, was subse-
Photon Number Splitting (PNS) attack. This has been achieved quently experimented for two users in [17]. Its low key rate was
through additional pulses that tends to raise an alarm whenever due to complexity in the system design. Subsequent work include
alteration is perceived in the intensity of the received pulses. This Orthogonal Frequency Division Multiplexing of QKD (OFDM-QKD)
[18] proposed in 2015. The system recorded better performance
only with activate decoding technique which subsequently add
⇑ Corresponding author.
to the cost and complexity of the design. Previously in 2012, Razavi
E-mail address: [email protected] (A. Taiwo).

https://doi.org/10.1016/j.optlastec.2017.11.034
0030-3992/Ó 2017 Elsevier Ltd. All rights reserved.
 A. Taiwo et al. / Optics and Laser Technology 101 (2018) 312–318 313

proposed a thrilling multi-access QKD system based on OCDMA The patterns are then treated as sets of binary digit and subse-
system [19]. The system was demonstrated with both passive quently converted to their respective decimal equivalents by add-
and active decoders. The active system is a form of ‘‘listen-befor ing their corresponding decimal values horizontally.
e-send” approach whereby both communicating ends pay atten-
tion to the channel to ensure it is unengaged before transmission.
The work however made use of Optical Orthogonal Code (OOC),
which has complexity in its derivation as well as longer code
length, with heavy spectral dependency on the light source [20,
21].
This work proposed a new code, which is suitable for multi-
plexed QKD system. As established in [19] that a single weight
code has proximity in performance to time division based system
than multi-weight code, the proposed code is also a single-
weight code which at the same time, is secured against channel
attack. One other unique characteristics of the code is that it could The obtained decimal values {x = 384, 192, 48, 24, 6, and 3} are
equally be used as time-dependent code with all users sharing the divided by the common factor ‘‘3” to form the new pattern {128,
same frequency spectrum at distinctive time interval. 64, 16, 8, 2, 1}. These are subsequently converted to their respec-
The paper is organized as follows; Section 1 contains the intro- tive binary form to generate the usable code sequence.
ductory part while Section 2 vividly describes the proposed code
derivation. The system setup will be addressed in Sections 3 and
4 will explain the mathematical derivation of the system. The
results will be discussed in section 5 and the concluding part will
be addressed in Section 6.

2. Proposed code design

The code begins with the formation of a matrix with a logical

pattern of binary digit ‘‘1” and ‘‘0” denoting presence and absence
of a certain pulse. The bits are arranged in 3  2 matrix with the
first two bits position of the first row and last two bits position
of the second row occupied by bit ‘‘1” as shown in the matrix. With each row as a unique address location, each user can


1 1 0
 uniquely communicate with their respective partners on the chan-
nel with no or minimal interference from the adjacent users. The
0 1 1
chip locations are diverse wavelength corresponding to the column
The bit pattern can then be increased by diagonal repetition of value as shown in the final derivation above. One other unique
the matrices. characteristic of the proposed code is occasional presence of ‘‘all
zero” column. Unlike the conventional channel code for classical
1 1 0
2
 3
6 0 1 1 7 OCDMA system where overlapping of chips are used to determine
6 7 the level of security, the level of ambiguity in predicting the exact
1 1 0
6
 7
6
6
7
7 spectrum being used by a certain users in the derived quantum
6
6 0 1 1 7
7 code could as well serve as a source of security. Other benefits of
1 1 0 5 the code is flexibility in modifying the required number of users
6
7
4
0 1 1 to suit ones purpose.
Theoretically, the number of users can further be increased
The patterns are combined in a large matrix to form a single through the formation of matrix C i;j with the position of the bits
matrix with N column and M row. This ensures that the element corresponding to
are located along the major diagonal in the matrix.
C 0;0 C 0;n
2 3




1 1 0
2 3
6. .. .. 7
60 1 1 6 .. 7
C i;j ¼ 6
7 . . 7
6 7
4 C m;0


C m;n 5
6 7
1 1 0
6 7
6 7
6 7
6
6 0 1 1 7
7
1 1 05
6 7
4 For i ¼ 0; 1; 2; . . . . . . . . . :m and j ¼ 0; 1; 2; . . . . . . n.
0 1 1 The positions of ‘‘1” in Table 1 are at row (i) and column (j) loca-
tions described as
The emerging vacuum are subsequently filled with bit ‘‘0” to
obtain a full pattern of binary sequence. C 0;1 ¼ C 1;2 ¼ C 2;4 ¼ C 3;5 ¼ C 4;7 ¼ C 5;8 ¼ 1
1 1 0 0 0 0 0 0 0 In order to derive the location mathematically, the following steps
2 3
6
6 0 1 1 0 0 0 0 07
70 are followed.
6
6 0 0 0 1 1 0 0 0 07
7 Step 1: Count j from 0 and increase by 1 in each stage. j = 0, 1, 2,
6 7 3, 4 . . .
6
6 0 0 0 0 1 1 0 0 07
7
Step 2: To find column with all ‘‘0”. All element of a column will
0 0 0 0 0 0 1 1 05
6 7
4
be ‘‘0” if jmod3 ¼ 0.
0 0 0 0 0 0 0 1 1