Journal of Theoretical and Applied Information Technology

th
20 July 2013. Vol. 53 No.2
2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

IMPLEMENTATION OF STRONGER AES BY USING

DYNAMIC S-BOX DEPENDENT OF MASTER KEY
1
SLIMAN ARRAG, 2ABDELLATIF HAMDOUN, 3ABDERRAHIM TRAGHA, AND 4SALAH
EDDINE KHAMLICH
1,2 and 4
Laboratory of treatment of information Universite Hassan II Mohammedia, Casablanca, Morocco
3
Laboratory of Information Technology and Modeling, Universite Hassan II Mohammedia, Casablanca,
Morocco
E-mail: arragsliman@yahoo.fr, alhamdoun@yahoo.fr, 3a.tragha@univh2m.ac.ma,
1 2

4
khamlich.salah@gmail.com

ABSTRACT

In this paper we propose and we show a new approach of nonlinear transformation algorithm for AES
SBox to enhance the complexity of the SBox structure,we making AES stronger by using Dynamic
Sbox,with look up table Sbox and Key expansion as modified when we change the initial key, that
effectively providing a high resistance against differential cryptanalysis and especially the linear
cryptanalysis. The structure of the AES S-box has been expanded and modified to be accordance with the
proposed algorithm and to obtain good nonlinearity of the Sbox. This has been done without changing the
basic operations of AES.The proposed modifications of the Advanced Encryption and a modified Sbox are
implemented in Cyclone II Dvice by using VHDL language.
Keywords: AES, Dynamic S-box, Master key, FPGA, Vhdl.

1. INTRODUCTION ISE synthesis tool and Qaurtus II v9 for great

simulated and synthesized. The architectural design
The evolution of information technology and in has been described code in VHDL language. The
particular the increase in the speed of processing AES has for the moment not been broken but the
devices has necessitated the need to reconsider the cryptanalysis of Rijndael (AES) has not stopped, for
cryptographic algorithms used. The National that we used a new approach allows us to obtain
Institute of Standards and Technology of the United competitive performance. This approach will be
States (NIST) in cooperation with industry and studied its effect on the AES focusing on the S-
cryptographic communities [1] have worked Boxes wich look up table Sbox and Key expansion
together to create a new cryptographic standard. as modified when we change the initial key.
The primary objective was a federal standard Literatures [4], [10], [11], [12], [13], [18], [20],
(Federal Information Processing Standard FIPS) to [21], [22], [25] describe design and implementation
establish that specifies a cryptographic algorithm of AES and AES modified treatment in the FPGA.
with the ability to improved protect sensitive
2. AES RIJNDAEL ALGORITHM
information from the government. It was expected
that the algorithm can be used both in the The AES algorithm (or rijndael) [2], [3],
governmental structures of the United States and in [5] takes as input a block of 128 bits (16 bytes), the
the analysis of companies and private full key is 128, 192 or 256 bits. The 16 bytes are
sectors.After (mathematics, cryptography, statistics, swapped input according to a predefined table.
engineering, etc.) of the algorithms, NIST [8] These bytes are then placed in a matrix of 4x4
announced that the new standard uses the Rijndael elements and lines are rotated to the right. The
algorithm and since 2001, it is the foundation of the increment for the rotation varies with the number of
new encryption standard AES. the line. A linear transformation is then applied to
the matrix, it consists of the binary multiplication of
In this paper we presente the design and
each matrix element with polynomials from a
hardware implementation of the AES (Advanced
auxiliary matrix, this multiplication is subjected to
Encryption Security) in Cyclonne II, as using Xilinx

196
 Journal of Theoretical and Applied Information Technology
th
20 July 2013. Vol. 53 No.2
2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

special rules according GF (28) (or Galois field Table 1: Look Up Table Of S-BOX.
finite field). Linear transformation ensures better
diffusion (spread of bits in the structure) over 63 7c 77 7b f2 6b 6f c5 30 01 67 2b fe d7 ab 76
several turns.Finally, an XOR between the matrix ca 82 c9 7d fa 59 47 f0 ad d4 a2 af 9c a4 72 c0
and other matrix allows to obtain an intermediate
b7 fd 93 26 36 3f f7 cc 34 a5 e5 f1 71 d8 31 15
matrix. These operations are repeated several
times and define a "turn." For a 128, 192 or 256, 04 c7 23 c3 18 96 05 9a 07 12 80 e2 eb 27 b2 75
AES requires respectively 10, 12 or 14 rounds. 09 83 2c 1a 1b 6e 5a a0 52 3b d6 b3 29 e3 2f 84
53 d1 00 ed 20 fc b1 5b 6a cb be 39 4a 4c 58 cf
The following diagram describes briefly the d0 ef aa fb 43 4d 33 85 45 f9 02 7f 50 3c 9f a8
progress of encryption Fig 1: 51 a3 40 8f 92 9d 38 f5 bc b6 da 21 10 ff f3 d2
cd 0c 13 ec 5f 97 44 17 c4 a7 7e 3d 64 5d 19 73
60 81 4f dc 22 2a 90 88 46 ee b8 14 de 5e 0b db
e0 32 3a 0a 49 06 24 5c c2 d3 ac 62 91 95 e4 79
e7 c8 37 6d 8d d5 4e a9 6c 56 f4 ea 65 7a ae 08
ba 78 25 2e 1c a6 b4 c6 e8 dd 74 1f 4b bd 8b 8a
70 3e b5 66 48 03 f6 0e 61 35 57 b9 86 c1 1d 9e
e1 f8 98 11 69 d9 8e 94 9b 1e 87 e9 ce 55 28 df
8c a1 89 0d bf e6 42 68 41 99 2d 0f b0 54 bb 16

SHIFT_ROW is a function operating lags

(typically it takes into 4 pieces of 4 bytes and
operates left shifts of 0, 1, 2 and 3 bytes for the
tracks 1, 2, 3 and 4 respectively).

Figure. 1: Structure Of The AES Algorithm.

Three main criteria have been met in its design: Figure 3: Structure Of The Shiftrows
-Resistance against all known attacks.
-Speed code on the widest variety of platforms MIX_COL is a function which converts each
possible. input byte into a linear combination of input bytes
-Simplicity in design. and can be mathematically expressed by a product
Rijndael (1998) was strongly influenced by his marticiel on the Galois field (28) [15].
predecessor, Square algorithm (1997).Crypton and
Twofish algorithms also use transactions Square.
Rijndael is pronounced "Raindal".

BYTE_SUB (Byte Substitution) is a nonlinear

function operating independently on each block
from a table called substitution.

Figure 4: Structure Of The Mixcolumn

Figure 2: Structure Of The Subbyte

197
 Journal of Theoretical and Applied Information Technology
th
20 July 2013. Vol. 53 No.2
2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

The Add RoundKey operation is a simple bit by first column of the i-th key in the second column of
bit XOR operation between the data and the the first key-i.
roundkey (by using initial key or Key
expansion).
The circled means exclusive OR operation
(XOR).
K i is the ith subkey calculated by an algorithm from
the master key K.
Figure 9:XOR-Ing Between Modified Row And Second
Key Expansion operation: The algorithm for Row Of Master Key
generating the 10 turns of the round key is as
follows: The fourth column of the i-1 key is rotated This iteration continues for the other two columns
such that each element is moved one row. to generate the key to ith.

Figure 5:Rotated The Last Row

Figure 10: Example Key Extension
It then places the result by way of an algorithm
forwards Sub-Box that replaces all eight bits of the In addition, this process continues iteratively to
matrix with a corresponding 8-bit value of S-Box. generate all 10 keys As a final note, all these keys
(See figure 7 below for SubByte inverse). are stored in a static manner once they have been
calculated first as th key generated is necessary for
the tower (10-i) th decryption.

Decryption is the process of applying the reverse

operations in the reverse order and with subkeys
also in reverse order.
Figure 6:Subbyte The Row By Using Sbox Look Up Table
2.1 Attacks
To generate the first column of the i-th key, this The AES has for the moment not been
result is XOR-ed with the first column of the first i- broken and the exhaustive search ("brute force") is
key and a constant (row constant or Rcon) which the only solution. Rijndael was designed so as to
depends on i. make conventional methods such as linear and
differential analysis very difficult[17], [19].

3. DEFINING THE PROBLEM

One of the main functions used in AES is
the "Substitution of bytes (SubBytes)". This
function performs a non-linear substitution, which
Figure 7:Look Up Table Of The Rcon is performed independently on each input byte. The
matrix which gives the relationship between the
input and output bytes, called (S-BOX) in the AES
algorithm is invertible.

Each matrix must meet the following criteria:

balance, non-linearity, completeness, strict
avalanche criterion, low Table XOR ,order
Figure 8:Operation Xor Between First Row Of Key And
Last Row After The Modefication And First Row Of Rcon
diffusion, Invertability, the static criteria
(independence between the input and data output,
The second column is generated by XOR-ing the the independence between the output and the input
data, the independence between the output and the

198
 Journal of Theoretical and Applied Information Technology
th
20 July 2013. Vol. 53 No.2
2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

output data), the dynamic criteria (Dynamic A recalculation inverse S-BOX is used to decrypt
independence between the data input and output, the plaintext.
the independence between dynamic the input and
output of data, dynamic independence between the By using the above described method 256
output and output data), specific criteria (Entire S- substitution matrices have been obtained. One of
BOX and non-contradiction). These criteria are them SBOXAES 0hex is equal to original SBOX
defined and described in detail in [9], [14], [23]. suggered in AES core. Some of them are depicted
Requirements defined so that each S-BOX must in Table 1 and Table 2 , respectively, SBOX
meet are determined by the need to be stable computed by XOR with byte equal to 24hex and 6F
algorithm for linear cryptanalysis and differential at hex.
a time.
Table 2: SBOX 24 (S-BOX 24 hex ).
Therefore, to meet the requirements set forth above,
new substitution matrices must be found, which
should be applied in the algorithm parameters or
47 58 53 5F D6 4F 4B E1 14 25 43 0F DA F3 8F 52
function key values and at the same time, these S-
boxes must have characteristics same or higher EE A6 ED 59 DE 7D 63 D4 89 F0 86 8B B8 80 56 E4

than those used in the AES standard. 93 D9 B7 02 12 1B D3 E8 10 81 C1 D5 55 FC 15 31

20 E3 07 E7 3C B2 21 BE 23 36 A4 C6 CF 03 96 51
4. PROPOSITION FOR GENERATING 2D A7 08 3E 3F 4A 7E 84 76 1F F2 97 0D C7 0B A0
NEW S-BOX DEPENDING OF MASTER
77 F5 24 C9 04 D8 95 7F 4E EF 9A 1D 6E 68 7C EB
KEY
F4 CB 8E DF 67 69 17 A1 61 DD 26 5B 74 18 BB 8C

Having the substitution matrix used in 75 87 64 AB B6 B9 1C D1 98 92 FE 05 34 DB D7 F6

AES core as a basis, it is necessary to identify other E9 28 37 C8 7B B3 60 33 E0 83 5A 19 40 79 3D 57
same matrices with the same or better 44 A5 6B F8 06 0E B4 AC 62 CA 9C 30 FA 7A 2F FF
characteristic. The principal SBOX (showing in
C4 16 1E 2E 6D 22 00 78 E6 F7 88 46 B5 B1 C0 5D
figure ), used in the cryptographic operation is
regarded as a base.. C3 EC 13 49 A9 F1 6A 8D 48 72 D0 CE 41 5E 8A 2C

9E 5C 01 0A 38 82 90 E2 CC F9 50 3B 6F 99 AF AE

Based on the S-BOX suggested in AES (SBOX 54 1A 91 42 6C 27 D2 2A 45 11 73 9D A2 E5 39 BA

AES) and depending on key used to encrypt the C5 DC BC 35 4D FD AA B0 BF 3A A3 CD EA 71 0C FB
new plaintext substitution matrices are compted
A8 85 AD 29 9B C2 66 4C 65 BD 09 2B 94 70 9F 32
(SBOXxor) by the following process:
Table 3: SBOX 6F (SBOX 6F hex ).
First of all select one byte from master key
0C 13 18 14 9D 04 00 AA 5F 6E 08 44 91 B8 C4 19
(initial key) Key[i];
Computting new SBOXxor, where each cell is A5 ED A6 12 95 36 28 9F C2 BB CD C0 F3 CB 1D AF

equal to XOR with selected byte, D8 92 FC 49 59 50 98 A3 5B CA 8A 9E 1E B7 5E 7A

SBOXxor[x,y]=SBOXAES[x,y] Key[i]; 6B A8 4C AC 77 F9 6A F5 68 7D EF 8D 84 48 DD 1A
A substitution matrix newly calculated is used 66 EC 43 75 74 01 35 CF 3D 54 B9 DC 46 8C 40 EB
for plaintext encryption. 3C BE 6F 82 4F 93 DE 34 05 A4 D1 56 25 23 37 A0
BF 80 C5 94 2C 22 5C EA 2A 96 6D 10 3F 53 F0 C7
Decryption process will the following approach
3E CC 2F E0 FD F2 57 9A D3 D9 B5 4E 7F 90 9C BD
is used:
A2 63 7C 83 30 F8 2B 78 AB C8 11 52 0B 32 76 1C

Selecte same byte from key - Key[i]; 0F EE 20 B3 4D 45 FF E7 29 81 D7 7B B1 31 64 B4

Computing new SBOXxor, where each cell is 8F 5D 55 65 26 69 4B 33 AD BC C3 0D FE FA 8B 16

equal to XOR with selecte byte, 88 A7 58 02 E2 BA 21 C6 03 39 9B 85 0A 15 C1 67
SBOXxor[x,y]=SBOXAES[x,y] Key[i]; D5 17 4A 41 73 C9 DB A9 87 B2 1B 70 24 D2 E4 E5
Computing inverse matrices by using 1F 51 DA 09 27 6C 99 61 0E 5A 38 D6 E9 AE 72 F1
SBOXxor, SBOXxor INV=INV (SBOXxor)
8E 97 F7 7E 06 B6 E1 FB F4 71 E8 86 A1 3A 47 B0
(structure of the SBOXxor inv showing in the
figure); E3 CE E6 62 D0 89 2D 07 2E F6 42 60 DF 3B D4 79

199
 Journal of Theoretical and Applied Information Technology
th
20 July 2013. Vol. 53 No.2
2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

Note: in the table 1 we considered 24 hex first one generated will used for finding a value that is used
byte on master key, and same idea for 6F hex . to rotate the S-box.
The same round keys are used for AddRoundKey
4.1 New algorithm using the SBOX depending stage as well. Suppose for a particular round j, if
on the AES key the round key value is :
Now, with Sbox dependent initial key, AES will be X"6172726167736C696D616E5F6D697469".
much stronger. [24] We now present how the
property above of Sbox can be used to master the Here XOR operation of all the bytes is taken.
function key by using one of the two cases (we can
also use other cases) depending on the level of the 35(Hex)=6172726167736C696D6
security requirement. In case the demand moderate 16E5F6D697469.
level of security Case 1 can be used. For more high
security requirements case 2 may be adopted. The resulting byte value 35(Hex) is used to XOR-
ing the Sbox (change the sbox by using XOR
operation between each byte of fixe Sbox(showing
in Table 1) and the 35 (hex) ). The resulting Sbox is
used during the Sub Bytes operation.

On the basis of the surveyed substitution matrices

and depending on the particular value of the
encryption key used in Advanced encryption
standard the following algorithm is offered:

Encryption:

Chose a key initial for AES;

The first byte of the Key master is selected
Figure 11:Our Suggestion For Dynamic Sbox Depending Key[1];
Of Initial Key Computing new SBOXkey[1] =
SBOXAESKey[1] (showing in figure 11);
Case 1: Continue according to the algorithm set out in
Here the various round keys are produceded using a AES by using new calculated S-BOXkey[1].
expansion key algorithm which is similar to that in
AES core key extension algorithm. The round keys Decryption:
thus produceded will used for find a value that is
used to change the static Sbox.
Chose a key initial for AES;
The similar round keys are used for Add RoundKey
step as well. Supposed for a particular round j, if The first byte of the Key is selected Key[1];
the round key value is : Computing new SBOXkey[1] =
SBOXAESKey[1];
X"6172726167736C696D616E5F6D697469", Computing inverse SBOXkey[1]INV=INV
eqyivalent arragsliman_miti in code ascii. (SBOXkey[1])=INV (SBOXAESKey[1]);
Continue as described in the AES algorithm with
The first byte 61(Hex) is used to XOR-ing the Sbox set out in AES by using new computed
(change the sbox by using XOR operation between SBOXkey[1]INV.
each byte of fixe Sbox(showing in Table 1) and the
first byte of the master key). The resulting Table 4 : Comparaison Between AES Core And
SBOXxor [eache byte,61 hex] is used during the Our Suggestion.
Subbyte operation.
AES Dynamique
Case 2: AES (Our
Here different round keys are generated using a key suggestion)
expansion algorithm which is similar to that of AES Block Length 128-bit same
key expansion algorithm. The round keys thus Key Lenght 128-192-256 same
bits

200
 Journal of Theoretical and Applied Information Technology
th
20 July 2013. Vol. 53 No.2
2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

Number of For key length same Table 5 : I/O functional descriptions of proposed and
Rounds 128-bit. modified AES-128.
10 Rounds
Composed of 4 Composed of 4
transformations, transformations,
namely: namely:
Function
Rounf Function Pin name I/O
-ByteSub using -ByteSub using description
SBOX SBOXxor key I System
CLK
-Shift Row -Shift Row frequency
-Mix Column -Mix Column Rst I System reset
-AddRoundKey. -AddRoundKey. Plaintext bits
(for
For last round For last round Encryption)
Mix Column is Mix Column is Donner
I Encryption
eliminated eliminated bits (for
SBOX Fixed initial Key Decryption)
Dependent Key for
Key Expansion Use the master Use the master I Encryption
key and static key and Clef
or
SBOX Dynamic decryption
SBOXxor key Encryption
bits (for
Encryption)
Sortie
5. IMPLEMENTATION OF MODIFIED O Plaintext bits
(for
PIPELINED AES IN FPGA
Decryption)
Implementation uses the VHDL
programming language that nowadays is commonly Table 6: Comparative Table Between Different
a language used very established for FPGA [16]. implementation constitute AES algorithm
The drawing & the software of the simulation is
Quartus II v9.1.

cry ptage_aes_pip Implementation FPGA Device

donner[127..0] sortie[127..0] Peak Total Total

Total Total virtual registers memory
clef [127..0]
pins logic memory bits
clk elements Megabyte
rst Crypt_aes_pip 386 13552 232 2432 327680
Decry_aes_pip 386 15827 242 2432 327680
Crypt_aes_pip_ 386 57366 496 3840 0
SBOXxor key[i]
Figure 12: Cryptage AES 128 Pipeline
Decrypt_aes_pip_ 386 19471 379 2560 327680
SBOXxor key[i]
Figure 13: Decryptage AES 128 Pipeline
Note: in this different implementation we
decry ptage_aes_pip use two optimization:
Architectural optimization by using pipelined
donner[127..0] sortie[127..0] architecture of AES-128 [12], [13].
clef [127..0] Algoritmic optimization by using modified
clk structure of mixcolumn block (Properties of the
binary calculation) [26], [27].
rst
Depending on comparative table we can notice that
the first architectural of the Crypt_aes_pip
inst
implementation in. occupied more than (13552
units) of the unit when the second Decry_aes_pip
require implementing approximately in. (15827
units) in total capacity of the device, on the other

201
 Journal of Theoretical and Applied Information Technology
th
20 July 2013. Vol. 53 No.2
2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

the third architecture, Crypt_aes_pip_SBOXxor Ciphering :( figure15 )

key[i], need (57366 tranches) of the device, and last
implementation,Decrypt_aes_pip_SBOXxor key[i], Plaintext: hamdoun_&_tragha
occupied (19471 logic elements).
Key: arragsliman_miti
The first conclusion here ,the advantage in our new
implementation is the SBOX and key expansion is
now dependent on entire initial key ( first byte of cyphertext:,A![208]2[237]<W[160]Xl[218][225][1
master key( case 1) or Xor-ing between each byte 9]|[30]
of initial key (case 2)).
But the disadvantage is that it consumes little extra
time and more logic elements.
Second Conclusion we find our new implemention
(crypt_aes_pip_SBOXxor key[i] &
Derypt_aes_pip_SBOXxor key[i]) is more
efficacious than architecture of the first and second
implementation showing in table 5, the occupying
number of resources of the device.
Figure 15:Simulation Of Crypt_Aes_Pip_Sboxxor Key[I]
6. SIMULATION & INTERPRETATION
The schemas descripted of the simulation Deciphering :( figure 16)
the processes for the setting in implementation
crypt_aes_pip_SBOXxor key[i] (AES-128 based to plaintext :,A![208]2[237]<W[160]Xl[218][225][19
change the static sbox by using XOR operation ]|[30]
between each byte of fixe Sbox (showing in Tble
1and the first byte of the master key (case1) or Key: arraglsiman_miti
(case2) ) are presented below, in Figure (15). The
overall length of the encryption process is (126 S),
Cyphertext: hamdoun_&_tragha
otherwise the simulation of
decrypt_aes_pip_SBOXxor key[i] ( change the
sbox by using XOR operation between each byte of
fixe Sbox(showing in Table 1) and the first byte of
the initial key (case1) or by using case2 when we
Xor-ing each byte of masetr key), presented
following, in figure (16).encryption is the time to (
40 s) and some decoding (s).

Figure 16:Simulation Of Decrypt_Aes_Pip_Sboxxor

Key[I]

Note: during our implementation and simulation we

used two different software,the first is altered UP
Simulator and the second is Quartus II v9.1.

7. CONCLUSION
The first conclusion, in this paper new
substitution matrices have been developed by XOR
with byte key and chosen from existing AES S-
BOX. These matrices were tested with simulation
Figure 14 : Compile The Circuit software developed by Quartus II V.9.1. Analysis of
the results shows that the characteristics of the new
256 S-BOX are identical, based on which the
conclusion was reached that it is possible to use

202
 Journal of Theoretical and Applied Information Technology
th
20 July 2013. Vol. 53 No.2
2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

each for encryption. It will not lead to a [5] Olivier Frider ETR6 Advanced Encryption
deterioration of the stability of the AES linear System , cole dingnieurs du Canton de
cryptanalysis and differential. An algorithm for the Vaud, Mai 2004.
use of these matrices is proposed, as it is based on a [6] Ashwini M. D, Mangesh S. D and Devendra N.
preselected byte of the key used, and depending on K ,FPGA Implementation of AES Encryption
the result in XOR operation, one of the S-BOX 256 and Decryption, Proceeding of International
is selected. Conference On Control, Automation,
Second conclusion, we find the changes Communication And Energy Conservation -
proposed (Sbox dependent of initial key) in our 2009.
paper although consumes little extra time and more [7] Daemen J. and Rijmen V., Rijndael: The
logic elements, but can be implemented without Advanced Encryption Standard. Dr. Dobbs
changing the block size keys ( 128, 192 or 256). Journal, March 2001.
Even if the original AES algorithm is very secure, [8] NIST, DRAFT NIST Special Publication 800-
These proposed changes in the treatment of the 131, Recommendation for the Transitioning of
algorithm will encrypt the information by Cryptographic Algorithms and Key Sizes,
performing high diffusion and confusion. It also Federal Information Processing Standards
increases the complexity of the AES algorithm Publication (FIPS PUB) 197, National Institute
several times,so AES will be much stronger. of Standards and Technology (NIST), January,
2010.
8. POSSIBLE FUTURE WORK [9] Nikolai Stoianov, AES S-BOX generator:
analysis of requirements, International Science
From our perspective is to do: Conference 2009 Communication and
information systems, Shoumen, Bulgaria,2010.
Implementation and simulation the other optimize
structure and description based on that same [10] Qin H., Nonmember, SASAO T. and IGUCHI
principle of AES algorithm to encrypt and decrypt Y.,Members ,A Design of AES Encryption
data such as images and sounds. Circuit with 128 bit keys using Look-UP Table
Study of FPGA components and choosing the Ring on FPGA,IEICE TRANS. INF. &
most suitable architectures for the implementation SYST.,VOL.E89-D,NO.3 MARCH 2006.
of encryption algorithms. [11] Rahman T., Pan S. and Zhang Q., Design of a
A combination between the AES algorithm and High Throughput 128-bit (Rijndael Block
genetic algorithm to increase the security. Cipher), Proceeding of International
Implementation of the AES pipeline architecture Multiconferrence of Engineers and computer
to optimize the encryption and decryption time and scientists 2010 Vol II IMECS 2010, March 17-
number of clock cycles. 19,2010, Hongkong.
New instructions ensure a faster and more [12] Hodjat A. and Varbauwhede I.,A 21.54 Gbits
affordable encryption of data, as well as better Fully Pipelined AES Processor on FPGA,
security. IEEE Symposim on Field-Programmable
Custom Computing Machines,April 2004.
REFRENCES: [13] Jarvinen et al, A fully pipelined memoryless
17.8 Gbps AES-128 encrypter,International
Symposium on Field Programmable Gate
[1] NIST, Advanced Encryption Standard (AES),
arrays,pp.207-215.2003.
Springfield,VA, Nov. 2001
[14] INDECT Consortium, D8.2: Evaluation of
[2] Simon Singh : Histoire des codes secrets. De
Components, June, 2010, http://www.indect-
lEgypte des pharaons lordinateur quantique.
project.eu/files/deliverables/public/deliverable
Paru chez J.-C. Latts, Paris, 1999.
8.2.
[3] Stallings W. Cryptography and Network
[15] Rijndael mix column, available at:
Security: Principles and Practices.4th ed.
Pearson Education, Inc. pp. 63-173. 2006. http://en.wikipedia.org/wiki/Rijndael_mix_col
umns
[4] Bhupathi Kakarlapudi and Nitin Alabur, FPGA
Implementations of S-box vs. T-box iterative [16] Mroczkowski P., Implementation of the block
architectures of AES. cipher Rijndael using Altera FPGA, May 2000.
[17] Eli Biham and Nathan Keller, Cryptanalysis of
Reduced Variants of Rijndael, In Proceedings

203
 Journal of Theoretical and Applied Information Technology
th
20 July 2013. Vol. 53 No.2
2005 - 2013 JATIT & LLS. All rights reserved.

ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195

Of The Third Advanced Encryption Standard

Conference. NIST, April 2000.
[18] Zambreno J., Nguyen D. and Choudhary A.,
Exploring Area/Delay Tradeoffs in an AES
FPGA Implementation,FPL 2004, LNCS 3203,
pp. 575585, 2004.
[19] T. Jacobsen and L. R. Knudsen, The
Interpolation Attack on Block Ciphers, Fast
Software Encryption, LNCS 1267, E. Biham,
Ed., Springer Verlag, 1997, pp. 28 40.
[20] Kenny D., Energy Efficiency Analysis and
Implementation of AES on an FPGA,
University of Waterloo,2008.
[21] X.Ji-peng ,Z.Xue-cheng, G.Xu, Ultra-low
power S boxes architecture for AES, The
journal of China Universities of post and
telecommunications.vol.15,issue1,March 2008.
[22] M.Priya Zach, K.Rahimunnisa and K.Suresh
Kumar, Compact AES Architecture Using
Efficient S-Box Implementation, IEEE
International Conference on Computational
Intelligence and Computing Research, 1373-
1376, 2011.
[23] Nikolai Stoianov, One Approach of Using Key-
Dependent S-BOXes in AES, MCSS 2011,
CCIS 149, pp. 331337, 2011. Springer-Verlag
Berlin Heidelberg, 2011.
[24] William Stallings, Cryptography and Network
Security, Third Edition, Pearson Education,
2003.
[25] Edwin NC Mui, Practical Implementation of
Rijndael S-Box Using Combinational Logic.
[26] A. Sliman, H. Abdellatif, T.Abderrahim, K.
Salah eddine. Implementation of The
Encryption algorithm AES under VHDL
language In FPGA, by using different
architecture of mixcolumn. Call of paper
WOTIC'11 13-15 Octobre, ENSEM de
Casablanca.
[27] A. Sliman, H. Abdellatif, T.Abderrahim, K.
Salah eddine, Implementation of The
Encryption algorithm AES under VHDL
language In FPGA by using different
architecture of mixcolumn, International
Journal of VLSI design & Communication
Systems (VLSICS) Vol.3, No.4, August 2012.

204