DES Block Ciphering Based Genetic Biometric Keys - Ubiquitous Computing and Communication Journal

Document Sample
DES Block Ciphering Based Genetic Biometric Keys - Ubiquitous Computing and Communication Journal Powered By Docstoc
					Special Issue on Ubiquitous Computing Security Systems



                  DES Block Ciphering Based Genetic Biometric Keys

                                               Mofreh A. Hogo
    Electrical Engineering Technology Dept., Benha Higher Institute of Technology, Benha University, Egypt.
{Temporary Address: Information Sys. Dept., Faculty of Information Sys.& Computers Taif University, Taif, Saudi Arabia}
                                       Email: mofreh_hogo@hotmail.com


                                                  ABSTRACT
                E-business and sensitive Internet applications are growing very fast, so the
                needs to protect such applications are increased. Encryption algorithms play a
                main role in security. This work introduces a new key generation technique that
                can be used for enhancing the design of block cipher encryption algorithms,
                especially for DES to increase its key space; to be applicable in sensitive
                applications. This is based on the use of biometric key using the voice of
                speakers, extracting important features as formants and pitch of the speakers
                because it is unique for each one; in addition it applies crossover operations of
                genetic algorithm on biometric key to permute it. Finally it combines between
                stream ciphering and block ciphering algorithms (encryption of the crossover-
                biometric key using the RC4). The output Key-Stream bytes from the RC4 is
                then introduced to the DES algorithm to obtain the target ciphertext. The paper
                introduces also the idea of generating multi biometric keys by using different
                crossover operation types based on a specific time-slice. The paper introduces
                also the different steps for generating the Biometric-Crossover-Stream Key.
                Moreover the proposed DES provided added complexity due to the generation
                of new biometric key than normal DES; it is small, and can be neglected if one
                considers the trade-off between security and the complexity of the algorithm
                especially for the critical systems. The paper introduces a comparison between
                the proposed biometric keys and weak and semiweak keys; the biometric
                generated keys were neither semiweak nor weak keys so the generated keys are
                possibly acceptable.

                Key Words: Biometric Key, Genetic Algorithm, Key-Stream, Time-Slice,
                Formants, Pitch, RC4, DES, 3DES, and Key-Space.


1    INTRODUCTION                                             128-bit keys, while RC6 used various (128,192,256)
                                                              bit keys [1-5]. Asymmetric key encryption or public
     Many encryption algorithms are widely available          key encryption is used to solve the problem of key
and used in information security. They can be                 distribution. In Asymmetric keys, two keys are used;
categorized into Symmetric (private or one key) and           private and public keys. Public key is used for
Asymmetric (public or two keys) encryption. In                encryption and private key is used for decryption
Symmetric keys encryption or secret key encryption,           (E.g. RSA and Digital Signatures). Because users
only one key is used to encrypt and decrypt data. The         tend to use two keys: public key, which is known to
key should be distributed before transmission                 the public and private key which is known only to
between entities. Keys play an important role in the          the user, there is no need for distributing them prior
cryptography. If weak key is used in algorithm then           to transmission. However, public key encryption is
everyone may decrypt the data. Strength of                    based on mathematical functions, computationally
Symmetric key encryption depends on the size of               intensive and is not very efficient for small mobile
key used; algorithm using longer key is harder to             devices [1]. Asymmetric encryption techniques are
break than using smaller key. There are many                  almost 1000 times slower than Symmetric
examples of strong and weak keys of cryptography              techniques, because they require more computational
algorithms like RC2, DES, 3DES, RC6, Blowfish,                processing power [2]. DES: (Data Encryption
and AES. RC2 uses one 64-bit key. DES uses one                Standard), was the first encryption standard to be
64-bits key, triple DES (3DES or EDE) uses three              recommended by NIST (National Institute of
64-bits keys while AES uses various (128,192,256)             Standards and Technology). DES is (64 bits key size
bits keys. Blowfish uses various (32-448); default            with 64 bits block size); based on Federal Register




UbiCC Journal – Volume 4                                                                                           677
Special Issue on Ubiquitous Computing Security Systems



1973, the NBS issued a public request for proposals            The goal of this work is to introduce a novel
for a standard cryptographic algorithm with the           method for generating biometric keys that satisfy
following criteria: Algorithm must provide a high         enough large key space and introduce better security
level of security, completely specified and easy to       performance. The paper introduces a hybridization
understand, security of the algorithm must reside in      method for generating the biometric key by the use
the key; the security should not depend on the            of biometric voice keys, genetic algorithm
secrecy of the algorithm, must be available to all        operations, and combination between the RC4 and
users, adaptable for use in diverse applications,         3DES in order to increase the key space.
economically implementable in electronic devices,              The rest of the paper is organized as follows:
efficient to use, able to be validated, and exportable.   Section 2 reviews DES and RC4 encryption
Since that time, many attacks and methods recorded        algorithms. Section 3 introduces the biometric and
the weaknesses of DES, which made it an insecure          formants features extraction, selection and
block cipher [3-5]. The same DES algorithm and key        preprocessing. Section 4 introduces the Genetic
are used for both encryption and decryption (except       algorithm and the different types of crossover; as
for minor differences in the key schedule). The key       well as the comparison with the weak and semiweak
length is 56 bits. (The key is usually expressed as a     keys in DES. Section 5 introduces the design of
64-bit number, but every eighth bit is used for parity    genetic biometric keys. Section 6 introduces the
checking and is ignored. These parity bits are the        encryption of the biometric-crossover key using RC4
least-significant bits of the key bytes.) The key can     and the results analysis. Finally section 7 is reserved
be any 56-bit number and can be changed at any            for the conclusion and the future work.
time. Although it is showing signs of old age, it has
held up remarkably well against years of                  2   THEORETICAL REVIEW OF THE DES
cryptanalysis and is still secure against all but             AND RC4 ALGORITHMS
possibly the most powerful of adversaries. A handful
of numbers are considered weak keys, but they can              DES operates on a 64-bit block of plaintext.
easily be avoided; where all security rests within the    After an initial permutation, the block is broken into
key. At its simplest level, the algorithm is nothing      a right half and a left half, each 32 bits long. Then
more than a combination of the two basic techniques       there are 16 rounds of identical operations, called
of encryption: confusion and diffusion. The               Function f, in which the data are combined with the
fundamental building block of DES is a single             key. After the sixteenth round, the right and left
combination of these techniques (a substitution           halves are joined, and a final permutation (the
followed by a permutation) on the text, based on the      inverse of the initial permutation) finishes off the
key. This is known as a round. DES has 16 rounds; it      algorithm. In each round the key bits are shifted, and
applies the same combination of techniques on the         then 48 bits are selected from the 56 bits of the key,
plaintext block 16 times; the algorithm uses only         the right half of the data is expanded to 48 bits via an
standard arithmetic and logical operations on             expansion permutation, combined with 48 bits of a
numbers of 64 bits at most, so it was easily              shifted and permuted key via an XOR operation and
implemented in late 1970s hardware technology. The        sent through 8 S-boxes producing 32 new bits, and
repetitive nature of the algorithm makes it ideal for     permuted again. These four operations make up
use on a special-purpose chip. Initial software           Function f. The output of Function f is then
implementations were clumsy, but current                  combined with the left half via another XOR. The
implementations are better.                               result of these operations becomes the new right half;
3DES is an enhancement of DES; it is 64 bit block         the old right half becomes the new left half. These
size with 192 bits key size. In this standard the         operations are repeated 16 times, making 16 rounds
encryption method is similar to the one in the            of DES. If Bi is the result of the ith iteration, Li and Ri
original DES but applied 3 times to increase the          are the left and right halves of Bi, Ki is the 48-bit key
encryption level and the average safe time. It is a       for round i, and f is the function that does all the
known fact that 3DES is slower than other block           substituting and permuting and XORing with the
cipher methods [3]. AES is a block cipher having          key, then a round looks like:
variable key length of 128, 192, or 256 bits; default
                                                               Li = Ri-1, and Ri = Li-1 f (Ri-1, Ki).
256. It encrypts data blocks of 128 bits in 10, 12 and
14 rounds depending on the key size. AES                  2.1. Decrypting DES
encryption is fast and flexible; it can be implemented        After all the substitutions, permutations, XORs,
on various platforms especially in small devices [6].     and shifting around, on the contrary, the various
AES has been carefully tested for many security           operations were chosen to produce a very useful
applications [3], [7,8]. Performance evaluation of        property: The same algorithm works for both
symmetric encryption algorithms is introduced             encryption and decryption. With DES it is possible to
in [9].                                                   use the same function to encrypt or decrypt a block.
                                                          The only difference is that the keys must be used in




UbiCC Journal – Volume 4                                                                                        678
Special Issue on Ubiquitous Computing Security Systems



the reverse order. That is, if the encryption keys for   the X9.17 and ISO 8732 standards [11,12]. K1 and
each round is K1 K2 K3... K16 then the decryption keys   K2 alternate to prevent the meet-in-the-middle attack
are K16 K15 K14, ..., K1.                                previously described. If C = EK2(EK1 (EK1(P))),
                                                         then a cryptanalyst could precompute EK1(EK1(P)))
2.2. Double Encryption                                   for every possible K1 and then proceed with the
    A simple way of improving the security of a          attack. It only requires 2n + 2 encryptions. Triple
block algorithm is to encrypt a block twice with two     encryption with two keys is not susceptible to the
different keys. First encrypt a block with the first     same meet-in-the-middle attack described earlier.
key, and then encrypt the resulting ciphertext with      But Merkle and Hellman developed another time-
the second key. Decryption is the reverse process.       memory trade-off that could break this technique in
                                                         2n - 1 steps using 2n blocks of memory [13]. For
    C = EK2(EK1(P)) then P=DK1(DK2(C));                  each possible K2, decrypt 0 and store the result in
the resultant doubly-encrypted ciphertext block          memory. Then, decrypt 0 with each possible K1 to
should be much harder to break using an exhaustive       get P. Triple-encrypt P to get C, and then decrypt C
search. Instead of 2n attempts (where n is the bit       with K1. If that decryption is a decryption of 0 with a
length of the key), it would require 22n attempts. If    K2 (stored in memory), the K1 K2 pair is a possible
the algorithm is a 64-bit algorithm, the doubly-         candidate. The earliest standard that defines the
encrypted ciphertext would require 2128 attempts to      algorithm (ANS X9.52, published in 1998) describes
find the key.                                            it as the "Triple Data Encryption Algorithm
                       Ciphertext                        (TDEA)" i.e. three operations of the Data Encryption
                                                         Algorithm specified in ANSI X3.92 and does not use
                                                         the terms "Triple DES" or "DES" at all. FIPS PUB
                                                         46-3 (1999) defines the "Triple Data Encryption
            DES           DES-1      DES                 Algorithm (TDEA)", but also uses the terms "DES"
                                                         and "Triple DES". The encryption algorithm is:
                                                         Ciphertext = EK3(DK2(EK1(plaintext))); i.e., DES
Plaintext                                                encrypt with K1, DES decrypt with K2, then DES
             K1             K2        K1 Ciphertext      encrypt with K3. Decryption is the reverse:
                                                              Plaintext = DK1(EK2(DK3(ciphertext)));
                  -1                     -1
            DES            DES       DES                  i.e., decrypt with K3, encrypt with K2, then decrypt
                                                         with K1. Each triple encryption encrypts one block of
                                                         64 bits of data. In each case the middle operation is
                                                         the reverse of the first and last. The standards define
                       Ciphertext                        three keying options [14]:
                                                         1. Keying option 1: All three keys are independent
      Figure 1: Triple-des block diagram                      the strongest, with 3 x 56 = 168 independent key
2.3. Triple Encryption                                        bits.
Triple Encryption with Two Keys a better idea,           2. Keying option 2: K1 and K2 are independent, and
proposed by Tuchman in [10], operates on a block              K3 = K1; provides less security, with 2 x 56 = 112
three times with two keys: with the first key, then           key bits is stronger than simply DES encrypting
with the second key, and finally with the first key           twice, e.g. with K1 and K2, because it protects
again. He suggested that the sender first encrypt with        against meet-in-the-middle attacks.
the first key, then decrypt with the second key, and     3. Keying option 3: All three keys are identical, i.e.
finally encrypt with the first key. The receiver              K1= K2= K. is no better than DES, with only 56
decrypts with the first key, then encrypts with the           key bits.
second key, and finally decrypts with the first key.     This option provides backward compatibility with
C = EK1(DK2(EK1(P))); P = DK1(EK2(DK1(C)))               DES, because the first and second DES operations
This is sometimes called encrypt-decrypt-encrypt         simply cancel out. It is no longer recommended by
(EDE) mode [11]. Figure 1 shows the block diagram        the National Institute of Science and Technology
of the 3DES. If the block algorithm has an n-bit key,    (NIST) and not supported by ISO/IEC 18033-3.
then this scheme has a 2n-bit key. The curious           In general Triple DES with three independent keys
encrypt-decrypt-encrypt pattern was designed by          (keying option 1) has a key length of 168 bits (three
IBM to preserve compatibility with conventional          56-bit DES keys), but due to the meet-in-the-middle
implementations of the algorithm: Setting the two        attack the effective security it provides is only 112
keys equal to each other is identical to encrypting      bits. Keying option 2, reduces the key size to 112
once with the key. There is no security inherent in      bits. However, this option is susceptible to certain
the encrypt-decrypt-encrypt pattern, but this mode       chosen-plaintext or known-plaintext attacks[15][16]
has been adopted to improve the DES algorithm in




UbiCC Journal – Volume 4                                                                                   679
Special Issue on Ubiquitous Computing Security Systems



and thus it is designated by NIST to have only 80        phase, for each value i from zero to (N-1), the KSA
bits of security[17].                                    calculates a value for j by adding (modulo N) the
                                                         previous value of j, the ith element of S and the ith
2.4. DES/3DES: Applications
                                                         element of K (the IV prepended to the secret key)
  In general, cryptography is used to protect data
                                                         modulo l, the length of K. Finally, the values of S[i]
while it is being communicated between two points
                                                         and S[j] are swapped. This is repeated for every
or while it is stored in a medium vulnerable to
                                                         element in the S array. The final product is an array,
physical theft. Communication security provides
                                                         S, that is the same length as the plaintext/CRC
protection to data by enciphering it at the
                                                         combination and is scrambled using the secret key as
transmitting point and deciphering it at the receiving
                                                         an index offset for each swap. The scrambled S array
point. File security provides protection to data by
                                                         is then fed into the PRGA and another succession of
enciphering it when it is recorded on a storage
                                                         N swaps occur, however each time these swaps
medium and deciphering it when it is read back from
                                                         occur, an output value is calculated. These output
the storage medium. In the first case, the key must be
                                                         values (z) are the ultimate keystream bytes that will
available at the transmitter and receiver
                                                         be used to encrypt the plaintext data. Table 1 shows
simultaneously during communication. In the second
                                                         the c code paraphrasing the operation of the detailed
case, the key must be maintained and accessible for
                                                         RC4 algorithm.
the duration of the storage period. FIPS 171 provides
approved methods for managing the keys used by the           Table 1: RC4 algorithm
algorithms specified in this standard. Applications of
                                                             KSA(K)                                   PRGA(K)
3DES can be found everywhere, where data has to be
                                                             Initialization:               Initialization:
secured. Data storage and networking are centers of
                                                              For i = 0 ... N - 1           i=0
interest; DES/3DES is used for:
1. Virtual Private Networking (VPN): In Secure IP               S[i] = i                    j=0
(IPsec)-based implementations, in SSH(secure shell)-          j=0                          Generation Loop:
based implementations, and in SSH2-based                     Scrambling:                    i =i +1
implementations                                               For i = 0 ... N - 1           j = j + S[i]
2. DES/3DES is used for secure extranet protocols,             j = j + S[i] + K[i mod l]    Swap(S[i], S[j])
such as Citrix Extranet or European Network                     Swap(S[i], S[j])            Output z = S[S[i] + S[j]]
Exchange.
3. DES/3DES is used for secure Storage devices,          3   FORMANTS & BIOMETRIC SECURITY
such as Smartcards. A prominent example for the              METRIC
species is “SIM Application Toolkit”.
4. XML encryption to secure E-Government is                  A formant is a peak in an acoustic frequency
dealing with DES/3DES.                                   spectrum which results from the resonant frequencies
5. E-commerce payment transactions.                      of any acoustic system. It is most commonly invoked
                                                         in phonetics or acoustics involving the resonance
2.5. RC4: Stream Ciphering Algorithm                     frequencies of the vocal tract or musical instruments
    RC4 encryption is the encryption used in most         [19]. Formants are distinguishing or meaningful
software applications today (to include HTTPS and        frequency components to human speech. A speech
SSL), which makes it arguably the most commonly          sound wave does not actually travel through the
used encryption method in the world. No wonder           vocal tract and out into the air. Rather, the air in the
then, when the architects of wireless network            vocal tract behaves like a spring that vibrates back
security sat down to design their protocol in 1997,      and forth in standing wave. Resonant frequencies
they incorporated RC4 into their design; in the form     match the frequencies of the waves that will fit the
of WEP (Wired Equivalent Privacy). The RC4               tube. The general form for calculating nth formants
consists mainly of two main steps; the Key-              Fn presented in [20] is:
Scheduling Algorithm (KSA) and the Pseudo-
                                                                          Fn = (2*N-1) * C /4L;
Random Generation Algorithm (PRGA) [18]. The
ultimate product of the KSA and PRGA is the              where: N = resonance number, C = Speed of sound,
keystream that is XORed with the plaintext data and      and L = length of the vocal tract. By definition, the
CRC combination to produce the encrypted                 information that humans require to distinguish
cyphertext ready for transmission. This process          between vowels can be represented quantitatively by
begins by creating an array of values (S[ ]) of a        the frequency content of the vowel sounds. The
length (N) equal to that of the length of the            formants with the lowest frequency F1, F2, F3,F4 are
plaintext/CRC combination. This array is initialized     enough to disambiguate the vowel. The first two
by setting the values of each element in the array to    formants, F1, F2 are primarily determined by the
be equal to the corresponding index of the element       position of the tongue F1 has a higher frequency
(Ex: S[0]=0, S[1]=1, …, S[N]=N). In its scrambling       when the tongue is lowered and F2 has higher




UbiCC Journal – Volume 4                                                                                        680
         Special Issue on Ubiquitous Computing Security Systems



                                                                  typical vowel based ARPABET[23]. The work in
     Table 2: Formant frequencies for typical vowels              [24] presented the use of formants and its arc tan for
         ARPABET                                                  key generation.
                     IPA       Typical
         Symbol for                        F1   F2       F3
                    Symbol      Word
           vowel                                                  3.1. Experiments For The Proposed Formants
            IY         /i/      Beet     270    2290    3010          Extraction and Normalization
            IH         /I/       Bit     390    1990    2550          Experiments were done for formants extraction
            EH        /ε/       Bet      530    1840    2480      using the praat software tool to extract the different
            AE        /æ/       Bat      660    1720    2410      formants features and analyze the biometric voice
            AH        /Λ/       But      520    1190    2390      key. The steps for extracting the first four formants
            AA        /a/       Hot      730    1090    2440
                                                                  frequencies and F0 are as following:
            AO        /c/      Bought    570     840    2410
            UH        /U/       Foot     440    1020    2240
                                                                  • Capture the speaker speech, the preprocessing
            UW        /u/       Boot     300     870    2240      step to extract the following features from the
            ER         /з/      bird     490    1350    1690      speaker voice : F1: First Formant, F2:Second
                                                                  Formant, F3:Third Formant, F4:Fourth Formant,
                                                                  and Pitch F0. Table 3 shows the results from using
                                                                  praat on different vowels. The results obtained are
                                                                  completely different from each other and
                                                                  distinguishable easily; there is no similarity between
F2(HZ)




                                                                  them and the differences between them are big
                                                                  enough to discriminate between them as shown in
                                                                  Figure 3.
                                                                  • Convert the selected formants from the decimal
                                                                  values to the corresponding ASCII code. Only 8
                                                                  digits are selected including both of the integer and
                                                                  the fraction to constitute the 64-bit value (consider
                              F1(HZ)
                                                                  the integer digits and complete the 8 digit from the
          Figure 2: Wide banded formants of different             fraction); Tables 4,5 show the results of extracted
          sounds                                                  formants after the conversion to ASCII
            Table 3: Comparison between two biometric             representation as 64 bits.
            Formants             ‘e’                   ‘a’        • The speakers are changed and swapped each time
             Time_s            0.929601            0.563084       slice, as another security concept in Key generation
                                                                  step. The results from this step are represented by
             F0_Hz            119.575946          104.500974
                                                                  the different fromants frequencies from F1 to F4 and
             F1_Hz           1209.319034         1009.009646      the pitch value F0, as shown in Figures 4,5; the
             F2_Hz           3161.302028         3448.814559      results obtained are completely discriminated.
             F3_Hz           5788.546047         4693.620730
             F4_Hz           8143.632944         8369.372678      3.2.To make identification or matching for
                                                                  unknown speaker
     frequency when the tongue is forward. Each formant           The following algorithm is implemented:
     corresponds to a resonance in the vocal tract. Vowels
     will almost repetitive sound is distinguished by             1. Read a   sound file and store it into a vector.
     virtue of a very low third formant (below 2000 Hz).          2. Divide    the signal in time domain into 10 equal
     have four or more distinguishable formants.                     blocks. Select the block with maximum power
     Formants move about in a range of approximately                 content, normalize the selected block.
     1000 Hz for a male adult. Nasals usually have an             3. Determine the formants F1 to F4, and F0.
     additional formant around 2500 Hz. The liquid                4. Calculate the Euclidean distance between this set
     usually has an extra formant at 1500 Hz, while the              of formants frequencies obtained from the speaker.
     Plosives modify the placement of formants in the             5. Calculate the Euclidean distance between the set of
     surrounding vowels. Bilabial sounds cause a                     frequencies obtained from the User, and each of the
     lowering of the formants. Alveolar sounds cause less            set of frequencies corresponding to the five vowels.
     systematic changes in neighboring vowel formants,            6. The minimum distance criterion is used for
     depending partially on exactly which vowel is                   decision making to determine the vowel. Figure
     present. The time-course of these changes in vowel              6(a) shows the structure of speaker’ identification
     formant frequencies are referred to as formant                  system; while Figure 6(b) shows the way of
     transitions [21,22]. Each vowel can be placed on a              matching process; where the minimum distance
     graph, where F1, F2 are represented on the Figure 2             with database elements identify the speaker
     shows the wide banded formants of different sounds,             exactly.
     while Table 2 illustrates the formants frequencies for




         UbiCC Journal – Volume 4                                                                                   681
Special Issue on Ubiquitous Computing Security Systems




                                                 Resulting Signal                                                             Resulting Signal
                        -2                                                                            -3
                       10                                                                            10

                        -3
                       10                                                                             -4
                                                                                                     10

                        -4
                       10
                                                                                                      -5
                                                                                                     10
                        -5
                       10
         3d b P ower




                                                                                       3d b P ower
                                                                                                      -6
                                                                                                     10
                        -6
                       10
                                                                                                      -7
                                                                                                     10
                        -7
                       10

                                                                                                      -8
                        -8                                                                           10
                       10


                        -9                                                                            -9
                       10                                                                            10
                             0   500   1000    1500      2000   2500   3000   3500   4000                  0   500   1000   1500      2000   2500   3000   3500   4000
                                                      Frequency                                                                    Frequency


                                   Figure 3: Power against the formants frequencies for the two letters 'a', and 'e'




                                              Figure 4: Formant, power, pitch value and spectrum for 'a'




                                              Figure 5: Formant, power, pitch value and spectrum for 'e'




UbiCC Journal – Volume 4                                                                                                                                                 682
Special Issue on Ubiquitous Computing Security Systems



                                Features                           11001011+11011111 = 11001111
                               extraction                     • Two point crossover
                                             Decision    two crossover point are selected, binary string from
                                classifier               beginning of chromosome to the first crossover point
                                                         is copied from one parent, the part from the first to
                                 Dbase                   the second crossover point is copied from the second
                                                         parent and the rest is copied from the first parent.
         Figure 6(a): Voice matching

                                                                  11001011 + 11011111 = 11011111

                                                              • Uniform crossover
                                                         bits are randomly copied from the first or from the
                                                         second parent.



                                                                   11001011 + 11011101 = 1101111
     Figure 6(b): Matching of unknown speaker
                                                            • Arithmetic crossover
4   GENERATION OF BIOMETRIC KEYS                         some arithmetic operation is performed to make a
    USING GA (CROSSOVER)                                 new offspring.

     A genetic algorithm (GA) is a search technique
used in computing to find exact or approximate
solutions to optimization and search problems            11001011 + 11011111 = 11001001 (AND)
[25][26]. Genetic algorithms are categorized as
global search heuristics. Genetic algorithms are a       5   Design of Genetic Biometric Keys
particular class of evolutionary algorithms (also
known as evolutionary computation) that use                   The crossover operations are implemented using
techniques inspired by evolutionary biology such as      c++ to find the different types of crossover
inheritance, mutation, selection, and crossover (also    operations. The different combinations of F0 to F4
called recombination). GA steps are listed below:        are the inputs and the output samples for each type as
Choose Initial Population                                following in Tables 6 (a) to (e).
Evaluate Each Individual's Fitness
Determine Population's Average Fitness                   5.1 Crossover Biometric Keys Versus Weak And
Repeat                                                   Semiweak Keys
Select Best-Ranking Individuals To Reproduce
                                                              In cryptography, a weak key is a key; which
Mate Pairs At Random
Apply Crossover Operator                                 when used with a specific cipher, makes the cipher
Apply Mutation Operator                                  behave in some undesirable way. Weak keys usually
Evaluate Each Individual's Fitness                       represent a very small fraction of the overall key
Determine Population's Average Fitness                   space, which usually means that A cipher with no
Until Terminating Condition (E.G. Until At Least One     weak keys is said to have a flat, or linear, key space.
Individual Has The Desired Fitness Or Enough             The block cipher DES has a few specific keys termed
Generations Have Passed).                                "weak keys" and "semi-weak keys". These are keys
The complete steps of GA are not needed in this          which cause the encryption mode of DES to act
work; just we apply the crossover operations in its      identically to the decryption mode of DES (albeit
various forms. There are many crossover techniques       potentially that of a different key). In operation, the
exist for organisms; which use different data            secret 56-bit key is broken up into 16 subkeys
structures to store themselves.                          according to the DES key schedule; one subkey is
     • Single point crossover                            used in each of the sixteen DES rounds[27][28].
One crossover point is selected, binary string from           The weak keys of DES are those which produce
beginning of chromosome to the crossover point is        sixteen identical subkeys. To ensure that the
copied from one parent, and the rest is copied from      extracted biometric keys are neither semiweak nor
the second parent. The following example illustrates     weak keys; and to complete the proposed system
how this type of crossover works:                        correctly; a comparison between Tables 16,17,and 18
                                                         is done by a simple matching software. The matching




UbiCC Journal – Volume 4                                                                                   683
Special Issue on Ubiquitous Computing Security Systems




Table 4: First example of biometric formants
    Formants for 'e'                           ASCII code formants 64-bit binary equivalent key
F0_Hz 119.575946 0011 0001 0011 0001 0011 1001 0011 0101 0011 0111 0011 0101 0011 1001 0011 0100
F1_Hz 1209.319034 0011 0001 0011 0010 0011 0000 0011 1001 0011 0011 0011 0001 0011 1001 0011 0000
F2_Hz 3161.302028 0011 0011 0011 0001 0011 0110 0011 0001 0011 0011 0011 0000 0011 0010 0011 0000
F3_Hz 5788.546047 0011 0101 0011 0111 0011 1000 0011 1000 0011 0101 0011 0100 0011 0110 0011 0000
F4_Hz 8143.632944 0011 1000 0011 1001 0011 1100 0011 0011 0011 0110 0011 0011 0011 0010 0011 1001


Table 5: Second example of biometric formants
    Formants for 'a’                             ASCII code formants 64-bits equivalent key
F0_Hz 104.500974 0011 0001 0011 0000 0011 0100 0011 0101 0011 0000 0011 0000 0011 1001 0011 0111
F1_Hz 1009.009646 0011 0001 0011 0000 0011 0000 0011 1001 0011 0000 0011 0001 0011 0000 0011 1001
F2_Hz 3448.814559 0011 0011 0011 0100 0011 0100 0011 1000 0011 1000 0011 0001 0011 0100 0011 1001
F3_Hz 4693.620730 0011 0100 0011 0110 0011 1001 0011 0011 0011 0110 0011 0010 0011 0000 0011 0111
F4_Hz 8369.372678 0011 1000 0011 0011 0011 0110 0011 1001 0011 0011 0011 0111 0011 0010 00110110

    between these tables was found to be zero                   varying based on time slice; as for each category of
matches. It means that; the extracted biometric key             the crossover generated key shown in Table 6; and
and the crossover operation done on these keys                  for each Time-Slice Ti the inputs to the RC4 are the
provide possible or strong keys, thus avoiding the              plaintext selected from a specific type of crossover
semiweak and weak keys in DES.                                  operation from Table 6 (i.e., say from Table 6 (a)),
                                                                and the Key used in RC4 will be selected from
6     ENCRYPTION OF THE BIOMETRIC KEY                           another type of crossover operation from Table 6
      WITH RC4 AND RESULTS ANALYSIS                             (i.e., say from Table 6(b)). In the next Time-Slice
                                                                Ti+1 the inputs to the RC4 will be swapped; as the
This step introduces a new combination between two              input will be from Table 6 (b), and the key will be
different symmetric cryptography algorithms; the                selected from Table 6 (a). The following section
stream cipher (RC4) and the block cipher (DES); to              introduces an illustrative example of the key-
guarantee security and increase the key space. The              Generation of the crossover biometric formants key
idea of using RC4 aimed at increasing Key-Space.                using the RC4
The approach of applying the RC4 stream cipher is               .

    Table 6: The possible crossover of different keys for speaker 'e'
    Table 6(a): Single point crossover : 40-24
          F0           0011 0001   0011 0001   0011 1001   0011 0101   0011 0111   0011 0101   0011 1001   0011 0100
          F1           0011 0001   0011 0000   0011 0000   0011 1001   0011 0000   0011 0001   0011 0000   0011 1001
     F0-F1: Key1       0011 0001   0011 0001   0011 1001   0011 0101   0011 0111   0011 0001   0011 0000   0011 1001
     F0-F1: ASCII          1           1           9           5           7           1           0           9
          F0           0011 0001   0011 0001   0011 1001   0011 0101   0011 0111   0011 0101   0011 1001   0011 0100
          F2           0011 0011   0011 0100   0011 0100   0011 1000   0011 1000   0011 0001   0011 0100   0011 1001
     F0-F2: Key2       0011 0001   0011 0001   0011 1001   0011 0101   0011 0111   0011 0001   0011 0100   0011 1001
     F0-F2: ASCII          1           1           9           5           7           1           4           9
          F0           0011 0001   0011 0001   0011 1001   0011 0101   0011 0111   0011 0101   0011 1001   0011 0100
          F3           0011 0100   0011 0110   0011 1001   0011 0011   0011 0110   0011 0010   0011 0000   0011 0111
     F0-F3: Key3       0011 0001   0011 0001   0011 1001   0011 0101   0011 0111   0011 0010   0011 0000   0011 0111
     F0-F3: ASCII          1           1           9           5           7           2           0           7
          F0           0011 0001   0011 0001   0011 1001   0011 0101   0011 0111   0011 0101   0011 1001   0011 0100
          F4           0011 1000   0011 0011   0011 0110   0011 1001   0011 0011   0011 0111   0011 0010   0011 0110
     F0-F4: Key4       0011 0001   0011 0001   0011 1001   0011 0101   0011 0111   0011 0111   0011 0010   0011 0110
     F0-F4: ASCII          1           1           9           5           7           7           2           6




UbiCC Journal – Volume 4                                                                                          684
Special Issue on Ubiquitous Computing Security Systems




 Table 6(b):Two point crossover: 16-32-16

       F1        0011 0001   0011 0000   0011 0000   0011 1001   0011 0000   0011 0001   0011 0000   0011 1001
       F0        0011 0001   0011 0001   0011 1001   0011 0101   0011 0111   0011 0101   0011 1001   0011 0100
  F1-F0: Key5    0011 0001   0011 0000   0011 1001   0011 0101   0011 0111   0011 0101   0011 0000   0011 1001
  F1-F0: ASCII       1           0           9           5           7           5           0           9
       F1        0011 0001   0011 0000   0011 0000   0011 1001   0011 0000   0011 0001   0011 0000   0011 1001
       F2        0011 0011   0011 0100   0011 0100   0011 1000   0011 1000   0011 0001   0011 0100   0011 1001
  F1-F2: Key6    0011 0001   0011 0000   0011 0100   0011 1000   0011 1000   0011 0001   0011 0000   0011 1001
  F1-F2: ASCII       1           0           4           8           8           1           0           9
       F1        0011 0001   0011 0000   0011 0000   0011 1001   0011 0000   0011 0001   0011 0000   0011 1001
       F3        0011 0100   0011 0110   0011 1001   0011 0011   0011 0110   0011 0010   0011 0000   0011 0111
  F1-F3: Key7    0011 0001   0011 0000   0011 1001   0011 0011   0011 0110   0011 0010   0011 0000   0011 1001
  F1-F3: ASCII       1           0           9           3           6           2           0           9
       F1        0011 0001   0011 0000   0011 0000   0011 1001   0011 0000   0011 0001   0011 0000   0011 1001
       F4        0011 1000   0011 0011   0011 0110   0011 1001   0011 0011   0011 0111   0011 0010   0011 0110
  F1-F4: Key8    0011 0001   0011 0000   0011 0110   0011 1001   0011 0011   0011 0111   0011 0000   0011 1001
  F1-F4: ASCII       1           0           6           9           3           7           0           9


 Table 6(c): Uniform crossover: 4-4-8-8-4-4-8-8-4-4-4-4
       F2      0011 0011 0011 0100 0011 0100 0011 1000 0011 1000 0011 0001               0011 0100   0011 1001
       F0      0011 0001 0011 0001 0011 1001 0011 0101 0011 0111 0011 0101               0011 1001   0011 0100
  F2-F0: Key9   00110001 0011 0100 0011 1001 00110101 0011 1000 0011 0101                00111001    00111001
  F2-F0: ASCII     1         4         9         5         8         5                       9           9
       F2      0011 0011 0011 0100 0011 0100 00111000 0011 1000 0011 0001                0011 0100   0011 1001
       F1      0011 0001 0011 0000 0011 0000 0011 1001 0011 0000 0011 0001               0011 0000   0011 1001
  F2-F1: Key10 00110001 0011 0100 0011 0000    1001    0011 1000 0011 0001               0011 0000   00111001
  F2-F1: ASCII     9         4         0         9         8         1                       0           9
       F2      0011 0011 0011 0100 0011 0100 0011 1000 0011 1000 0011 0001               0011 0100   0011 1001
       F3      0011 0100 0011 0110 0011 1001 0011 0011 0011 0110 0011 0010               0011 0000   0011 0111
  F2-F3: Key11 0011 0100 0011 0100 0011 1001 00110011 0011 1000 0011 0010                00110000    00111001
  F2-F3: ASCII     4         4         9         3         8         2                       0           9
       F2      0011 0011 0011 0100 0011 0100 0011 1000 0011 1000 0011 0001               0011 0100   0011 1001
       F4      0011 1000 0011 0011 0011 0110 0011 1001 0011 0011 0011 0111               0011 0010   0011 0110
  F2-F4: Key12 00111000 0011 0100 0011 0110 00111001 0011 1000 0011 0111                 00110010    00111001
  F2-F4: ASCII     8         4         6         9         8         7                       2           9




 Table 6(d): Uniform crossover: 8-8
       F3        0011 0100   0011 0110   0011 1001   0011 0011   0011 0110   0011 0010   0011 0000   0011 0111
       F0        0011 0001   0011 0001   0011 1001   0011 0101   0011 0111   0011 0101   0011 1001   0011 0100
  F3-F0: Key13   00110001    0011 0110   0011 1001   00110101    0011 0110   0011 0101   0011 1001   00110111
  F3-F0: ASCII       1           6           9           5           6           5           9           7
       F3        0011 0100   0011 0110   0011 1001   0011 0011   0011 0110   0011 0010   0011 0000   0011 0111
       F1        0011 0001   0011 0000   0011 0000   0011 1001   0011 0000   0011 0001   0011 0000   0011 1001
  F3-F1: Key14   0011 0100   0011 0000   0011 1001   0011 1001   0011 0110   0011 0001   0011 0000   0011 1001
  F3-F1: ASCII       4           0           9           9           6           1           0           9
       F3        0011 0100   0011 0110   0011 1001   0011 0011   0011 0110   0011 0010   0011 0000   0011 0111
       F2        0011 0011   0011 0100   0011 0100   0011 1000   0011 1000   0011 0001   0011 0100   0011 1001
  F3-F2: Key15   0011 0100   0011 0100   0011 1001   0011 1000   0011 0110   0011 0001   0011 0000   0011 1001
  F3-F2: ASCII       4           4           9           8           6           1           0           9
       F3        0011 0100   0011 0110   0011 1001   0011 0011   0011 0110   0011 0010   0011 0000   0011 0111
       F4        0011 1000   0011 0011   0011 0110   0011 1001   0011 0011   0011 0111   0011 0010   0011 0110
  F3-F4: Key16   0011 0100   0011 0011   0011 1001   0011 1001   0011 0110   0011 0111   0011 0000   0011 0110
  F3-F4: ASCII       4           3           9           9           6           7           0           6




UbiCC Journal – Volume 4                                                                                    685
Special Issue on Ubiquitous Computing Security Systems



 Table 6(e) :Uniform crossover: 8-8
       F4          0011 1000 0011 0011 0011 0110 0011 1001 0011 0011 0011 0111 0011 0010 0011 0110
       F0          0011 0001 0011 0001 0011 1001 0011 0101 0011 0111 0011 0101 0011 1001 0011 0100
  F4-F0: Key17     0011 1000 0011 0001 0011 0110 0011 0101 0011 0011 0011 0101 0011 0010 0011 0100
  F4-F0: ASCII             8        1           6         5             3            5         2             4
       F4          0011 1000 0011 0011 0011 0110 0011 1001 0011 0011 0011 0111 0011 0010 0011 0110
       F1          0011 0001 0011 0000 0011 0000 0011 1001 0011 0000 0011 0001 0011 0000 0011 1001
  F4-F1: Key18     0011 1000 0011 0000 0011 0110 0011 1001 0011 0011 0011 0001 0011 0010 0011 1001
  F4-F1: ASCII             8        0           6         9             3            1         2             9
      F4-F2        0011 1000 0011 0011 0011 0110 0011 1001 0011 0011 0011 0111 0011 0010 0011 0110
       F2          0011 0011 0011 0100 0011 0100 0011 1000 0011 1000 0011 0001 0011 0100 0011 1001
  F4-F2: Key19     0011 1000 0011 0100 0011 0110 0011 1000 0011 0011 0011 0001 0011 0010 0011 1001
  F4-F2: ASCII             8        4           6         8             3            1         2             9
       F4          0011 1000 0011 0011 0011 0110 0011 1001 0011 0011 0011 0111 0011 0010 0011 0110
       F3          0011 0100 0011 0110 0011 1001 0011 0011 0011 0110 0011 0010 0011 0000 0011 0111
  F4-F3: Key20     0011 1000 0011 0110 0011 0110 0011 0011 0011 0011 0011 0010 0011 0010 0011 0111
  F4-F3: ASCII             8        6           6         3             3            2         2             7


     Table 7: Encryption of ASCII(crossover) biometric keys using RC4:
     Example of biometric crossover formants F0-F1 using the F1-F0,F1-F2,F1-F3,F1-F4 keys
                                        F1-F0: Key       F1-F2: Key            F1-F3: Key     F1-F4: Key
              Selected Key
                                        10957509          10488109              10936209       10693709
        F0-F: Key: 111957109            ’ÓIŸnµvQ         Ê@_…ÁÕ"_              _ i_«GÚØ       S_®__ÙÞ_




              Speaker
            Interchange



       Formants Extraction      64-Bit Digitizer Based
      F1,F2,F3,F4, …and F0     ASCII Code Conversion                                          16-IP & F(R,L)



           Formants,            Formant Interchanging
          Selection ,and       (Crossover Computation)        RC4 Encryption       16-Key
                                                                Algorithm                          F (R,K)
           Normalize               K1, K2,K3,…….                                   blocks

                                                                                              Ciphertext


                               Time slices generator                                Decryption F (L,R)



         Figure 7: Overall structure of the proposed des using biometric keys and genetic algorithms




UbiCC Journal – Volume 4                                                                                         686
Special Issue on Ubiquitous Computing Security Systems



6.1. The proposed algorithm for generating the new       6.4. Sequence Of 3DES encryption
         proposed Key works as following:                CIPHERTEXT = EK3(DK2(EK1(plaintext))); i.e.,
1. The Inputs to the RC4 are:                            DES encrypt with K1, DES decrypt with K2, then
 (a).The constructed crossover-biometric formants        DES encrypt with K3.
      frequencies. For example F0-F1 Table 6(a).
 (b). The key to encrypt it which is either selected     6.5. Decryption Steps Using the Proposed Genetic
      randomly or selected from other crossover-                 Biometric Keys:
      biometric formants. For example F1-Fo Table        The decryption is a reverse process of the encryption
      6(b).                                              and can be done as follows:
  2. Apply the RC4 Encryption algorithm to the           PLAINTEXT = DK1(EK2(DK3(ciphertext))); decrypt
  inputs selected in step 1to generate a keystream       with K3, encrypt with K2, then decrypt with K1.
  bytes.                                                 1. Submit the voice of the speaker.
  3. Output from this step is the Key-Stream obtained    2. Extract the formants frequencies.
  from the RC4.                                          3. Construct the equivalent genetic biometric
  4. The Generated output from step 3 acts as input          formants signature.
  key to be introduced to the DES algorithm.             4. Encrypt the genetic biometric key using RC4.
                                                         5. Submit the key to the key generation unit in DES
6.2.     Changes in this combination are as              6. Decrypt the cipher text using the DES.
         following:
• Change the speakers to find other formants each        7   CONCLUSION AND FUTURE WORK
  different time slices.
• Change the time slice itself, (the time slice may be        DES is considered to be insecure due to the 56-
  period     of time       or     amount     of data     bit key size being too small (DES keys have been
  encrypted(100MB)).                                     broken in less than 24 hours). TDES are theoretically
• Change the crossover type operations to construct      attacked, ADES is more robust to cryptanalysis.
  different Tables similar to Table 6 periodically.           This work presented a novel contribution in
• Change the input to RC4 to be a key and the key to     securing the DES algorithm using new techniques for
  be as input swapping between the inputs and the        generating biometric formants keys in combination
  key introduced to RC4.                                 with the genetic algorithm operations as Crossover,
• Construct new biometric keys by finding different      and the combination between the two encryption
  combinations between the different formants            algorithms DES and the RC4 for encrypting the
  frequencies of speakers.                               obtained biometric key from the crossover. The work
• Swap between the speakers formants frequencies.        in this paper introduced details for the new proposed
 The step of encrypting the biometric formants key       DES system, starting from the speakers' voice
using the RC4 is shown in Table 7, and Figure 7.         capturing, formants extraction, and normalization.
Table 7 illustrates the encryption of F0-F1 using             The crossover operation done on the biometric
different Keys as F1-F0, F1-F2, F1-F3, F1-F4, and        keys in different forms ensures that the keys after the
the corresponding key-Stream generated.                  crossover were possibly strong keys and avoided the
The complexity added to the normal DES will be the       weak and semiweak keys in DES. Applying RC4 for
time needed to generate the formants frequencies for     encrypting the biometric key also increased the key-
the speakers, and the crossover operation, and the       Space for the DES. The complexity of the proposed
time needed to encrypt the crossover-biometric           algorithm may be slightly increased than the
formats keys using RC4. The added complexity was         traditional one; in the same time the key-space added
the less than n; where n is the number of bytes to be    is too high. The new proposed DES proves its
encrypted using DES. In other word the overall Big-      capabilities to be applicable in sensitive information
O of the proposed DES system will increase slightly;     systems; requiring high security.
and this increasing compared with the added Key-              The proposed DES is claimed to be more secure
Space increased is negligible. The proposed              and robust due to the uniqueness of formants of
decryption DES using the biometric formants will be      speaker; who sends the message. Attacking the
as the reverse of the encryption in the proposed DES     proposed, requires overcoming multi-layers:
based biometric formants. The overall structure of       speakers,     utterances,    formants,      digitization
the new proposed DES is shown in Figure 7.               technique, sequence of crossover and its variant
                                                         types, the predefined time slices, and the encryption
6.3. The proposed 3DES                                   of the keys using RC4, in addition to the core
The proposed 3DES is constructed as same as the          algorithm of DES. The big O of the proposed
proposed DES moreover it is needed to generates 3        algorithm is increased due to the added functions and
biometric keys that can be selected from Table 7; as     operations; but this increase can be neglected if one
K1=(’ÓIŸnµvQ),       K2=      (Ê@_…ÁÕ"_),       and      considers the trade-off between security and the
K3=(_ i_«GÚØ).                                           complexity of the algorithm especially for the critical




UbiCC Journal – Volume 4                                                                                    687
Special Issue on Ubiquitous Computing Security Systems



systems. Military applications, high category E-         [12]ISO DIS 8732, “Banking—Key Management
commerce, E-banking, and political issues can                 (Wholesale),” Association for Payment
benefit very much by applying this algorithm.                 Clearing Services, London, Dec 1987.
    The future extension of this work will be the        [13]R.C. Merkle and M. Hellman, “On the Security
implementation of a complete system using the new             of Multiple Encryption,”Communications of
proposed DES software and hardware device using               the ACM, v. 24, n. 7, 1981, pp. 465–467.
FPGA. Moreover the introducing of multimodal DES         [14]http://en.wikipedia.org/wiki/Triple_DES
using biometric keys in different forms as formants,     [15]Ralph Merkle, Martin Hellman: "On the Security
fingerprints, and iris as biological and unique keys.         of Multiple Encryption", Communications of
As well as the use of the Linear Feedback Shift               the ACM, Vol 24, No 7, pp 465–467, July
Registers LFSR in combination with this proposed              1981.
system for generating the biometric key based voice      [16]Paul van Oorschot, Michael J. Wiener, "A
signature.                                                    known-plaintext attack on two-key triple
                                                              encryption", EUROCRYPT'90, LNCS 473,
8   REFERENCES                                                1990, pp 318–325.
                                                         [17]Elaine Barker, William Barker, William Burr,
[1] Ruangchaijatupon,P.,Krishnamurthy,”Encryptio              William       Polk,      and    Miles     Smid,
     n and Power Consumption in Wireless LANs-                "Recommendation for Key Management – Part
     N,’’ The Third IEEE Workshop on Wireless                 1: General(Revised), NIST Special Publication
     LANs – September 27-28, 2001- Newton,                    800-57, May, 2006.
     Massachusetts.                                      [18]D. Hulton, Practical Exploitation of RC4
[2] Hardjono, ''Security In Wireless LANS And                 Weaknesses in WEP Environments, Dachb0den
     MANS,'' Artech House Publishers 2005.                    Labs, February 2002.
[3] W.Stallings, ''Cryptography and Network              [19]http://person.sol.lu.se/SidenyWood/prrate/whatf
     Security 4th Ed,'' Prentice Hall , 2005,PP. 58-          orm.html, Tutorial, Beginners guide to Praat
     309.                                                     ,March 2008
[4] Coppersmith, D. "The Data Encryption Standard        [20]Stanely.chen, Ellen Eide and Michaael A.
     (DES) and Its Strength Against Attacks."I BM             Picheny,      "Advanced Speech Recognition,
     Journal of Research and Development, May                 Topics in signal Processing", September, 22,
     1994,pp. 243-250.                                        2005., ELEN E6884.
[5] Bruce Schneier. The Blowfish Encryption              [21]http://en.wikipedia.org/wiki/Formant, Formant-
     Algorithm Retrieved October 25, 2008,                    Wikipedia: the free encyclopedia, 2007.
     http://www.schneier.com/blowfish.html.              [22]George Musser, "Forming Formants", Scientific
[6] K. Naik, D. S.L. Wei, Software Implementation             American, National Geographic Channel,
     Strategies for Power-Conscious Systems,”                 October 20, 2007.
     Mob-ile Networks and Applications- 6, 291-          [23]D. Coppersmith,"IBM journal of Research and
     305, 2001.                                               Development; jan/Mar 2000; 44, 1/2 ABI/ C.
[7] Daemen, J., and Rijmen, V. "Rijndael: The            [24]Zaki. T. Fayed, "Proposed Biometric key for
     Advanced Encryption Standard."D r. Dobb's                DES Scheme Applying Formant's Arc-
     Journal, March 2001,PP. 137-139.                         tangents",
[8] N. El-Fishawy ,"Quality of Encryption                [25]Holland, John H (1975), Adaptation in Natural
     Measurement of Bitmap Images with RC6,                   and Artificial Systems, University of Michigan
     MRC6,       and    Rijndael    Block     Cipher          Press, Ann Arbor.
     Algorithms", International Journal of Network       [26]Koza, John (1992), Genetic Programming: On
     Security, , Nov. 2007, PP.241–251.                       the Programming of Computers by Means of
[9] D. S. Abdul. Elminaam, H. M. Abdul Kader, M.              Natural Selection, MIT Press. ISBN 0-262-
     M. Hadhoud, "Performance Evaluation of                   11170-5.
     Symmetric Encryption Algorithms", IJCSNS            [27]William Stallings; "cryptography and Network
     International Journal of Computer Science and            Security; Principles and Practice"; Prentice
     Network Security, VOL.8 No.12, December                  Hall; 2003.
     2008, pp. 58-64.                                    [28]Bruce Schneier, "Applied Cryptography", John
[10]K. Van Espen and J. Van Mieghem, “Evaluatie               Wiley &Sons, Inc, 2004, ISBN: 471128457.
     en Implementatie van Authentiserings algorit-
     men,” graduate thesis, ESAT Laboratorium,
     Katholieke Universiteit Leuven, 1989.
[11]ANSI X9.17 (Revised), “American National
     Standard for Financial Institution Key
     Management (Wholesale),” American Bankers
     Association, 1985.




UbiCC Journal – Volume 4                                                                                688

				
DOCUMENT INFO
Description: UBICC, the Ubiquitous Computing and Communication Journal [ISSN 1992-8424], is an international scientific and educational organization dedicated to advancing the arts, sciences, and applications of information technology. With a world-wide membership, UBICC is a leading resource for computing professionals and students working in the various fields of Information Technology, and for interpreting the impact of information technology on society. www.ubicc.org