Editing RC4 (section)

===RC4A===
[[Souradyuti Paul]] and [[Bart Preneel]] have proposed an RC4 variant, which they call RC4A.<ref>{{Citation |chapter=A New Weakness in the RC4 Keystream Generator and an Approach to Improve the Security of the Cipher |author1=Souradyuti Paul |author-link1=Souradyuti Paul |author2=Bart Preneel |author-link2=Bart Preneel |chapter-url=http://homes.esat.kuleuven.be/~psourady/publication-info/PP04-bias_rc4.htm |year=2004 |title=Fast Software Encryption, FSE 2004 |series=Lecture Notes in Computer Science |volume=3017 |publisher=Springer-Verlag |isbn=978-3-540-22171-5 |pages=245–259 |doi=10.1007/978-3-540-25937-4_16 |access-date=2011-11-04|doi-access=free }}</ref>

RC4A uses two state arrays {{mono|S1}} and {{mono|S2}}, and two indexes {{mono|<var>j1</var>}} and {{mono|<var>j2</var>}}.  Each time {{mono|<var>i</var>}} is incremented, two bytes are generated:
# First, the basic RC4 algorithm is performed using {{mono|S1}} and {{mono|<var>j1</var>}}, but in the last step, {{mono|S1[<var>i</var>]+S1[<var>j1</var>]}} is looked up in {{mono|S2}}.
# Second, the operation is repeated (without incrementing {{mono|<var>i</var>}} again) on {{mono|S2}} and {{mono|<var>j2</var>}}, and {{mono|S1[S2[<var>i</var>]+S2[<var>j2</var>]<nowiki>]</nowiki>}} is output.

Thus, the algorithm is:

 <span style="color: green;">''All arithmetic is performed modulo 256''</span>
 i := 0
 j1 := 0
 j2 := 0
 '''while''' GeneratingOutput:
     i := i + 1
     j1 := j1 + S1[i]
     [[Swap (computer science)|swap values]] of S1[i] and S1[j1]
     '''output''' S2[S1[i] + S1[j1]<nowiki>]</nowiki>
     j2 := j2 + S2[i]
     swap values of S2[i] and S2[j2]
     '''output''' S1[S2[i] + S2[j2]<nowiki>]</nowiki>
 '''endwhile'''

Although the algorithm required the same number of operations per output byte, there is greater parallelism than RC4, providing a possible speed improvement.

Although stronger than RC4, this algorithm has also been attacked, with Alexander Maximov<ref>{{Citation |title=Two Linear Distinguishing Attacks on VMPC and RC4A and Weakness of RC4 Family of Stream Ciphers |author=Alexander Maximov |url=http://eprint.iacr.org/2007/070 |date=22 February 2007 |id=Cryptology ePrint Archive: Report 2007/070 |access-date=2011-11-04}}</ref> and a team from NEC<ref name="nec">{{Citation |title=The Most Efficient Distinguishing Attack on VMPC and RC4A |url=http://www.ecrypt.eu.org/stream/papersdir/037.pdf |year=2005 |author1=Yukiyasu Tsunoo |author2=Teruo Saito |author3=Hiroyasu Kubo |author4=Maki Shigeri |author5=Tomoyasu Suzaki |author6=Takeshi Kawabata}}</ref> developing ways to distinguish its output from a truly random sequence.