Cryptographic data integrity with serial bit processing and pseudo-random generators

• US 6,069,954 A
• Filed: 05/09/1997
• Issued: 05/30/2000
• Est. Priority Date: 05/29/1996
• Status: Expired due to Fees

• Current Assignee: Moreau, Thierry
• Original Assignee: Moreau, Thierry
• Inventors: Moreau, Thierry
• Primary Examiner(s): Swann, Tod R.
• Assistant Examiner(s): KABAKOFF, STEPHEN ERIC
• Application Number:
  US08/853,455
• Time in Patent Office: 1,117 Days
• Field of Search: 380/9, 380/28, 380/29, 380/42, 380/43, 380/44, 380/45, 380/46, 380/47, 380/268
• US Class Current: 380/28
• CPC Class Codes: H04L 9/0668 producing a non-linear pseudorandom sequence

Abstract

A new structure for a secret key cryptography algorithm uses double exclusive-or (XOR) encryption (e_i =k1_i XOR m_i XOR k2₁), analogue to a stream cipher arrangement, but providing data integrity protection. The double XOR encryption creates an inner sequence, for example s_i =k1_i XOR m_i for the encryption process, s_i =k2_i XOR e_i for the decryption process, which is unknown to the adversary and is used as a starting point for feedback on the very key streams used in the encryption of following bits of the message. In its preferred embodiment, this structure is the Frogbit data integrity algorithm which uses 10 independent pseudo-random sources. The Frogbit algorithm is well suited to the design and implementation of "semi-proprietary" ciphers, where the overall design is publicly known and peer-reviewed but a significant amount of the implementation specification is left open to private customization. The Frogbit algorithm is a means to create computer programs and microprocessor-based devices that are harder to crack with reverse engineering than with the prior art.

Claims

1. A cryptographic data integrity binary representation generator system for use with a second processor acting on said binary representation, or a transformation thereof, to produce a signal indicating, with an overwhelming probability whether said binary representation was generated according to said generator system, said generator system comprising:

means for consecutively transforming a cleartext message bit m_i into a bit e_i using bits k1_i and k2_i from keys k1 and k2 according to the following equation;

space="preserve" listing-type="equation">e.sub.i =k1.sub.i XOR m.sub.i XOR k2.sub.i ;

and

means for generating successive bits k1_i+1 and k2_i+1 from a decision machine having finite state information using (k1_i XOR m_i) and k2_i as inputs, said keys k1 and k2 having initial state information defining said bits k1_i and k2_i, wherein said binary representation consists of a string of consecutive bits e_i, e_i+1, e_i+2, etc.

2. The binary representation system as claimed in claim 1, wherein said decision machine comprises a plurality of pseudo-random generators each having its own state information, said inputs of said decision machine influencing which one of said plurality of pseudo-random generators is used to obtain said bits k1_i+1 and k2_i+1.

3. The binary representation system as claimed in claim 2, wherein the said plurality of pseudo-random generators has ten pseudo-random generators.

6. The binary representation generator system as claimed in claim 1, wherein a string of said clear text message bits mi represent a cryptographic key.

4. The binary representation generator system as claimed in 2, wherein said plurality of pseudo-random generators make use of different algorithms.

5. The binary representation generator system as claimed in 3, wherein said plurality of pseudo-random generators make use of different algorithms.