Error detection/correction and fault detection/recovery – Pulse or data error handling – Error/fault detection technique
Patent
1997-12-16
2000-03-28
Nguyen, Hoa T.
Error detection/correction and fault detection/recovery
Pulse or data error handling
Error/fault detection technique
380 42, G06F 1110
Patent
active
060444886
DESCRIPTION:
BRIEF SUMMARY
The present invention refers to a method for generating a check word for a bit string to check the integrity and the authenticity of the bit string, the bit string comprising at least one set of data consisting of a plurality of data words, each having the same word length defined by a number of bits. This method will subsequently be referred to as the message digest method, the message digest being the check word. Further, the invention refers to a method for sending bit strings with secured integrity and authenticity using a message digest generated with a method of the type mentioned above.
In contrast to encoding methods, message digest methods do not aim at secrecy, but are used to secure the integrity of data.
To this end, a hash function is typically applied to a data set of a bit string, a message digest being calculated from the data and a secret key, i.e., a key (only) known to the sender and the receiver, also referred to as "seed". When the sender generates such a message digest and, together with the data, sends it to the receiver, the receiver can determine, whether unauthorized modifications have been made to the data (authenticity check). To do so, the receiver calculates the message digest, using the same seed, and checks, whether the newly calculated message digest corresponds to the one sent. An intruder may be able to manipulate the data, however, he usually cannot calculate a digest matching the data, since he does not dispose of the secret key (seed).
Standard methods for generating a message digest are among others N-hash, MD5 (Rivest 1992) and SHA (Metzger and Simpson 1995).
Such methods are also employed for securing data transfer via networks. For example, present efforts aim at enlarging the IP protocol correspondingly (Atkinson 1995).
A problem of the known methods when used in high speed networks is their unsatisfactory speed.
It is an object of the present invention to provide a very fast and sufficiently secure message digest method that is particularly suitable for large amounts or streams of data.
The object is solved, according to the present invention, with a method for generating a check word for a bit string to check the integrity and authenticity of the bit string, the bit string comprising at least one set of data consisting of a plurality of data words, each having the same word length defined by a number of bits, and wherein number of bits as the data words, word and a random number, subjected to a first linking operation, subjected to a second linking operation, and are linked, the result of this linking being the check word of the set of data.
In the method of the present invention, the data words are combined with random numbers, preferably by simple operations, such as modulo-2.sup.wordlength -addition and/or modulo-2.sup.wordlength -multiplication. The sequences in which these operations are effected and in which the random numbers are combined with the data words result from the secret key. Suitably, the key also holds information on the generation of the random numbers.
Besides the two above-mentioned linking operations, a bit-by-bit exclusive-OR operation may also be applied to the operands of the operand pairs.
The generation of the check word becomes more secure against unauthorized manipulation when the results of all operations applied to the first and second operands are linked alternately by a first linking operation and a second linking operation.
The statements made above in connection with the operations are also true for the first and second linking operations.
Preferably, the bit string comprises a plurality of data sets, each having the same number of data words, and the random number, with which the data words of the first and second operand pairs of a data set are linked, is the same for each data set. In this respect, it is advantageous to alter the sequence of the random numbers, according to which these are linked to the data words of the first and second operand pairs, per data set.
The check words obtained for the respective data sets are
REFERENCES:
patent: 5347650 (1994-09-01), Arditti et al.
IBM Technical Disclosure Bulletin, vol. 28, No. 4, Sep. 1985, pp. 1426-1427, "Authentification Technique Using Manipulation Detection Codes".
"Zur Fehlererkennungsfahigkeit Von Random Codes am Beispiel des Message Authentification Code" B. Friedrichs, pp. 145-152.
IEEE Transactions on Computers, vol. 41, No. 4, Apr. 1992, pp. 505-507, C. J. Mitchell, "Authenticating Multicast Internet Electronic Mail Messages Using a Bidirectional . . . ".
Securicom 88, Worldwide Congress on Computer and Communications Security & Protection, Jan. 1988, pp. 91-110, M. Campana et al., "Comment Utiliser Les Fonctions de Condensation . . . ".
Chase Shelly A
IBM
Nguyen Hoa T.
LandOfFree
Process for generating a check word for a bit sequence for verif does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Process for generating a check word for a bit sequence for verif, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process for generating a check word for a bit sequence for verif will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1335805