Method and apparatus for generating error detection check bytes for a data record

Claim 2

Apparatus for generating a set of CRC check bytes for a variable-length record formed as a sequence of data bytes and error correction check bytes inserted into said sequence of data bytes at preselected intervals, comprising a computing system adapted to operate on said sequence of data bytes and error correction check bytes in order to generate said set of CRC check bytes, characterised in that said computing system comprises first computing means for sequentially receiving said sequence of data bytes and error correction check bytes in said record and for multiplying all said data bytes and error correction check bytes, except for a first preselected set of said error correction check bytes, by a matrix of the form T , where n is an integer and different from any integer used for computing said error correction check bytes, in order to generate a first sub-set of CRC check bytes, second computing means for sequentially receiving said sequence of data bytes and error correction check bytes in said record and for multiplying all said data bytes and error correction check bytes, except for a second preselected set of error correction check bytes, by a matrix of the form T -(n+1) in order to generate a second sub-set of CRC check bytes, and means for combining said first and second sub-sets of CRC check bytes so as to generate said set of CRC check bytes.

Claim 3

Apparatus as claimed in claim 1, for operating on a record formed as a sequence of data bytes and a set of first level error correction check bytes and a set of second level error correction check bytes, characterised in that said first preselected set of error correction check bytes comprises all said second level error correction check bytes, and said second preselected set of error correction check bytes comprises a predetermined number of first level error correction check bytes.

Claim 4

Apparatus as claimed in claim 1 or claim 2 characterised in that said first and second sub-sets of CRC check bytes contain the same number of check bytes.

Claim 5

Apparatus as claimed in claim 3 characterised in that said first and second sub-sets of CRC check bytes each contain two check bytes.

Claim 8

A method as claimed in claim 5, for operating on a record formed as a sequence of data bytes and a set of first level error correction check bytes and a set of second level error correction check bytes, characterised in that said first preselected set of error correction check bytes comprises all said second level error correction check bytes, and said second preselected set of error correction check bytes comprises a predetermined number of said first level error correction check bytes.

Claim 9

A method as claimed in claim 5 or claim 6 characterised in that said first and second sub-sets of CRC check bytes contain the same number of check bytes.

Claim 10

A method as claimed in claim 7 characterised in that said first and second sub-sets of CRC check bytes each contain two check bytes.

Claim 11

Error correction apparatus, for performing an error correction operation on a record including a plurality of CRC check bytes (CRC) generated by a method as claimed in claim 5, said apparatus comprising a computing system for computing the value of two CRC error syndromes for use in said error correction operation, characterised in that said computing system comprises first error syndrome computing means for computing the value of a first CRC error syndrome S (1) adjusted for error correction in accordance with the relation where -n is any integer, i is the byte count in a sequence D i of all the data bytes and error correction check bytes of said record, except for a first preselected set of said error correction check bytes, and E i (D i ) is the sequence of error correction check bytes corresponding to the sequence, second error syndrome computing means for computing the value of a second CRC error syndrome S -(n+1) (1) adjusted for error correction in accordance with the relation where i is the byte count in a sequence Z i of all the data bytes and error correction check bytes in said record, except for a second preselected set of said error correction check bytes, and E 1 (Z i ) is the sequence of error correction check bytes corresponding to the sequence, Z . , and means for utilising said two CRC error syndromes for performing said error correction operation.

Claim 12

Error correction apparatus as claimed in claim 9, for performing an error correction operation on a record having first and second level error correction check bytes, comprising second level processing means for processing second level error correction bytes, characterised in that said second level processing means comprises first partial processing means for computing a first partial value of the CRC error syndrome adjusted for second level error correction in accordance with the relation where i is the byte count in a sequence D i of all the data bytes in the record and all said first level error correction check bytes but excluding all said second level error correction check bytes, and E 2 (D i ) is the sequence of second level error correction bytes corresponding to the sequence, D . , first combining means for combining the first partial value of the CRC error syndrome computed by the first partial processing means with the value of the CRC error syndrome computed by said first error syndrome computing means to produce a first total value of the CRC error syndrome, second partial processing means for computing a second partial value of the CRC error syndrome adjusted for second error correction in accordance with the relation where i is the byte count in a sequence Z i of all data bytes in the record and a predetermined number of said first level error correction check bytes but excluding all said second level error correction check bytes, and E 2 (Z i ) is the sequence of second-level error correction bytes corresponding to the sequence, Z., second combining means for combining the second partial value of the CRC error syndrome computed by said second partial processing means with the value of the CRC error syndrome computed by said second error syndrome computing means to produce a second total value of the CRC error syndrome, and means for utilising said first and second total values of said CRC error syndrome for performing said error correction operation.

Claim 13

A method of correcting errors in a record including a plurality of CRC check bytes (CRC) generated by a method as claimed in claim 5, said method of correcting errors comprising performing a computing operation to compute the value of two CRC error syndromes for use in said error correction operation, characterised in that said computing operation comprises computing the value of a first CRC error syndromes S -n (1) adjusted for error correction in accordance with the relation where -n is any integer, i is the byte count in a sequence D i of all the data bytes and error correction check bytes of said record, except for a first preselected set of said error correction check bytes, and E 1 (D i ) is the sequence of error correction bytes corresponding to the sequence, D i , computing the value of a second CRC error syndrome S -(n+1) (1) adjusted for error correction in accord ance with the relation where i is the byte count in a sequence Z i of all the data bytes and error correction check bytes in said record, except for a second preselected set of error correction check bytes, and E 1 (Z i ) is the sequence of error correction check bytes corresponding to the sequence, Z i , and utilising said two C R C error syndromes to correct said errors.

Claim 14

A method of correcting errors as claimed in claim 11, for correcting errors in a record having first and second level error correction check bytes, said method comprising processing second level error correction bytes, characterised in that said second level processing operation comprises computing a first partial value of the CRC error syndrome adjusted for second level error correction in accordance with the relation where i is the byte count in a sequence D i of all data bytes in the record and all said first level error correction check bytes but excluding all said second level error correction check bytes, and E 2 (D i ) is the sequence of second-level error correction bytes corresponding to the sequence, D., combining the first partial value of the CRC error syndrome computed by the first partial processing means with the value of the CRC error syndrome computed by said first error syndrome computing means to produce a first total value of the CRC error syndrome, computing a second partial value of the CRC error syndrome adjusted for second error correction in accordance with the relation where i is the byte count in a sequence Z i of all data bytes in the record and a predetermined number of said first level error correction check bytes but excluding all said second level error correction check bytes, and E 2 (Z i ) is the sequence of second-level error correction bytes corresponding to the sequence, Z., combining the second partial value of the CRC error syndrome computed by said second partial processing means with the value of the CRC error syndrome computed by said second error syndrome computing means to produce a second total value of the CRC error syndrome, and utilising said first and second total values of said CRC error syndrome to correct said errors.