Replacement of bad data bit or bad error control bit

Details Replacement of bad data bit or bad error control bit Replacement of bad data bit or bad error control bit

1999-05-07

2002-07-23

Ton, David (Department: 2133)

Error detection/correction and fault detection/recovery

Pulse or data error handling

Digital data error correction

C714S799000, C714S723000

Reexamination Certificate

active

06425108

ABSTRACT:

FIELD OF THE INVENTION
This invention relates to use of error control bits to compensate for corrupted memory cells.
BACKGROUND OF THE INVENTION
Error control of a digital signal transferred from one site to another is often implemented by use of error control bits (parity, CRC, syndrome polynomials) that are associated with a segment of the signal to be transferred. However, one or more of the digital signal bits may become corrupted through a process that is not associated with, or that does not permit correction by use of, the error control bits associated with that segment. This may occur when a memory cell becomes corrupted and will not accept and hold a particular voltage value, corresponding to a “zero” or a “one.” In this situation, an entire digital signal segment may become unusable although only one or a few of the digital signal bits in memory are “bad.”
Segments containing data bits and segments containing the associated error control bits are traditionally stored in fixed locations in separate regions in a digital computer memory. This approach is acceptable if a data bit segment is either correct, as interrogated using a selected error control procedure, or if the data bit is erroneous but can be corrected using the error control procedure. However, if the data bit segment contains one or more bit errors that cannot be so corrected, an entire region in memory can be compromised and rendered unusable.
What is needed is an approach that can reduce or minimize the disruption of a data bit segment in which one or more erroneous and uncorrectable data bits are found and should allow subsequent identification of any such data bit segments. Preferably, the approach should not affect a data bit segment that either contains no errors or contains only correctable data bit errors. Preferably, the approach should involve processing of only the bits in the data bit segment and in the associated error control bit segment.
SUMMARY OF THE INVENTION
These needs are met by the invention, which (1) performs an error check on a data bit segment to determine if one or more data bit errors is present, and (2) where one or more data bit errors is present, determines whether the bit error(s) is correctable. Where the first question is answered “yes” and the second question is answered “no”, the system replaces one or more data bits in this corrupted data bit segment by one or more error control bits, and flags the segment in which the replaced bit(s) is located, in order to disable application of the error control check(s) for that modified data bit segment.


REFERENCES:
patent: 4475194 (1984-10-01), LaVallee et al.
patent: 4479214 (1984-10-01), Ryan
patent: 4903268 (1990-02-01), Hidaka et al.
patent: 5233614 (1993-08-01), Singh
patent: 5640353 (1997-06-01), Ju
patent: 5974576 (1999-10-01), Zhu
patent: 5996096 (1999-11-01), Dell et al.

Affiliated with

Also associated with

Rating

Replacement of bad data bit or bad error control bit does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Replacement of bad data bit or bad error control bit, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Replacement of bad data bit or bad error control bit will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2895822