Error detection/correction and fault detection/recovery – Pulse or data error handling – Digital data error correction
Reexamination Certificate
2001-01-02
2003-10-07
Tu, Christine T. (Department: 2133)
Error detection/correction and fault detection/recovery
Pulse or data error handling
Digital data error correction
C714S800000
Reexamination Certificate
active
06631492
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to error correction for multi-track digital data storage.
BACKGROUND OF THE INVENTION
The demand for increasing data storage capacity and data transfer rate motivates the design of linear-format digital data storage systems with more heads, narrower track pitch, higher media transport speed, and greater in-track data density. These trends can be seen particularly in the development of digital tape drives, including both optical tape drives and magnetic tape drives.
As data rate and data density are increased, the deleterious effects of media defects become more serious. For example, small imperfections in a recording medium such as a magnetic tape or an optical tape may not disturb the recording or readout of large data marks recorded at a low data density. But for higher density recording with smaller data marks the same imperfections may inhibit the recording of data marks or distort the readout signals. As another example, consider a tape recording medium that is undamaged by the transport mechanism in a low-speed linear tape drive. The same tape recording medium could be scratched in the transport of a high-speed, high capacity tape drive, resulting in the loss of one or more data tracks. What is needed is a method for digital error-correction that is suitable for multi-track linear data recording that efficiently corrects data errors due to scratches or media defects.
When the error rate is high, the overall reliability of a digital data storage system may be improved by a rewrite strategy. Newly recorded data is read to determine the relative number of digital errors. If the system Error Correction Code (ECC) cannot correct the observed error rate with adequate margin, the data is rewritten. The rewrite strategy will restrict the net data transfer rate unless the system can read the data immediately after it is written and measure the error rate without substantial delay. Therefore it is important that the error correction method provide high-speed error detection.
SUMMARY OF THE INVENTION
It is an object of this invention to provide an efficient and reliable error correction in a multi-track digital data recording system.
It is a further object of this invention to provide for error correction that reliably corrects errors caused by longitudinal scratches (oriented parallel to the data tracks and media transport direction) and other sorts of media defects.
It is a further object of this invention to support high-speed error detection and correction.
These objects are achieved by an error correction method for multitrack data recording and read out of recorded multitrack digital data, comprising the steps of:
a) organizing input digital data into a sequence of data words;
b) encoding each data word as a fixed-length transverse error correcting code word having a fixed number of symbols which represent digital data and parity;
c) recording each transverse error correcting code word onto a plurality of longitudinal data tracks on a recording medium with no more than two code symbols from each code word recorded on each track;
d) reading recorded transverse error correcting code words by a plurality of readout transducers, with at least one transducer for each data track, that reproduce the code word symbols of each transverse error correcting code word from the plurality of longitudinal data tracks; and
e) processing the transverse error correcting code words to recover the data words and identify and correct symbol errors such as caused by media defects or transducer failures by using the error correcting parity.
ADVANTAGES
The present invention has as an advantage that data errors resulting from longitudinal scratches and recording head failures may be corrected efficiently with minimal ECC overhead. The method is compatible with multi-track linear recording at high data rates and high data density. The method also provides rapid measurement of digital errors to support re-write strategies.
It is a further advantage of the present invention that it may be integrated with a multi-level ECC system to provide reliable detection and correction of random errors and large defects that may occur along with media scratches and head failures. The method further provides for incorporation of timing synchronization patterns into the data tracks with no increase in overhead. The present invention is particularly effective when it records ECC code words in a linear format; that is, the tracks are parallel to the media motion.
REFERENCES:
patent: 4719628 (1988-01-01), Ozaki et al.
patent: 4996690 (1991-02-01), George et al.
patent: 5487077 (1996-01-01), Hassner et al.
patent: 5841794 (1998-11-01), Inoue et al.
“Principles of Optical Disc Systems” by K. S. Immink, Section 7.2.2, pp. 247-254, Philips Research Laboratories, Eindhoven.
“Error-Correcting Codes”, by W. Wesley Peterson et al., Second edition, Section 9.2 and 9.7, pp. 271-307.
Eastman Kodak Company
Owens Raymond L.
Tu Christine T.
LandOfFree
Multitrack data recording and read out of recorded... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Multitrack data recording and read out of recorded..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multitrack data recording and read out of recorded... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3175315