Browse Prior Art Database

Error Correction Code Pointer Control with Pointer Preprocessing

IP.com Disclosure Number: IPCOM000117903D
Original Publication Date: 1996-Jul-01
Included in the Prior Art Database: 2005-Mar-31
Document File: 4 page(s) / 137K

Publishing Venue

IBM

Related People

Bentley, SR: AUTHOR [+2]

Abstract

Disclosed is a method of analyzing condensed error pointers to select the preferred pointers prior to their use in an Error Correction Code (ECC) decoder.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 50% of the total text.

Error Correction Code Pointer Control with Pointer Preprocessing

      Disclosed is a method of analyzing condensed error pointers to
select the preferred pointers prior to their use in an Error
Correction Code (ECC) decoder.

      When too many pointers are provided for a given code block,
the pointer preprocessor algorithm attempts to delay the use of
enough pointers to permit error correction to proceed while
optimizing the effectiveness of the ECC decoder by increasing the
probability that the correct error pointers will be selected by the
time they are needed.

Assume a system that:
  o  Uses a product code (an outer-code, inner-code code-block)
where:
     -  the outer code is an error correction code and,
     -  the inner code is an error detection code (generates error
     -  location pointers for the outer code),
  o  Interleaves the inner code rows or Minimum Interleave Elements
      (MIE's) over multiple product code code-blocks,
  o  Creates additional error pointers as a function of:
     -  the modulation code and,
     -  quality estimates of each track's timing,
  o  Employs the error detection and location strengths of the error
      correction code.

      Such a system requires that the error pointers generated in the
interleaved domain be properly deinterleaved and synchronously
applied to the outer-code decoder with the deinterleaved outer code
code-words.  One such system has combined inner code and modulation
error pointers with synchronization pointers to form three new
generic pointers:
  VPTR     Valid Pointer, indicates a true error was detected
            within the EDC row
  SHFLG    Second Half Flag, indicates that there is a high
            probability that the error is in the second half of the
            MIE row
  EXPTR    Expanded Pointer, can indicate an error even if the
            Valid Pointer is not set.  (It is set whenever the
            dataflow determines there is some likelihood an error is
            present in the MIE even if no error was explicitly
            detected.)

      A detailed description of the Pointer Pre-Processing is given
below for a product code where the outer code can correct up to 8
errors in known locations.  The inner code covers 32 columns of outer
codes.  This pre-processing operation is also shown in the flow
charts of Figs. 1 and 2.

      When the deinterleaver circuit indicates a code-block is ready
for correction, the Pointer Control Circuit (PCC) receives the VPTR,
SHFLG, & EXPTR for each MIE and loads them into a 4 bit x 32 bit
FIFO.  The fourth bit in the FIFO is the Pointer Select (PSEL) flag
that is used to indicate whether the corresponding VPTR or EXPTR has
been selected to become active at the next opportunity.  As the
pointers are received from the deinterleaver, the PCC counts the
number of...