Error detection/correction and fault detection/recovery – Pulse or data error handling – Digital data error correction
Reexamination Certificate
2000-11-28
2004-04-06
Moise, Emmanuel L. (Department: 2133)
Error detection/correction and fault detection/recovery
Pulse or data error handling
Digital data error correction
C714S767000
Reexamination Certificate
active
06718505
ABSTRACT:
The invention relates to a method and an apparatus for correcting errors in a stream of data by using a method of decoding a cross-interleaved Reed-Solomon code (CIRC).
BACKGROUND OF THE INVENTION
There has been a demand for increased quality of audio recording and read back systems as well as of other types of communication.
One problem with recording media are the defects in the media resulting in areas with improperly recorded digital data or in data that cannot be read back reliably. As a result errors occur in the read back data as well as in transmissions of digital data. During recording, read back and transmission of digital data, errors occur in the digital data with some finite probability. The data is typically composed of binary units, a group of binary units (such as 8) makes up a data byte, and groups of bytes (such as 2) make up a data word. Additionally the data is arranged into blocks of data (such as 32 or 28 byte blocks).
There are two different types of errors. The first type is a single bit error which is the substitution of one of two possible values of a binary bit for its opposite value. Such errors usually occur randomly in a digital signal. A second type of error consists of a continuous sequence of erroneous bits. Such errors are referred to as burst errors. The length of these bursts and their frequency of occurrence are also random.
One error correction code that is typically used in compact audio discs is a so-called CIRC correction code. CIRC is an acronym for crossinterleaved Reed-Solomon code. The CIRC utilises a two-step process. In passing through a first encoder, 24 consecutive data bytes representing a data polynomial are divided by a generator polynomial. In this process, four parity bytes are added. The result is a block of 24 data bytes and 4 parity bytes (C
1
code). This consecutive sequence of bytes is interleaved or dispersed among other encoded data bytes.
The interleaved data is passed through a second encoder. The second encoder is identical to the first except that the bytes are presented in a different sequence due to interleaving, parity bytes together with data bytes are being encoded, blocks of 28 bytes (28 data bytes plus 4 parity bytes) are being encoded instead of 24 byte blocks, and 4 additional parity bytes are added. The result of the second encoding process is a 32 byte block (C
2
code) composed of 24 data bytes and 8 parity bytes.
In the case of the CIRC correction code, the encoding process of the (28, 24) Reed-Solomon code is performed for twenty-four data symbols (bytes), with each symbol consisting of 8 bits. In typical audio applications each audio sample comprises 16 bits and is formed of two symbols of 8 bits each. Thus, each 8-bit symbol is either the upper or lower side of an audio sample of one of the two channels of stereophonic audio data.
The encoded data is recorded optically and subsequently read back. After the encoded data is read back, there will likely be single bit or burst errors due to recording, read back or transmission problems.
Conventional error correction methods are known for use in decoding CIRC. Examples of such methods are disclosed in U.S. Pat. No. 4,546,474, U.S. Pat No. 4,476,562 and U.S. Pat No. 4,497,058. According to the conventional methods for decoding CIRC errors, the processing is run on the basis of a so-called erasure correction method. In the erasure correction method the location of error symbols is indicated by means of pointer information. Error correction is performed on this error symbol. In the case of above-mentioned C
1
and C
2
codes, detection and correction up to double errors can be performed. However, if the error location is already know, then error correction up to 4 erasures can be performed. Therefore, in order to raise the error-correction capability, the implementation of the erasure correction method for error correction has been preferred. In addition, the erasure correction method has been found to be particularly effective in correcting burst errors.
According to the conventional method used for decoding CIRC errors, correction of up to two errors is performed in the C
1
decoder. If triple errors or more occur, which of course are not corrected, C
1
pointer information is sent to the C
2
decoder in the next stage, so that error correction is performed in the C
2
decoder utilising C
1
pointer information.
Multiple use of the CIRC correction codes in decoding leads to an increase in correction capability of error correction systems. However, conventional decoding methods (CIRC decoders) are incapable of multiple processing of blocks of CIRC codes. In order to solve this problem, one can either provide conventional CIRC decoders with additional means or devices which allow multiple processing or make conventional CIRC decoders useful for multiple processing.
One method (conventional CIRC decoder) for multiple processing of CIRC codes has been disclosed in U.S. Pat No. 4,852,099. The known method uses the erasure correction method to increase the correction capability of error correction systems. It is proposed to perform C
1
decoding and C
2
decoding twice in a specific order. That order, for example, might be, C
1
decoding, followed by C
2
decoding, followed by C
1
decoding, and followed by C
2
decoding.
According to this example, C
1
code words are supplied to a C
1
decoder in which actual decoding of the (32, 28) Reed-Solomon code is performed. Error correction of up to two errors is performed. If three or more errors are detected by the C
1
decoder, a C
1
pointer is set for all symbols in C
1
code words. Then data and error pointers corrected by means of C
1
decoder are further processed in a deinterleave processing stage. An output of the deinterleaver is supplied to the C
2
decoder. Erasure correction of up to four erasures is executed in the C
2
decoder utilising C
1
pointer information. Upon completion of erasure correction in the C
2
decoder the C
1
pointer is cleared and no pointer information is transferred to the second C
2
decoding cycle.
In a second cycle data from the C
2
decoder is supplied to the interleaver that returns the data to the same arrangement as it was when it was reproduced. Thereafter, the processing in the second decoding cycle corresponds to the processing in the first decoding cycle. By using this decoding method, multiple processing of CIRC codes can be performed, but additional hardware (interleaver block) is needed.
Another method used in decoding CIRC utilising multiple processing of CIRC is disclosed in U.S. Pat No. 4,637,021. Error detection and error correction is achieved by processing blocks of digital data bytes with a C
1
decoder and a C
2
decoder. In order to maximise the rate at which data is processed, decoders C
1
and C
2
actually operate concurrently on data stored in a system memory with the C
1
decoder operating on the data ahead of the C
2
decoder. According to the disclosed decoding method of the CIRC correction code, error processing up to double-error correction is executed in C
1
decoding in first stage, and double-error correction is executed in C
2
decoding at next stage by referring to C
1
pointer information that is derived from the C
1
decoder. In this case, C
1
pointer information is not used to increase the error correction capability of error correction systems. C
1
pointer information is used to check the quality of the decoding process.
In the method according to the U.S. Pat No. 4,852,099, data bytes are read by means of C
1
and C
2
decoders from the system memory according to the following sequence. A first C
1
decoder processes a C
1
block of data bytes (32 bytes). A first C
2
decoder then processes a C
2
block of data bytes (28 bytes) that has already been processed by the C
1
decoder; this concludes a first pass. A second C
1
decoder processes data bytes that have already been processed by the first pass, and a second C
2
decoder processes data bytes that have already been processed by the first pass and the second C
1
decoder; this con
Kravtchenko Alexander
Rominger Friedrich
Vogel Jürgen
Kurdyla Ronald H.
Laks Joseph J.
Moise Emmanuel L.
Thomson Licensing S.A.
Tripoli Joseph S.
LandOfFree
Method and apparatus for error correction in a process of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for error correction in a process of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for error correction in a process of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3261508