JPS6248409B2 - - Google Patents

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to an error correction decoding circuit, and more specifically, to an error correction decoding circuit, which corrects or corrects code errors when an image signal is converted into a digital signal, transmitted or recorded, and then reproduced. This invention relates to a circuit that corrects a reproduced image.

[Background of the invention]

Digital magnetic recording devices such as digital VTRs that record and play back PCM-encoded television signals use error correction codes to correct code errors in order to eliminate the effects of code errors that occur during playback. . Then, only for errors for which code errors cannot be corrected, correction is performed by replacing the signal with a correction signal that has a strong correlation with, for example, a television signal from one scanning line before. By the way, in error correction codes, there is a limit to the code errors that can be corrected. For example, in Hamming codes, one
Bit errors are corrected correctly, but errors of 2 or more bits are either not corrected (hereinafter referred to as uncorrectable), or corrected but the incorrect correction is not performed. It is done. When a correction operation is performed, it is impossible to determine whether a correct correction or an erroneous correction has been made in the corrected code.

On the other hand, if the error cannot be corrected because the error correction operation is not performed, it is possible to determine whether the error is uncorrectable by, for example, dividing by the generator polynomial of the error correction code and checking the remainder. That is, if the error cannot be corrected, it can be determined that there is an error in the code word. For this reason, even if correction is performed only in cases that cannot be corrected, some types of code errors may be incorrectly corrected, and the problem is that image quality may deteriorate due to the effects of these incorrectly corrected errors. There is.

On the other hand, in digital VTRs, etc., code errors that cannot be corrected or are incorrectly corrected using error correction codes are bursts in which code errors are concentrated in one place due to dropouts caused by scratches on the tape, etc. This is incorrect.

[Purpose of the invention]

Therefore, an object of the present invention is to provide a circuit that converts image information into a digital signal, transmits or records it as an error correction code, and decodes and reproduces the signal.
The purpose is to reduce deterioration in image quality due to the above-mentioned erroneous correction.

[Summary of the invention]

In order to achieve the above object, the present invention treats a plurality of code words of an error correction code as one block, records or transmits the code by temporally interleaving the code in block units, and performs deinterleaving (inverse processing of interleaving) during playback. ) is converted into the original time-series signal, and then correction is performed using an error correction code, and the correction is controlled according to the number of error correction codes included in the block that are no longer correctable. shall be.

[Embodiments of the invention]

The present invention will be explained in detail below using the drawings. FIG. 1 is a time chart for explaining the operating principle of the present invention.

FIG. 1a shows one block of an error correction code, and W ₁ , W ₂ , W ₃ and W ₄ each correspond to one error correction code word. One block of codes consisting of these four error correction codes is recorded after, for example, interleaving shown in FIG. Assume that a dropout occurs during reproduction, and a long burst error that cannot be corrected by an error correction code occurs in the area indicated by the arrow in FIG. When this code is converted into the original time-series code by deinterleaving processing, the code sequence shown in c of the same figure is obtained. Here, assume that the error correction code is capable of correcting a 1-bit code error, for example. In this case, each of the code words W ₁ , W ₂ , W ₃ , and W ₄ included in the block includes a plurality of errors (indicated by diagonal lines) that are greater than or equal to correctable code errors. In other words, when the burst error is longer than a certain length, each of the deinterleaved code words W ₁ ......W ₄ has a correctable code error (length 1), as shown by the diagonal line (length 2). The above code errors will be included.

Therefore, when an error correction operation is performed, as shown in d of the same figure, corrections are impossible or incorrect, and in d of the same figure, W ₁ , W ₃ , and W ₄ are uncorrectable, and W ₂ Mistakes will be corrected. on the other hand,
As mentioned above, it is possible to determine which codewords are uncorrectable by performing operations such as dividing the error-corrected code by the generator polynomial of the error-correcting code and finding the remainder. It is. Therefore, as shown in Figure e, it is determined that W ₁ , W ₃ , and W ₄ of the code words after error correction contain errors and cannot be corrected.
On the other hand, although an erroneous correction was actually performed on W ₂ , since the correction operation was already performed, it is determined in this determination that there is no code error. Therefore, W ₁ , W ₃ , which were uncorrectable,
In the conventional method shown in Figure f, which corrects only the part corresponding to W ₄ , the incorrectly corrected part,
W ₂ continues to contain errors and the image quality deteriorates.
On the other hand, in the present invention, for the corrected code, as shown in FIG. . In the case of e in the figure, the number is 3. For example, if there are two or more uncorrectable code words in a block,
By performing control to modify all code words included in that block, the code of _W2 , which has been erroneously corrected, is also modified as shown in g of the same figure.

As described above, when a long burst error exceeding the correction range occurs in a block, almost all of the error correction code words included in the block contain more errors than can be corrected. Therefore, in this case, the error correction codeword included in the block is almost impossible to correct or is incorrectly corrected. On the other hand, the probability that the correcting code will erroneously correct a code error that exceeds the correction range is approximately 0.1 to 0.3, depending on the error correcting code used. Therefore, if a long burst error that cannot be corrected occurs, 70 to 90% of the code words in the block will be uncorrectable, and the remaining 10 to 30% will be incorrectly corrected. Therefore, as in the present invention, by controlling the modification of the code after error correction according to the number of uncorrectable code words included in the block,
It becomes possible to efficiently remove code words that have been incorrectly corrected.

FIG. 2 is a block diagram showing an embodiment of an error correction decoding circuit according to the present invention, which is constructed based on the above principle. After an error correction code is added to the input code in the encoder 1, a plurality of error correction code words are treated as one block as shown in FIG. interleave the codes in time,
Record on VCR. On the other hand, the reproduced code from the VTR is converted into the original time-series code in units of blocks in the deinterleave processing circuit 3, and then subjected to error correction processing in the decoder 4. Then, the code after error correction is processed by the 1-block delay circuit 5.
A delay of one block is carried out at . On the other hand, the uncorrectable code word detection circuit 6 detects uncorrectable code words from the error-corrected code, and outputs a signal as shown in FIG. 1e. On the other hand, the counting circuit 7 measures the number of uncorrectable code words included in each block, and if this number is 2 or more, all code words included in the block are corrected; For example, the selection circuit 8 is controlled so that only the code word detected by the uncorrectable code word detection circuit 6 is corrected. Therefore, from the output of the selection circuit 8, a code is obtained in which the erroneously corrected code word is corrected by the correction code.

According to the present invention, for example, if some codewords in a block are correctly corrected and the remaining codewords are uncorrectable or incorrectly corrected, the correctly corrected codewords can also be corrected. There is no problem with being exposed. however,
The degree of image quality deterioration due to erroneous correction due to erroneous correction is far greater than the image quality deterioration caused by correcting correctly corrected code words, and is not a problem at all in practice.

As described above, according to the present invention, even for long burst errors that are erroneously corrected by an error correction code, the correction mode can be switched depending on the number of uncorrectable code words included in the block. It becomes possible to easily correct a code word that has been incorrectly corrected using a correction code, and the effect obtained is significant.

[Brief explanation of the drawing]

FIG. 1 is a time chart for explaining the principle of the present invention, and FIG. 2 is a block diagram showing the configuration of an embodiment of an error correction decoding circuit according to the present invention. 1... Encoder, 2... Interleave processing circuit,
3...Deinterleave processing circuit, 4...Decoder,
5... 1 block delay circuit, 6... Uncorrectable code word detection circuit, 7... Counting circuit, 8... Selection circuit.

Claims (1)

[Claims] 1. In a circuit that decodes a digital signal consisting of a plurality of error correction code words as one block, there is a circuit that performs error correction based on the above error correction code word, and a code that does not undergo correction even though there is a code error. a detection circuit for detecting words; and means for measuring the number of words to be corrected in one block based on the output of the detection circuit, and when the output of the measuring means is a certain number or more, An error correction decoding circuit characterized in that the error correction decoding circuit is configured to modify all code words in the code word with a highly correlated modification signal.