JPH03231532A - Error correction processing method for digital signal - Google Patents

Abstract

PURPOSE:To apply reproduction processing of an information data with fidelity by applying error correction processing of an error location of an input data or interpolation processing without error correction processing based on the comparison processing result between an error number and an error location number. CONSTITUTION:When, e.g. an error number NE and an error location number NA are coincident, it is discriminated that a correctable error is mixed in an information data DA. When the error number NE and the error location number NA are not coincident, it is discriminated that the correction disable error is mixed in the information data DA. Thus, when the production of an error in a correction enable range is recognized, the error correction processing of the information data DA is implemented. When the production of an error over the correction enable range is recognized, the error correction processing of the information data DA is not implemented but the interpolation processing is executed.

Description

[Detailed description of the invention] [Table of contents] Overview Industrial field of application Prior art (Figure 7) Problem to be solved by the invention (Figure 8) Means for solving the problem (Figure 1) Operation implementation Example (i) Description of the first embodiment (Figures 2 to 4) (ii) Description of the second embodiment (Figures 5 and 6) Effects of the invention [Summary] Digital signal error correction processing method In particular, with regard to correction processing methods when multiplexed information data, etc. contains errors that exceed the error correction and detection capabilities, it is necessary to correct the information data without erroneously correcting it or leaving it uncorrected. The purpose of this method is to detect when impossible errors have been mixed in and perform accurate error correction processing.The first method is error correction processing of digital signals in which error correction codes are added to information data. Then, the digital signal is subjected to error detection calculation processing, the number of errors is determined and the number of error positions is detected based on the calculation result, the number of errors is compared with the number of error positions, and the number of errors is compared with the number of error positions. The second method includes performing error correction processing on the error position of the information data or interpolation processing without performing the error correction processing based on the comparison processing result, and the second method includes error correction of the first digital signal. In the processing method, when the information data is shortened from a data string of a standardized digital signal, error correction processing is performed on the information data based on conditions for shortening the digital signal and a result of the error position detection processing. Contains and constitutes.

[Industrial Application Field] The present invention relates to an error correction processing method for digital signals. More specifically, the present invention relates to a method for error correction processing of digital signals. More specifically, the present invention relates to a method for correcting errors in digital signals. More specifically, the present invention relates to a method for correcting errors in digital signals. This relates to a processing method.

In recent years, with the increase in density and functionality of information networks, in the field of data communication and processing, multiplexed information data, such as high-definition signals, are
(MIIZ) to compress the band and transmit it.
(Multiple 5ubnyquist Sa
2. Description of the Related Art Digital signals in which error correction codes are added to audio data, etc. of the +wpling encoding system are transmitted, and techniques for reproducing the signals are used.

According to this, burst errors, in which errors are locally concentrated in information data, may occur for some reason during the transmission of the digital signal or in the recording medium thereof. As a result, the digital signal may contain more errors than the error correction processing circuit can correct. In such a case, the correction processing circuit may erroneously correct the information data or leave it uncorrected.

Therefore, there is a need for an error correction processing method that can perform accurate error correction even when information data contains more errors than the error correction and detection capabilities.

(Prior Art) Figs. 7 and 8 are explanatory diagrams of a conventional digital signal error correction processing method.

FIG. 7 is a block diagram illustrating an example of an error correction processing method for an audio signal using the MUSE method.

In the figure, a frame deinterleave processing circuit 1. Bit deinterleave processing circuit 2. Error correction processing circuit 3. It consists of a word deinterleave processing circuit 4 and an interpolation processing circuit 5.

The error correction processing circuit 3 includes syndrome calculation circuits 3a and 8.
2-bit shift register 3b, other calculation circuit 3c, error number determination circuit 3d, error position detection circuit 3e, AND circuit 3
f and an adder 3g.

The function of this device is to input the MUSE signal DIN which has undergone word, bit and frame interleaving processing, and first perform frame and bit deinterleaving processing.

Next, the bit deinterleaved MUSE signal DIN is subjected to error correction processing. Thereafter, the corrected MUSE signal DIN is subjected to word deinterleaving processing and its interpolation processing, and is output to the audio reproduction circuit.

FIG. 8 is a flowchart of a conventional error correction processing method.

In the figure, first, in step P1, the syndrome of the MUSE signal DIN subjected to bit deinterleaving processing is calculated. At this time, 82-bit shift register 3
MUSE in enhanced mode (82, 67) in which a 15-bit error correction code is added to b and the syndrome calculation circuit 3a.
A signal DIN is input.

Next, in step P2, the number of errors is determined using the error number discriminant. At this time, the error number determining circuit 3d determines that correction is not permitted in the enhanced mode based on the syndrome calculation results when there is no error and when there is a 3-bit error.

Also, if there is a 1 or 2 bit error, permission for correction is identified.

Next, in step P3, the number of error positions is calculated in parallel using the error position equation, and the bits at the error positions are inverted. At this time, characteristic values indicating the error position are substituted into the error position equation for the number of bits (-cycles) of the data string of the MUSE signal. In the error correction process, the focus of the error position is inverted every time a solution to the error position equation is found and the correction permission identification in step P2 is activated.

Thereby, errors within the correction capability of the error correction processing circuit 3 can be corrected.

[Problem to be solved by the invention]

By the way, according to the conventional example, a MUSE signal in which a burst error or the like has occurred due to some reason may contain errors exceeding the correction ability of the error correction processing circuit 3. However, the error correction processing circuit 3 always performs correction processing on the assumption that there is an error within the correction capability range.

Therefore, the following problems may occur.

■There are cases where bits in the correct position are incorrectly corrected without correcting the bits in the incorrect position.

This is because when calculating the number of error positions in step P3 of FIG. 8, the characteristic value indicating the error position is substituted into the error position equation by the number of focuses (-cycles) of the data string of the MUSE signal. . That is, error correction processing is performed every time a solution to the error location equation is found. As a result, the number of errors determined in step P1 differs from the number of error positions found by going around the data string of the MUSE signal in step P3, and the focus at the correct position is incorrectly corrected without correcting the focus at the true error position. It's something you end up doing.

(2) Even if the result of determining the number of errors is that there are two errors in the bits of the data string in step P2, the correction process may end without actually being able to find the error position.

One of the reasons for this is that the MUSE signal is converted into a standardized data string of 1 by the error correction method using the BCH code.
It has been shortened from 27 bits to 82 bits. Therefore, if more errors than the correction ability are added to these 82 bits, the result may be that there is an error in the apparently shortened part, which does not match the result of determining the number of errors within the original 83 bits of data. This is because the processing result is as follows.

As a result, data that actually required error correction processing is interpolated based on the error correction processed data of the MUSE signal, and as a result, when the audio data is played back,
There is a problem in that noise and the like are mixed in and the sound quality is degraded.

The present invention was created in view of the problems of the conventional example, and prevents uncorrectable errors from being mixed into the information data without erroneously correcting it or leaving it uncorrected. The purpose of the present invention is to provide an error correction processing method for digital signals that enables accurate error correction processing by detecting errors.

[Means for Solving the Problems] FIGS. 1(a) and 1(b) show diagrams of the principle of a digital signal error correction processing method according to the present invention.

The first method is to add an error correction code EB to the information data DA.
is an error correction process for the digital signal DS to which is added, first, in step pt, an error detection calculation process is performed on the digital signal DS, and then in step P2. In step P3, the number of errors NH is determined and the number of error positions NA is detected based on the calculation result, and then in step P4, the number of errors NE and the number of error positions NA are compared, and further,
In step P5, an error correction process is performed on the error position In of the information data DA based on the comparison process result, or in step P6, an interpolation process is performed on the information data DA without error correction process based on the comparison process result. The second method is characterized in that in the first digital signal error correction processing method, when the information data DA is shortened from the data string DP of the standardized digital signal DS, the digital signal DS The above object is achieved by performing error correction processing on the information data DA based on the shortening conditions and the detection processing result of the error position itn.

(Operation) According to the first method of the present invention, the number of errors N in the stencil P5 is
Based on the comparison processing result between B and the number of error positions NA, error correction processing is performed for the error position n of the information data DA or step P6
Based on the comparison processing result, interpolation processing is performed without performing error correction processing on the information data DA.

For example, when the number of errors NA and the number of error positions NA match, it is determined that a correctable error is mixed in the information data DA. Also, the number of errors NE and the number of error positions NA
If they do not match, it is determined that an uncorrectable error has been mixed into the information data DA.

Therefore, when it is recognized that an error within the correction capability range has occurred, error correction processing is performed on the information data DA. Further, if it is recognized that an error exceeding the correction capability range has occurred, the error correction process of the information data DA is not performed and the process shifts to interpolation process. As a result, the digital signal D that has caused a burst error etc. for some reason.
Even if S contains an error exceeding the correction capability range, it is possible to reduce as much as possible the error correction process that erroneously corrects a bit in the correct position as in the conventional example.

As a result, information data in which errors exceeding the correction capability range are interpolated based on data that has not been subjected to error correction processing, and as a result, it becomes possible to faithfully reproduce the information data.

Further, according to the second method of the present invention, when the information data DA is shortened from the data string DP of the standardized digital signal DS, the conditions for shortening the digital signal DS and the detection processing result of the error position n The data DA is subjected to error correction processing based on.

For example, a data string DP in which the digital signal DS is standardized
= When the data string DP that is allowed to be transmitted for M bits is shortened to m bits, when an error position wn is found in the data string DP = M-m, an uncorrectable error is detected in the information data DA. It is determined that it is mixed in. Furthermore, when error position n is found in the data string DP=m bits, it is determined that a correctable error is mixed in the information data DA.

Therefore, similarly to the first method, if an error within the correction capability range occurs, error correction processing is performed on the information data DA. Further, if an error exceeding the correction capability range has occurred, the error correction process of the information data DA is not performed and the process shifts to interpolation process. As a result, even if the digital signal DS with burst errors, etc. contains errors beyond the correction capability range, similar to the first method, the error correction processing required in the conventional example can be reduced as much as possible. becomes possible.

As a result, information data in which errors exceeding the correction capability range are interpolated based on data that has not been subjected to error correction processing, and as a result, it becomes possible to faithfully reproduce the information data.

〔Example〕

Next, embodiments of the present invention will be described with reference to the drawings.

2 to 5 are diagrams illustrating a digital signal error correction processing method according to an embodiment of the present invention.

(i) Description of First Embodiment FIG. 2 is a block diagram illustrating an error correction processing method for a MUSE signal according to a first embodiment of the present invention.

In the figure, 1] is an 82-bit shift register,
Focus deinterleaved input data (MUSE
The signal) inputs DI and shifts up the I focus one by one. Note that the audio data of the MUSE signal will be explained with reference to the data format shown in FIG.

12 has a syndrome calculation circuit and a parity calculation circuit 1
2a, a syndrome Sl calculation circuit 12b, and a syndrome Sl calculation circuit 12c.

The parity calculation circuit 12a uses a generator polynomial G (x) = (x
+1) (x' +x3+ 1) (x'+x"+x"
+x+1), the parity p is calculated from the first term (x+1).

The syndrome 51 calculation circuit 12b calculates the second term (x 7+
Syndrome 51=a is calculated from x3+1). 83 calculation circuit 12c calculates the third term (x7+x3
+x”+x+1) to calculate the syndrome S3.

13 is another calculation circuit, and syndrome 5l=a
5 which calculates the calculated value S3/Sl from the syndrome 51=a.
14 is an error number judgment circuit consisting of a 3151 calculation circuit 13b and an adder 13c which adds Sl' and S3/Sl and outputs an added value S.
=a and the output result data of the adder 13c are input, and the correction permission data D1 and the number of errors NE are output.

Table 1 shows the correction permission criteria of the determination circuit 14. this is,
Calculated values p, s1. from each calculation circuit 12.13. S3.
SL'' satisfies the discriminant in the table, and the relationship between the number of errors in input data DIN and permission/disapproval of correction is shown in a table.

(Left below) Table 1 15 is an error position detection circuit, and syndrome 51=a
It inputs the output result data of the adder 13c and outputs the error position n, the correction execution data D2, and the number NA of error positions. The detection circuit uses the error position equation F(x)=
bX” + aX+1 is the calculated value X = α, α indicating the error position
2. α3. Substitute ... in order. As a result, the solution when the equation becomes F(x)-0 is detected as the error position n of the data string. Furthermore, the number of solutions to the equation is detected as the number NA of error positions in the data string.

Up to this point, it is the same as in the conventional example, but in the first embodiment of the present invention, an error number/error position number determination circuit 16 is provided.

The number of errors/number of error positions determination circuit 16 receives the number of errors NE and the number of error positions NA and outputs comparison result data D3. When the comparison result data D3 is NE)NA, it means that an error exceeding the correction capability range of the error correction processing circuit concerned is mixed, and when NE=NA, the error correction of the error correction processing circuit concerned is included. The content contains errors that are within the range of ability.

17 is a three-input logic output circuit, and each data D1 to D3
The corrected data D5 is obtained by performing AND logical operation on the three inputs.
This outputs the following.

Reference numeral 18 denotes an adder, which inverts the focus of the error position of the input data DIN, which is sequentially shifted and amplified, based on the corrected data D5, and outputs output data DOT.

FIGS. 3(a) and 3(b) show the MU according to each embodiment of the present invention.
FIG. 3 is a data format diagram of an SE signal.

Figure (a) shows the MU in 4-channel mode (A mode).
This is the audio data format of the SE signal.

In the figure, the audio data of the MUSE signal consists of a 16-bit frame synchronization signal, a 22-bit control code, compressed difference data of 16 x 67 bits of audio, and a 16-bit correction code, and the 135-bit data is transmitted as one frame. It is something that This data is word, bit, and frame interleaved to efficiently correct errors during transmission.

For example, write processing is performed sequentially in the horizontal direction, and pig sending processing is performed sequentially in the frame direction.

Furthermore, according to the transmission standard of the MUSE signal, the write data string (127, 112) in the horizontal direction is
2°67) BCH is allowed. In this case, a correction code is added to the rear 15 bits of the 82-bit data string in the horizontal direction.

Figure (b) shows the MU in 2-channel mode (B mode).
This is the audio data format of the SE signal.

Note that the error correction/detection capabilities related to A and B modes are 1 bit error correction (SEC) and 2 bit error detection (DED) in normal mode, and 2 bit error correction (DEC) and 3 bit error detection (DED) in enhanced mode. Bit error detection (TED)
is possible.

FIG. 4 shows a flowchart of the MUSE signal error correction processing method according to the first embodiment of the present invention.

In the figure, when performing error correction processing on the enhanced mode MUSU signal (see Figure 3) in which a 15-pin error correction code is added to the audio data di-d32, first, in step P1, the syndrome of the input data DIN is calculated. Process. At this time, the input data (MUSE signal) DIN subjected to the bit deinterlip processing is input to the 82-bit shift register 11 and the syndrome calculation circuit 12.

In the parity calculation circuit 12a, a generator polynomial G(x)=(x+1
)(x'+x'+1)(x'+x"+x"+ x
+ 1 ), the parity p is calculated from the first term (x+1). The syndrome Sl calculation circuit 12b calculates the syndrome 5L=a from the second term (X stroke+x'+1). In the 83 calculation circuit 12C, the third term (x'+x''+
Syndrome S3 is calculated from x''+x+1).

In the other calculation circuit 13, a calculation value S12 is calculated from the syndrome 5l=a, and a calculation value 33/Sl is calculated from the syndrome 51=a. Also, the calculated value st” and 5
3151 are added and the addition [b is output.

In the next step, step P2, the number of errors NE is determined based on the syndrome calculation results.

At this time, the error number determination circuit 14 receives the parity p-n drome 51=a and the output result data of the adder 13c, and outputs the correction permission data Di and the error number NE. Its determination ability is capable of 2-bit error correction and 3-bit error detection in enhanced mode.

Next, in parallel, in step P3, the number NA of error positions in the input data Dllll is detected. At this time, in the error position detection circuit 15, the syndrome 51=a and the adder 1
3c is input, and the error position n, correction execution data D2, and number NA of error positions are output. Also,
In the circuit, the error position equation F(x) = bX' + aX
Calculated value X=α, α2 . which indicates the error position at +1. α...
・are substituted in order.

As a result, the solution when the equation F(x)=0 is detected as the error position n of the data string. Furthermore, the number of solutions to the equation is detected as the number NA of error positions in the data string.

Next, in step P4, the number of errors NE in the human data DIN is calculated.
and the number of error positions NA. At this time, the number of errors/number of error positions determining circuit I6 receives the number of errors NE and the number of error positions NA and outputs comparison result data D3. In addition, when the comparison result data D3 is NE+NA, the content indicates that an error exceeding the correction capability range of the error correction processing circuit is mixed. Furthermore, if NE=NA, it means that an error within the correction capability range of the error correction processing circuit is mixed.

Furthermore, if the comparison processing result is NE=NA in step P5, error correction processing is performed at error position n of the input data DIN. At this time, the three-input logic output circuit 17 performs three-input AND logic operation processing on each of the data D1 to D3, and outputs corrected data D5. Moreover, in the adder 18,
Input data DTN that is sequentially shifted and amplified is inverted based on the corrected data D5, and output data DOT is output by inverting the bit at the error position.

Alternatively, if the comparison processing result NE=#NA in step P6, the process proceeds to interpolation processing without performing error correction processing at error position n of input data DIN.

In this way, according to the first error correction method of the present invention,
In step P5, error correction processing is not performed on the error position n of the input data DIN based on the comparison processing result between the number of errors NE and the number NA of error positions, or in step P6, error correction processing is not performed on the input data DIN based on the comparison processing result. Interpolation processing is performed on .

For this reason, if a burst error occurs for some reason, the M
Even if the USE signal contains an error that exceeds the correction capability range, it is possible to reduce the power of the error correction process that erroneously corrects a bit in the correct position as in the conventional example.

As a result, the input data DIN in which errors exceeding the correction capability range are interpolated based on correct data that has not been subjected to error correction processing, making it possible to faithfully reproduce the audio data.

(11) Description of Second Embodiment The fifth leap line shows a blank diagram for explaining the MUSE signal error correction processing method according to the second embodiment of the present invention.

In the figure, the difference from the first embodiment is that in the second embodiment, a shortening condition determination circuit 19 is provided at the output stage of the error position detection circuit 15.

That is, the shortening condition determination circuit 19 determines that the error position n is 46.
It determines whether or not it has occurred with more than one bit, and outputs determination result data D4.

This is 12 times as shown in the data format in Figure 3.
If 82 bits of data are allowed in a data string standardized to 7 bits, rQ with no data in bits 1 to 45
, the error position is determined using the 45th and 46th bits of the data string as boundary standards.

In addition, 20 is a four-manpower logic output circuit, and each data D1
~D304 performs manual AND logical operation processing and outputs corrected data D6.

Other components with the same names and symbols have the same functions, so their explanations will be omitted.

FIG. 6 shows a flowchart illustrating the MUSE signal error correction processing method according to the second embodiment of the present invention.

In the figure, the difference from the first embodiment is that in the correction processing flow according to the second embodiment, input data D is
The error position n of IN is determined by adding a shortening condition.

That is, similarly to steps P1 to P3 of the first embodiment, step P4 is performed through steps P1 to P3 of the second embodiment.
, the input data DIN is input in step P4.
A shortening condition n≧46 is introduced for the error position n of , and a comparison process is performed between the number of errors NE and the number of error positions NA.

At this time, the data string DP in which the MUSE signal is standardized is
Data string DP that is allowed to be transmitted for 127 bits = 8
When shortened to 2 bits, the data string D P =127
When an error level iin≦45 is found in -82 bits, it is determined that an uncorrectable error is mixed in the input data DIN. Furthermore, when an error position n≧46 is found in the data string D P =82 bits, it is determined that a correctable error is mixed in the input data DIN.

Next, in step P5, the shortening condition n≧46. If the comparison result is NE=NA, error correction processing is performed at error position n of input data DIH.

Alternatively, in step P6, the shortening condition n≦45. Alternatively, if the comparison result is NE≠NA, the interpolation process is performed without performing the correction process for the error position In of the input data DIN.

In this way, according to the second error correction processing method of the present invention, when the input data DIN is shortened from 127 bits of the data string of the standardized MUSE signal, the shortening condition n≧46 and the error position The data DIN is subjected to error correction processing based on the detection processing result of n.

Therefore, similarly to the first method, when an error within the correction capability range occurs, input data DIN error correction processing is performed. Furthermore, if an error exceeding the correction capability range has occurred, the error correction process of the input data DIN is not performed and the process shifts to interpolation process.

With this, as in the first method, even if the MUSE signal that has caused a burst error etc. contains an error that is beyond the correction capability range, the error correction process as in the conventional example can be reduced as much as possible. becomes possible.

As a result, the input data DIN in which errors exceeding the correction capability range are interpolated based on data that has not undergone error correction processing, making it possible to faithfully reproduce the information data. , Table 2 shows the conventional example, No. 1., based on the experimental results of the present inventors. This is a comparison of the error correction processing method of the second embodiment.

The experimental conditions were MUSE audio data format audio mode A mode and correction code enhanced mode.

The basic data is silent, the number of data is 5000 frames,
The error correction processing methods of the conventional example and each embodiment are compared when an error bit is inserted.

Table 2 According to this, compared to the conventional example, 1. In the second embodiment, it is possible to reduce the number of erroneous corrections by half.

Although the embodiments of the present invention have been described with respect to the MUSE signal, similar effects can be obtained in a communication system in which information data is multiplexed, for example, error correction processing of packet communication data to which error bits are added.

〔Effect of the invention〕

As described above, according to the present invention, the error correction process for the error position of the input data or the interpolation process is performed without performing the error correction process based on the result of the comparison process between the number of errors and the number of error positions.

Therefore, even if the information data that has caused a burst error or the like for some reason contains an error that is beyond the correction capability, the error correction process that incorrectly corrects the bit in the correct position as in the conventional example is not performed. It becomes possible to reduce it as much as possible.

Furthermore, according to the present invention, when the data is shortened from a standardized information data string, it is possible to perform error correction processing on the data based on the shortening conditions and the detection processing result of the error position n.

This makes it possible to faithfully reproduce information data, which greatly contributes to improving the reliability of the digital signal processing device.

[Brief explanation of drawings]

FIG. 1 is a principle diagram of the digital signal error correction processing method according to the present invention; Part 2 is a block diagram illustrating the MUSE signal error correction processing method according to the first embodiment of the invention; 4 is a flow chart of the error correction processing method according to the first embodiment of the present invention. FIG. 5 is a data format diagram of the MUSE signal according to each embodiment of the present invention. FIG. 6 is a flowchart of the error correction processing method according to the second embodiment of the present invention, and FIG. FIG. 8 is a block diagram illustrating a signal error correction processing method. FIG. 8 is a flowchart of a conventional error correction processing method. (Explanation of symbols) D'S...Digital signal, NE...Number of errors, NA...Number of error positions, DA...Information data, DP...Standardized data string, EB...・Error correction code, n...Error rank 10

Claims (2)

[Claims] (1) Error correction processing of a digital signal (DS) in which an error correction code (EB) is added to information data (DA), which performs error detection calculation processing on the digital signal (DS), and calculates the result of the calculation processing. The number of errors (NE) is determined and the number of error positions (NA) is detected based on the number of errors (N
E) and the number of error positions (NA), and based on the result of the comparison process, perform error correction processing on the error position (n) of the information data (DA) or interpolation processing without performing the error correction processing. A digital signal error correction processing method characterized by: (2) In the digital signal error correction processing method according to claim 1, the information data (DA) is a standardized digital signal (
When the data string (DP) of the digital signal (DS) is shortened, the conditions for shortening the digital signal (DS) and the error position (n)
A digital signal error correction processing method, characterized in that error correction processing is performed on the information data (DA) based on the detection processing result.