JPH04371026A - Burst error correcting system - Google Patents

Abstract

PURPOSE:To correctly perform the error correction at the time of the occurrence of burst error deeper than the interleave depth in the burst error correcting system of a digital data communication system. CONSTITUTION:Communication data received by a receiver 10 is decoded by a deinterleave means 11. If error correction of decoded communication data is performed at random by a burst error detector 13, the burst occurence of error is forecasted, and burst error is detected. If burst error is detected, a part of forecasted burst error is corrected by a burst error correcting device 30, and error correction of decoded communication data is performed again by an error correcting device 12. Consequently, communication data which has burst error deeper than the interleave depth is correctly corrected.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a burst error correction method for a digital data communication system, and more particularly to a burst error correction method for correcting an error when a burst error larger than the interleave depth occurs.

[0002] In a digital data communication system using radio, there is a possibility that a transmitted signal may be erroneously received due to noise in the radio transmission. If the transmission signal is received incorrectly, the signal itself has no meaning; on the contrary, the erroneous signal has a large influence on others. For this reason, there are several methods for correcting errors in transmission signals. For example, a continuous transmission method that repeatedly transmits a transmission signal multiple times,
A return verification method in which the received transmission signal is returned to the sending side as is, the sending side checks whether the transmitted signal is correct, and if it is incorrect, it is sent again.The receiving side only checks whether there is an error, and if an error occurs. An error detection retransmission method (automatic
tic request (ARQ), and an error correction method (forward err.
or correction: FEC).

0003

[Prior Art] Conventionally, in order to improve communication quality, among these methods, the method is generally used because of its advantages such as not requiring a return communication path, easy control as a system, and small delay. Error correction methods have been used. This error correction method was able to correct communication data in which errors occurred randomly. However, errors in communication data often occur in burst-like clusters due to a fading phenomenon in which the received electric field strength changes over time. Ordinary error correction codes cannot correctly correct data for such errors that occur in bursts. Therefore, by interleaving a suitable number of bits of communication data and using this interleaved communication data, errors occurring in bursts have been dealt with.

0004

However, since the interleaving depth is set at the time of designing a digital communication system based on a trade-off between error correction capability and transmission delay, it cannot cope with burst lengths that change from time to time. Therefore, for errors with a burst length greater than the interleaving depth, there is a problem in that correct data is corrected when correcting it, that is, erroneous correction is performed. Furthermore, there is also the problem that errors propagate due to erroneous correction, and the number of errors increases due to error correction.

[0005] The present invention has been made in view of the above points, and uses an error correction method to correct correctly even when an error with a burst length larger than the interleave depth occurs in communication data. The purpose of this invention is to provide a burst error correction method that can perform the following steps.

[0006] Another object of the present invention is to provide a burst error correction method that allows codes in interleaved blocks to have error correction and error detection capabilities, and which can more accurately correct communication data with burst errors. It is to be.

Still another object of the present invention is to provide a burst error correction system that can detect burst errors from syndromes and correct communication data containing burst errors.

[0008]

[Means for Solving the Problems] FIG. 1 is a first principle explanatory diagram of the present invention that achieves the above object. In the figure, communication data received by a receiver 10 is interleaved, and first, deinterleaving is performed by a deinterleaving means 11. Next, error correction device 1
Error correction of the communication data received by 2 is performed,
If this error correction is performed randomly, it is predicted that a burst error has occurred, and a burst error is detected. If a burst error is detected, the burst error correction device 30 partially corrects the predicted burst error in the communication data and outputs it to the signal processing device 20. The signal processing device 20 performs signal processing on the corrected communication data.

FIG. 2 is a second diagram illustrating the principle of the present invention that achieves the above object. In the figure, communication data received by receiver 10 is interleaved;
The codes within the interleaved block have error correction and error detection code capabilities. First, deinterleaving is performed by the deinterleaving means 11. Next, error correction is performed on the received communication data by the error correction detection device 14, and the bits in each error correction code are matched with other correctly decoded codes to predict that a burst of errors has occurred. , a burst error is detected. If a burst error is detected, the burst error correction device 30 partially corrects the predicted burst error in the communication data and outputs it to the signal processing device 20. The signal processing device 20 performs signal processing on the corrected communication data.

FIG. 3 is a diagram illustrating the third principle of the present invention that achieves the above object. In the figure, communication data received by receiver 10 is interleaved;
First, deinterleaving is performed by the deinterleaving means 11. Next, error correction of the received communication data is performed by the error correction device 12, and syndromes are determined. This syndrome predicts burst errors and detects the burst error position. If a burst error is detected, the syndrome error correction device 31 corrects a portion of the predicted burst error in the communication data and outputs it to the signal processing device 20. The signal processing device 20 performs signal processing on the corrected communication data.

[0011]

In FIG. 1, when error correction is performed randomly by the error correction device 12, it is predicted that a burst error will occur, and the burst error will be detected. If a burst error is detected, the burst error correction device 30 corrects the code predicted to be a burst error. Then, the error correction of the communication data corrected by the error correction device 12 is performed again. Therefore, even communication data in which an error with a burst length greater than the interleaving depth has occurred can be correctly corrected.

In FIG. 2, an error correction detection device 14 predicts burst errors by matching the bits in each error correction code with other correctly decoded codes, and detects burst errors. If a burst error is detected, the burst error correction device 30 corrects the code predicted to be a burst error. Then, the error correction of the communication data corrected by the error correction device 12 is performed again. Therefore, even communication data in which an error with a burst length greater than the interleaving depth has occurred can be correctly corrected.

In FIG. 3, syndromes are determined by the error correction device 12. This syndrome predicts burst errors and detects the burst error position. If a burst error is detected, the syndrome error correction device 31 corrects the code predicted to be a burst error. Then, the error correction of the communication data corrected by the error correction device 12 is performed again. Therefore, even communication data in which an error with a burst length greater than the interleaving depth has occurred can be correctly corrected.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram illustrating the first principle of the burst error correction method of the present invention. In the figure, a receiver 10 receives communication data. Communication data to be sent and received is usually transmitted using BCH (Bose-Chaudh) as a single error correction code.
uri-Hocquenghem) codes, especially BCH (
7,4) code is used. Deinterleaving means 11 decodes the interleaved communication data into the original code. The error correction device 12 performs error correction on the decoded communication data, predicts that a burst error has occurred when this error correction is performed randomly, and detects the burst error. Here, a state in which error correction is performed randomly means a state in which error-corrected code symbols are randomly distributed. In addition, to detect random correction, the code after error correction is returned to the code symbol order before deinterleaving, and it is checked whether the correction location is a burst or random. The signal processing device 20 performs signal processing on the corrected communication data. The burst error correction device 30 corrects a portion of the predicted burst errors in the communication data in which the burst errors have been detected.

Next, the operation will be explained based on the first principle explanatory diagram. First, communication data received by receiver 10 is interleaved. Therefore, it is necessary to decode the interleaved communication data into the original code, so the received communication data is deinterleaved by the deinterleaving means 11. Then, the error correction device 12 performs error correction on the decoded communication data, and if this error correction is performed randomly, that is, if the error-corrected code symbols are dispersed, burst errors It is predicted that a burst error has occurred, and a burst error is detected.

If a burst error is detected,
The burst error correction device 30 partially corrects the predicted burst errors. This correction is performed by inverting the first or last few bits of the bits predicted to be burst errors so that the burst length becomes the interleave depth. The burst length is the length of consecutive erroneous bits. And error correction device 12
Error correction of the communication data corrected by is performed again. Finally, the corrected communication data is sent to the signal processing device 20.
Signal processing is performed by

FIG. 4 is a diagram showing an example of transmission data. As shown in the figure, the transmission data 50 is D0 to D6, D
7 ~D13, D14~D20,..., D35~D4
A bit string corresponding to a "row" such as 1 is a data unit stored in a memory or the like. Also, D0 to D35, D1
~D36, D2 ~D37,..., D6 ~D41
A bit string corresponding to a "column" such as ``column'' is a data unit for interleaving. In the figure, since the number of bit strings corresponding to a "column" is six, the interleaving depth is six. The bit string when transmitting interleaved communication data is as follows.

D0 D7 D14D21D28D35D
1 D8 D15D22D29D36...D2
6D33D40D6 D13D20D27D34D41

FIG. 5 is a diagram showing an example of received data. This received data 51 represents communication data received by receiver 10 and deinterleaved by deinterleaving means 11 in FIG. In the figure, DX
indicates a normally received code, and EX indicates an incorrectly received code.

FIG. 6 is a diagram showing an example of error-corrected data. This error-corrected data 52 is data obtained by performing error correction on the received data 51 in FIG. 5 using a BCH (7, 4) code and decoding it. From the comparison with Figure 5, E8, E9
, E15 and E16 are not corrected, and conversely E1
It can be seen that 3 and E20 are incorrectly corrected. Therefore, the data in the second and third rows as data units to be processed later were not decoded normally.

FIG. 7 is a diagram showing an example of data in which a portion of predicted burst errors are corrected. FIG. 1 shows data corrected by the burst error correction device 30. Since E13 and E20 are corrected in the error correction data 52 of FIG. 6, error correction is performed randomly. At this time, the burst error correction device 30 predicts that a burst error has occurred and corrects E9 and E1 of the original received data.
Make corrections in 6.

FIG. 8 is a block diagram showing a second embodiment of the present invention. In the figure, a receiver 10 receives communication data. For communication data to be transmitted and received, a BCH code, particularly a BCH (7,4) code, is usually used as a single error correction code. Furthermore, in order to check whether there is an error in the result despite performing single error correction using the BCH (7, 4) code, CRC (Cyclic R
A code symbol for performing an edundancy check is also added. Deinterleaving means 11 decodes the interleaved communication data into the original code. The error correction device 12 performs error correction on decoded communication data. The burst error detection device 13 predicts that a burst error has occurred and detects a burst error when a certain number of code symbols that have undergone error correction are gathered. The signal processing device 20 performs signal processing on the corrected communication data. The burst error correction device 30 corrects a portion of the predicted burst errors in the communication data in which the burst errors have been detected.

Next, the operation of the second embodiment will be explained. First, communication data received by receiver 10 is interleaved. Therefore, it is necessary to decode the interleaved communication data into the original code, so the received communication data is deinterleaved by the deinterleaving means 11. The communication data decoded by the error correction device 12 is
Single error correction is performed using an H(7,4) code. Furthermore, even though single error correction is performed using the BCH (7,4) code, if an error is detected by CRC and a certain number of erroneous code symbols have gathered, the burst error detection device 13 detects burst errors. It predicts that a burst error has occurred and detects the burst error.

If a burst error is detected,
The burst error correction device 30 partially corrects the predicted burst errors. This correction is performed by inverting the first two bits or the last two bits of the predicted burst error bits so that the burst length becomes the interleave depth. Then, error correction of the re-corrected communication data is performed again by the error correction device 12. Finally, the corrected communication data is subjected to signal processing by the signal processing device 20. Note that the data processed in the second embodiment and the transition of data are the same as those in the first embodiment, so a description thereof will be omitted.

FIG. 9 is a block diagram showing a third embodiment of the present invention. In the figure, a receiver 10 receives communication data. For communication data to be transmitted and received, a BCH code, particularly a BCH (7,4) code, is usually used as a single error correction code. Furthermore, a code for performing CRC is also added in order to check whether there is an error in the result despite single error correction using the BCH (7,4) code. Deinterleaving means 11 decodes the interleaved communication data into the original code. The error correction device 12 performs error correction on decoded communication data. The burst error detection device 13 predicts that a burst error has occurred when a certain number of error-corrected bits have been collected, and detects the burst error. Signal processing device 2
0 performs signal processing of corrected communication data. The burst error correction device 30 corrects a portion of the predicted burst errors in the communication data in which the burst errors have been detected.

Next, the operation of the third embodiment will be explained. First, communication data received by receiver 10 is interleaved. Therefore, it is necessary to decode the interleaved communication data into the original code, so the received communication data is deinterleaved by the deinterleaving means 11. The communication data decoded by the error correction device 12 is
Single error correction is performed using an H(7,4) code. Here, if error correction is performed randomly, the error correction device 12 predicts that a burst error has occurred, regardless of whether a long burst error has a higher probability of occurring than an error with a short burst length. , performs burst error detection.

If a burst error is detected,
The burst error correction device 30 partially corrects the predicted burst errors. This correction is performed by inverting the first two bits or the last two bits of the predicted burst error bits so that the burst length becomes the interleave depth. Then, the error correction of the communication data corrected by the error correction device 12 is performed again. Further, the burst error detection device 13 performs detection of burst error correction. Finally, the corrected communication data is subjected to signal processing by the signal processing device 20. Note that the data processed in the third embodiment and the data transition are the same as those in the first embodiment, so a description thereof will be omitted.

FIG. 2 is a second principle explanatory diagram (block diagram showing an embodiment) of the burst error correction system of the present invention. In the figure, a receiver 10 receives communication data. For communication data to be transmitted and received, a BCH code, particularly a BCH (7,4) code, is usually used as a single error correction code. Furthermore, in order to detect which of the individual error correction codes are incorrectly corrected, the codes within the interleaved block of communication data are provided with error correction and error detection capabilities, such as single error correction and double error detection capabilities; to have. Deinterleaving means 11 decodes the interleaved communication data into the original code. The error correction detection device 14 performs error correction detection on decoded communication data, and when error detection is performed within each error correction code, errors occur in bursts in conjunction with other correctly decoded code symbols. burst error is detected. The signal processing device 20 performs signal processing on the corrected communication data. The burst error correction device 30 corrects a portion of the predicted burst errors in the communication data in which the burst errors have been detected.

Next, the operation will be explained based on a second principle explanatory diagram (block diagram showing an embodiment). First, communication data received by receiver 10 is interleaved. Therefore, it is necessary to decode the interleaved communication data into the original code, so the received communication data is deinterleaved by the deinterleaving means 11. Then, error correction is performed on the decoded communication data by the error correction detection device 14, and the bits in each error correction code are matched with other correctly decoded codes to predict that a burst of errors has occurred, Burst error detection is performed.

If a burst error is detected,
The burst error correction device 30 partially corrects the predicted burst errors. This correction is performed by inverting the first two bits or the last two bits of the predicted burst error bits so that the burst length becomes the interleave depth. Then, the error correction of the communication data corrected by the error correction device 12 is performed again. Finally, the corrected communication data is subjected to signal processing by the signal processing device 20. Note that the data processed in the fourth embodiment and the data transition are the same as those in the first embodiment, so a description thereof will be omitted.

FIG. 3 is a third principle explanatory diagram (block diagram showing an embodiment) of the burst error correction system of the present invention. In the figure, a receiver 10 receives communication data. For communication data to be transmitted and received, a BCH code, particularly a BCH (7,4) code, is usually used as a single error correction code. Deinterleaving means 11 decodes the interleaved communication data into the original code. The error correction device 12 performs error correction on the decoded communication data, obtains a syndrome, predicts a burst error based on the syndrome, and detects a burst error position. The signal processing device 20 performs signal processing on the corrected communication data. The burst error correction device 30 corrects a portion of the predicted burst errors in the communication data in which the burst errors have been detected.

Next, the operation will be explained based on the third principle explanatory diagram (block diagram showing the embodiment). First, communication data received by receiver 10 is interleaved. Therefore, it is necessary to decode the interleaved communication data into the original code, so the received communication data is deinterleaved by the deinterleaving means 11.

The error correction device 12 then performs the following processing. That is, error correction of the decoded communication data is performed. Next, the syndrome is determined using the following procedure. For example, the modulus polynomial is "X3 +X+1"
and the generating polynomial is defined as "X3 +X+1". At this time, the remainder when the received data is divided by the generator polynomial is the syndrome.

For example, the received data is a bit string "011
1000", the receiving polynomial is "X5 +X
4 +X3''. Therefore, X5 +X4 +X3 = (X3 +X+1) (X2
+X)+X, so the remainder when the receiving polynomial is divided by the generating polynomial is X. Therefore, the syndrome of received data in this case is α1.

At this time, burst error prediction is performed as follows using addition operations from the Galois field operation table, which will be described later. That is, if the syndrome α1 as shown in the above example is obtained, it has the same meaning as if two syndromes α5 and α6 were obtained. If α1
If the single bit error correction obtained with the syndrome α5 and α6 is a random error, and the two bit errors obtained with the syndromes α5 and α6 are burst errors, the error correction device 12 predicts that a burst error has occurred. , detection of the burst error position is performed.

If a burst error is detected,
Regarding some of the burst errors predicted by the burst error correction device 30, in the above example, bits corresponding to α5 and α6 are re-corrected. Then, error correction of the re-corrected communication data is performed again by the error correction device 12. Finally, the corrected communication data is subjected to signal processing by the signal processing device 20.

FIG. 10 is a diagram showing a Galois field operation table. The Galois field arithmetic table 60 is used to predict the code position of a burst error from the syndrome when error correction based on the syndrome is performed randomly. For example, if the syndrome of the corrected communication data is α
0, there are four α0s in the lower left half of the Galois field operation table 60, namely 61, 62, 63 and 6
There is an α0 of 4. These addends of α0 are 0, α0, α1 from the row and column of Galois field operation table 60, respectively.
and α3, α2 and α6, and α4 and α5. By the way, when a burst error occurs, several bits of errors occur consecutively, so the code positions, that is, the exponent of α, should also be consecutive. Therefore, in the above example, since α4 and α5 are consecutive exponents, it can be predicted that the error occurs at these code positions as the code positions of the burst error. Here, the predicted error positions are D0 to D6, D7 in the communication data 50 of FIG.
~D13, D14~D20,..., D35~D41
Each of the "rows" of is α6 , α5 , α4 ,...
, α0.

In the above explanation, the interleaving depth is 6.
However, any interleaving depth set at the time of designing the digital data communication system may be used. Furthermore, although the BCH (7,4) code is used for single error correction, other BCH codes may also be used. And B.C.H.
In addition to the codes, error correction codes such as Hamming codes, Fire codes, and bulk codes may be used.

Although CRC was used as an error check after performing single error correction using the BCH code, other error checking methods, such as error detection codes such as parity check codes and constant weight codes, may also be used. .

[0040]

As explained above, in the present invention, the burst error detection device detects burst errors in communication data, and if a burst error is detected, the burst error correction device corrects the errors in the communication data. Since the configuration is configured such that burst errors are predicted and corrected, and the error correction of communication data decoded by the error correction device is performed again, communication data with burst errors can be correctly corrected.

Furthermore, by providing error correction and error detection capabilities to the codes in the interleaved blocks of communication data, it is possible to improve the ability to correctly correct communication data with burst errors.

Furthermore, the communication data received by the burst error correction device is subjected to error correction, a syndrome is obtained, a burst error is predicted using this syndrome, a burst error position is detected, and the communication data is corrected by the burst error correction device. The structure is configured so that burst errors are predicted and corrected from the error-corrected code, and the error correction of the communication data decoded by the error correction device is performed again, thereby further improving the ability to correctly correct communication data with burst errors. can be improved.

[Brief explanation of drawings]

FIG. 1 is a first principle explanatory diagram of the burst error correction method of the present invention.

FIG. 2 is a second principle explanatory diagram of the burst error correction method of the present invention.

FIG. 3 is a diagram illustrating the third principle of the burst error correction method of the present invention.

FIG. 4 is a diagram showing transmission data.

FIG. 5 is a diagram showing received data.

FIG. 6 is a diagram showing data subjected to error correction.

FIG. 7 is a diagram showing data in which a portion of predicted burst errors are corrected.

FIG. 8 is a block diagram showing a second embodiment of the present invention.

FIG. 9 is a block diagram showing a third embodiment of the present invention.

FIG. 10 is a diagram showing a Galois field operation table.

[Explanation of symbols]

10 Receiver 11 Deinterleaving means 12 Error correction device 13 Burst error detection device 14 Error correction detection device 20 Signal processing device 30 Burst error correction device 31 Syndrome error correction device

Claims (3)

[Claims] 1. A burst error correction system for a digital data communication system, comprising: a receiver (10) for receiving interleaved communication data; a deinterleaving means (11) for deinterleaving the received communication data; an error correction device (12) that performs error correction on the interleaved communication data, predicts that errors occur in bursts when the error correction is performed randomly, and detects burst errors;
), a burst error correction device (30) that corrects a part of the predicted burst error of the communication data when the burst error is detected, and a signal processing device that performs signal processing of the corrected communication data. (20) A burst error correction method characterized by comprising the following. 2. In a burst error correction method for a digital data communication system, codes in an interleave block are provided with error correction and error detection capabilities, and a receiver (1
0), a deinterleaving means (11) for deinterleaving the received communication data, and a deinterleaving means (11) for performing error correction on the deinterleaved communication data, and correcting the bits in each error correction code by other correctly corrected bits. an error correction detection device (14) that predicts that an error has occurred in bursts in accordance with the decoded code and detects the burst error; and if the burst error is detected, the predicted burst error of the communication data is detected; A burst error correction system comprising: a burst error correction device (30) that corrects a part of the communication data; and a signal processing device (20) that performs signal processing of the corrected communication data. 3. In a burst error correction system for a digital data communication system, a receiver (10) that receives interleaved communication data, a deinterleaving means (11) that deinterleaves the received communication data, and a receiver (10) that receives interleaved communication data; an error correction device (12) that performs error correction on the interleaved communication data, obtains a syndrome, predicts a burst error based on the syndrome, and detects a burst error position; , a burst error correction device (31) that corrects the code of the communication data from the burst error position, and a signal processing device (20) that performs signal processing of the corrected communication data. Error correction method.