KR20070012928A - Error detecting method of modulation decoder and error control method for optical data storage systems - Google Patents

Abstract

An error detection method of a modulation decoder for an optical data storing apparatus and an error control method therefor are provided to make an error control decoder have improved error correction capability in data reproduction by using an error characteristic in a decoding process of modulation codes used in an existing DVD(Digital Video Disc) device. Inter-different codewords of 250, 254, 243, and 254 exist in states 1 to 4 of EFMPlus modulation codes for a DVD device. Repeated codewords of 94, 90, 101, and 90 exist. When one of the codewords is inputted according to the respective states in a decoding process for an input codword in the corresponding state, the inputted codeword is recognized as an incorrect codeword and error information is transmitted.

Description

ERROR DETECTING METHOD OF MODULATION DECODER AND ERROR CONTROL METHOD FOR OPTICAL DATA STORAGE SYSTEMS}

1 is a block diagram illustrating an error control method for an optical data storage device according to the present invention.

2 shows the characteristics of the EFMPlus code used in the DVD device.

Figure 3 illustrates a common code capable of error detection during the EFMPlus decoding process for use in the present invention.

Figure 4 illustrates a common code capable of error detection during the EFMPlus decoding process for use in the present invention. (Continued in FIG. 3)

FIG. 5 illustrates a code for each state capable of error detection in an EFMPlus decoding process for use in the present invention.

FIG. 6 illustrates a code for each state capable of error detection in an EFMPlus decoding process for use in the present invention. (Continued in FIG. 5)

FIG. 7 illustrates codes for each state capable of error detection in an EFMPlus decoding process for use in the present invention. (Continued in FIG. 5)

8 illustrates a code for each state capable of error detection in an EFMPlus decoding process for use in the present invention. (Continued in FIG. 5)

FIG. 9 illustrates codes for each state capable of error detection in an EFMPlus decoding process for use in the present invention. (Continued in FIG. 5)

FIG. 10 illustrates a code for each state capable of error detection in an EFMPlus decoding process for use in the present invention. (Continued in FIG. 5)

FIG. 11 shows codes for each state capable of error detection in an EFMPlus decoding process for use in the present invention. (Continued in FIG. 5)

12 illustrates a code for each state capable of error detection in an EFMPlus decoding process for use in the present invention. (Continued in FIG. 5)

FIG. 13 illustrates a common code capable of error detection during decoding of an ETM code used in an HD-DVD device.

FIG. 14 shows a code A capable of error detection in the decoding process of an ETM code used in an HD-DVD apparatus.

FIG. 15 illustrates a code B capable of error detection in the decoding process of an ETM code used in an HD-DVD apparatus.

FIG. 16 illustrates a code B capable of error detection in the decoding process of an ETM code used in an HD-DVD apparatus. (Continued in FIG. 15)

FIG. 17 illustrates an error correction method using error detected by a modulation decoder of an optical storage device as erasure information.

18 shows a graph comparing error correction performance.

end. Problems of the Prior Art

  Conventional and next generation DVD devices related to the present invention generally use modulation codes to eliminate inter-symbol interference that may occur in optical discs, and use error correction codes to correct errors generated in the channel.

  In conventional DVD devices, the EFMPlus code is used as a modulation code, and the Reed-Solomon Product Code (RSPC) is used as an error correction code. When a certain noise is inserted into the optical disk or the playback speed increases, the probability of error is increased by receiving data having a relatively low signal-to-noise ratio during playback, and at the same time, the length of consecutive error It may be increased to exceed the allowable correction capability in the decoder for error correction. This makes reliable data reproduction impossible. Nevertheless, due to the limitation of the physical data format, it is impossible to change the structure of the modulation code and the error correction code to improve the error correction capability.

  Similar patents and background documents related to the present invention and description thereof are as follows.

(1) technical source

① Similar patent

Patent [1]: The overall technology related to the invention (design) relating to modulation codes used in conventional DVD devices is described in US Pat. No. 5,696,505 (METHOD OF CONVERTING A SERIES OF M-BIT INFORMATION WORDS TO). A MODULATED SIGNAL, METHOD OF PRODUCING A RECORD CARRIER, CODING DEVICE, DECODING DEVICE, RECORDING DEVICE, READING DEVICE, SIGNAL, AS WELL AS RECORD CARRIER, 9 December 1997).

Patent [2]: The overall technology associated with the present invention (design) relating to modulation codes used in conventional DVD devices is described in US Pat. No. 6,195,778, registered by PT Tran (DEMODULATION OF DVD CODEWORDS USING DEPENDENCY-SORTED TABLES FOR DUPLICATE DEPENDENT AND UNIQUE / NON-DEPENDENT MAPPINGS, February 27, 2001.

Patent [3]: The overall technology related to the invention (design) relating to modulation codes and error correction codes used in next-generation DVD devices is described in US Patent US 2005 / 0058034A1 (INFORMATION RECORDING MEDIUM, filed by H. Ando, S. Ashida). METHOD OF RECORDING INFORMATION THERETO, AND INFORMATION RECORDING / REPRODUCING APPARATUS, March 17, 2005).

Patent [4]: The overall description related to the present invention (modification) regarding modulation code and error correction code used in next generation DVD devices is described in US Patent US2004 / 0246863A1 filed by H. Ando, C. Noda, T. Kojima. INFORMATION RECORDING MEDIUM, INFORMATION REPRODUCING APPARATUS, AND INFORMATION RECORDING AND REPRODUCING APPARATUS (9 December 2004).

② Background literature

Technique [1]: A description of the modulation code used in a conventional DVD device is described in IEEE Transactions on Consumer Electronics / 1/1995, P491-495 published by K. A. S. Immink.

(2) Name of prior art (conventional technology composition)

① EFMPlus Coding Technology

  The EFMPlus encoder consists of an 8-bit input, 16-bit output, and four states. When the number of minimum and maximum zeros between 1 and 1 is d and k, respectively, each codeword satisfies the condition of (d = 2, k = 10). At this time, each state and codeword are configured as follows.

If the number of '0's at the end of the codeword is one or less, the next state is one.

If the number of '0's at the end of the codeword is 2 or more and 5 or less, the next state is 2 or 3.

If the number of '0's at the end of the codeword is 6 or more and 9 or less, the next state is 4.

In addition, since the condition of (d = 2, k = 10) must be satisfied even when the codeword and another codeword are connected, the beginning of the codeword in the state 1 is the fastest 2 and the maximum 9 ' 0 ', and in case of state 4, a maximum of one' 0 'is included.

  Codewords with the following states 1 and 4 do not have the same codeword. However, if a codeword having a next state of 2 or 3 has duplicate codewords, it checks the bit values of the first and 13th positions of the next following incoming codeword, and if the state is both '0', , Otherwise, state 3.

The encoder consists of three parts: input, output and state. This encoding process is defined by two logical functions: output function and next state function. When a particular codeword c _t transmitted by the encoder at any instant t is a function of the source word b _t input to the encoder, the codeword is determined by the particular state s _t of the encoder. Also, the next state s _{t + 1} corresponding to the instant _{t + 1} is expressed as a function of c _t and s _t . That is, the output function h (.) And the next state function g (.) Can be briefly expressed as follows.

The output function h (.) And the next state function g (.) Are both represented by four items with 351 entries. Some of this is shown in Table 1.

Table 1: Part of the EFMPlus Coding Table

i h (i, 1), g (i, 1) h (i, 2), g (i, 2) h (i, 3), g (i, 3) h (i, 4), g (i, 4) 0 1 2 3 4 5 6 7 8 0010000000001001,1 0010000000010010,1 0010000100100000,2 0010000001001001,2 0010000010010000,2 0010000000100100,2 0010000000100100,3 0010000001001000,3 0010000010010000,3 0100000100100000,2 0010000000010010,1 0010000100100000,2 0100010010000000,4 0010000010010000,2 0010000000100100,2 0010000000100100,3 0100000000010010,1 0010000010010000,3 0010000000001001,1 1000000100100000,3 1000000000010010,1 0010000001001000,2 1000000100100000,2 1001001000000000,4 1000100100000000,4 0010000001001000,3 1000010010000000,4 0100000100100000,2 1000000100100000,3 1000000000010010,1 0100010010000000,4 1000000100100000,2 1001001000000000,4 1000100100000000,4 0100000000010010,1 1000010010000000,4

  Table 1 shows the 16-bit output h (i, s) and the next state function g (i) for the source (input) word i expressed as an integer from 0 to 255, and for the specific input i when the encoder is in one of four states. , s). For example, suppose the encoder is initialized in state 1 and the input sequence is '8', '3', '4'. In this case, the response to the input '8' is h (8, 1) = '000000010010000', and the new state is g (8, 1) = 3. Thus, the response to input '3' starts at state '3', where h (3, 3) = '0100000001001000' and the next state is g (3, 3) = 2. If we continue for '4' in this way, it starts at state '2', where h (4, 2) = '0100000010010000', and the next state is '2'.

  The decoder converts the 16-bit codeword into an 8-bit data word. It cannot be converted to just 16-bit words, and it must examine the first and 13th symbols of the incoming codeword. That is as follows.

Thus, the decoding of the new code takes the form of converting (16 + 2) channel bits into 8 bits.

  The encoder defined above can freely adjust 351 source words. To enable the use of a single 26-bit sync word, seven additional words are bundled so that only 344 words remain. By handling only 256 source words, the remaining words can be used to minimize power in the low frequency band. Suppression or DC-control of low frequency components is made possible by the control of the digital sum value (DSV). The 88 remaining words are used like the corresponding channel codes for source words 0, ..., 87. A complete encoder can be represented as a "main table" and a "substitute table." For source words 0, ..., 87 the encoder selects a particular representation from the main table or replacement table that minimizes the absolute value of the DSV.

② Reed-Solomon Product Code (RSPC) technology

Product codes are constructed using two linear block codes arranged in a matrix. The length of the codeword, the length of the input source, and the minimum distance are n, k, and d, respectively, and two linear block codes are (n ₁ , k ₁ , d _{min, 1} ), (n ₂ , k ₂ , d, respectively). _{min, 2} ), the minimum distance of the final product code is expressed as the product of the minimum distances of each codeword, resulting in (n ₁ n ₂ , k ₁ k ₂ , d _{min, 1} d _{min, 2} ) a linear block code is formed.

  The input contained in one block of the Reed-Solomon product code (RSPC) used in a conventional DVD device is 172 bytes x 192 bytes. In the encoding process, 16 bytes of parity are first inserted into the outer code in each of 172 columns to form an RS (208, 192, 17) code, and then 10 bytes in every 208 lines in the inner code. Parity is inserted to form RS (182, 172, 11) codes. Thus, the total size of one block is 182 bytes × 208 bytes.

The present invention is intended to devise a method and apparatus that can be applied to an actual commercial optical disk device without physical changes by improving the correction ability for errors generated in the optical disk channel than the conventional one and not bringing about changes in the conventional standard specifications. .

① Description of FIG.

  1 is a view of the structure of the Reed-Solomon product code (RSPC) used in the optical data storage device and the error control method devised. Data encoded by RSPC is stored on an optical disc in units of rows after modulation encoding, and the conventional method in decoding thereof is as follows. First, an inner code is decoded in a row direction with respect to data received by a modulation decoder. At this time, in the case of the DVD device, an error of up to 5 bytes can be corrected with 10 bytes of parity for each line of the inner code. If the error occurs more than 5 bytes, the error of the corresponding line cannot be corrected, and the erasure declaration process is performed so that the position of the error can be known when decoding from the outer code. Perform. After performing an error correction process for each row, if there is an uncorrected row, the outer code is decoded in the column direction. In this case, after the error correction in the inner code, the position information of the error is already known through the erasure declaration, so all 16 bytes of parity can be used for error correction by performing the erasure decoding process. Up to 16 bytes of error correction are possible in each column.

  In the present invention, a method of detecting an error in a modulation decoding process has been devised to improve error correction capability compared to a conventional decoding process so that error correction can be performed even when more errors are generated. Therefore, the error detected during the modulation decoding process is designed to be used as the location information of the error in the inner code, thereby improving the error correction capability of the inner code from 5 bytes up to 10 bytes in each row.

② Description of FIG.

  2 shows the characteristics of an EFMPlus modulation code in a conventional DVD apparatus. The output symbols of the modulation decoder are classified into three cases: (1) correct symbols, (2) undetected error symbols, and (3) detectable error symbols. The encoding and decoding process for the EFMPlus code is performed using a look-up table with a condition of (d = 2, k = 10) and is not present in the look-up table because error propagation is two. If a non-wordword is input to the modulation decoder, one or two errors can be detected. Therefore, as a result of examining the EFMPlus code table, the maximum number of different codewords for each state is 254. On the other hand, the total number of available codewords having a length of 16 bits and satisfying the condition of d = 2 is 566 in total. Thus, when an error occurs in the disc, the minimum probability of detecting an error for the current codeword is (566-254) /566=0.55. Similarly, since the maximum number of codewords duplicated for each state is 101, the probability that an error occurs until the next codeword is detected by a current codeword error is 0.26% or more.

③ Explanation of 3rd and 4th

  3 to 4 show a common code capable of error detection that does not exist in the EFMPlus code for the DVD device. Of the total of 566 16-bit codewords that satisfy the condition of d = 2, 52 codes shown in FIG. 3 do not exist in any state. That is, such a code and a code that does not satisfy d = 2 are capable of error detection regardless of the state position to be decoded.

④ Explanation of Figs. 5 to 12

  5 to 12 illustrate codes capable of detecting an error that does not exist in each state in the EFMPlus decoding process for a DVD device. In the case of state 1, there are additionally 294 codes other than 52 common codes among codes satisfying the condition d = 2. For states 2, 3, and 4, there are no 288, 299, and 287 codes, respectively. Therefore, when a codeword input in the modulation decoding process does not exist in the current state to be decoded, that is, when a code corresponding to a specific state exists in FIG. 4, an error is detected.

⑤ Explanation of FIGS. 13 to 16

  13 to 16 illustrate a code capable of error detection in an ETM decoding process for a next generation HD-DVD device. In the case of ETM codes, error detection by the current codeword itself is not possible due to the decoding characteristics. However, it is possible to detect an error depending on what codeword is input. For example, if the current codeword is "010001010101" and the subsequent input codeword is "000100100101", the codeword cannot be decoded to a correct value. That is, the error can be detected.

⑥ Explanation of FIG. 17

  17 illustrates an error correction method using output information of a modulation decoder in an optical storage device. This figure is consistently applicable to both conventional DVD devices and next generation HD-DVD devices. The error correction process devised in the present invention is as follows.

(1) First, the number of errors (erasures) is calculated for each row by using error detection information input from the modulation decoder. If the number of errors exceeds 10 bytes, the row exceeds the error correction capability, and an erasure declaration is performed on the row so that the outer code can perform erasure decoding.

(2) If the number of erasers is 10 bytes or less, syndrome calculation and inspection are performed. If all syndrome values are 0, it means that no error occurs in the corresponding row, and therefore, an additional error correction process for the corresponding row is not performed. On the other hand, if the syndrome value is not zero, the eraser position polynomial and the error position polynomial are calculated.

(3) If (error count × 2 + number of erasers) exceeds 10 bytes through the erasure position polynomial and the error position polynomial, the error correction capability is exceeded, and the erasure decoding can be performed in the outer code. Perform an erasure declaration on that line.

(4) If the error correction capability is within the range, find the location of the error and then perform the error correction process together with the eraser.

(5) After an error correction process is performed for every row, if an error is not corrected and there are rows in which an erasure is declared, erasure decoding is performed for each column of the outer code.

⑦ Example: Description of FIG. 18

18 is a diagram comparing error correction performance of a signal received in an optical disc channel by inserting an arbitrary error. By default, burst errors generated up to 16 rows in length can be corrected by erasure decoding in the outer code, so the maximum correctable error length in conventional DVD devices is 2922 bytes (= 182 bytes). X 16 rows + 5 bytes x 2 rows). On the other hand, when the error information detected by the EFMPlus decoder designed in the present invention is used, a congestion error of up to 2932 bytes can be corrected. In addition, the maximum number of errors that can be corrected per block of RSPC is 3872 bytes in the case of the conventional DVD device, whereas up to 4832 bytes can be corrected by applying the proposed technique, resulting in 25% performance. It brings an improvement. As shown in the Figure 8, when inserting a random error Optionally, ^10-5 show the performance improvement of 1.5dB in the symbol error rate (symbol error rate, SER). In case of inserting 1 byte of concatenation error, the performance is improved by 2.8dB. However, as the length of the concatenation error increased, the error correction capability was further improved, and when the concatenation errors of 5 bytes and 10 bytes were inserted, the performance improvement was 5.8 dB and 10.5 dB, respectively.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to coding techniques for conventional and next generation DVD devices, and more particularly, to an error control decoder in data reproduction using error characteristics in the decoding process of modulation codes used in conventional DVD devices. A method and apparatus have been devised to improve the error correction capability. Through this invention, data can be reliably decoded even in a situation where consecutive errors occurring due to high-speed reproduction of various noises and data are frequent, thereby enabling a DVD device to reproduce data at a higher speed.

Claims (5)

An error detection capability of an error control decoder in data reproduction is improved by using a characteristic capable of detecting an error in decoding a modulation code. The method of claim 1, wherein in the states 1, 2, 3, and 4 of the EFMPlus modulation code for the DVD device, 250, 254, 243, and 254 different codewords exist, respectively, and 94 codewords are duplicated. There are 90, 101, and 90. In this case, when one of the codewords shown in FIGS. 3 to 12 is input for each state in the decoding process for the input codeword in the corresponding state, it is recognized as an incorrect codeword and transmits error information. The modulation code used in the next generation HD-DVD apparatus is an ETM (eight-to-twelve modulation) code, and when the codeword is input in the form shown in FIGS. In this case, the current codeword is recognized as an invalid codeword and the error information is transmitted. The error information transmitted through claim 2 is used as error location information for the inner code of the RSPC decoder used in the conventional DVD device, and the error location information inputted to the inner code of the RSPC decoder and the EFMPlus decoder output are used. Perform laser decoding. The error information transmitted through Clause 3 is used as error location information for the inner code of the extended RSPC decoder used in the next generation HD-DVD device, and the error location information and the ETM decoder output input to the inner code of the RSPC decoder are used. Erasure decoding is performed.

Images