JP3670616B2 - Signal reproduction method and signal reproduction apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an error correction process in which an error correction code is added to an information block of a specific unit, and more particularly to a signal reproduction method and a signal reproduction apparatus that perform a data demodulation correction process using the correlation between data demodulation and error correction process.
[0002]
[Prior art]
In a system in which digital data is recorded and transmitted in units of bytes (8 bits), the data is often processed by forming a Reed-Solomon error correction product code block. That is, (M × N) bytes of data are arranged in a matrix of M rows × N columns, a Po-byte error correction check word is added to the M-byte information portion for each column, and N-byte information for each row A Pi byte error correction check word is added to the part to construct a Reed-Solomon product code block of (M + Po) rows × (N + Pi) columns. By recording and transmitting this Reed-Solomon error correction product code block, random errors and burst errors can be corrected efficiently on the reproduction side and reception side.
[0003]
Such a Reed-Solomon error correction product code block has a total codeword size, that is, a ratio (redundancy rate) of an error correction check word portion (Pi × M + Po × N + Po × Pi) to (M + Po) × (N + Pi). The smaller, the higher the recording and transmission efficiency. On the other hand, the larger the error correction check words Pi and Po, the higher the correction capability for both random errors and burst errors.
[0004]
Here, even if the redundancy rate of the Reed-Solomon error correction product code block is the same, the M and N are smaller, and therefore the Reed-Solomon error correction product code block with smaller Pi and Po has larger M and N, and therefore Pi and Po. It is known that the correction capability is lower than that of a larger Reed-Solomon error correction product code block.
[0005]
It is known that if M and N are increased, Pi and Po can be increased even with the same redundancy rate, and it is known that high correction capability can be obtained, but this is realized if the following constraints are not satisfied. Can not.
[0006]
First, as a codeword length for constructing a Reed-Solomon codeword, there is a constraint that M + Po and N + Pi must be 255 bytes or less when the word length (symbol length) is 8 bits. Pi is the error correction code length of the PI sequence, and Po is the error correction code length of the PO sequence.
[0007]
Based on these conditions, optical disc standards such as DVD-ROM, DVD-RAM, and DVD-R have recently been announced as information recording media. Among these standards, ISO standards for DVD-ROM and DVD-RAM were determined as DIS16448 (80 mm DVD-ROM), DIS16449 (120 mm DVD-ROM), and DIS16825 (DVD-RAM).
[0008]
In this DVD standard, the RSPS (Reed-Solomon Product Code) method is adopted for the error correction coding processing method, and error correction is performed with an error check word having a smaller redundancy rate than the method used in the conventional optical disk system. The ability has been greatly improved.
[0009]
The DVD error correction method is basically the same as described above, but the base problem is how to set the target values of the random error correction capability and the burst error correction capability. These decisions must be made in consideration of the recording method of the recording medium and the occurrence of defects resulting from handling.
[0010]
Regarding the recording / reproducing system, in the optical disk system, the recording density determined from the recording / reproducing beam spot size resulting from the recording wavelength and optical system characteristics has a great factor in determining the error correction system. In particular, in determining the burst error correction capability, the defect length such as scratches caused by handling is required from experience, but the error correction capability is obtained by multiplying the physical defect length by the linear recording density to determine the burst error of information data. Therefore, it is necessary to increase the correction capability together with the improvement in recording density. Regarding the recording density, the reproduction system is described as an example as follows.
[0011]
Assuming that the light source wavelength is λ and the aperture of the objective lens is NA, the radius R of the spot size of the laser beam is expressed by the following equation (1).
[0012]
R = 0.32λ / NA (1)
As shown in the equation (1), in order to reduce the radius of the spot size of the laser light, the wavelength λ may be shortened or the objective lens NA may be increased.
[0013]
Now, in DVD, the adopted wavelength is 650 nm and NA is 0.6. As an error correction method, a Reed-Solomon product code and (M × N) = (192 × 172) bytes of an information data block are PI = 10 bytes and PO = 16 bytes, respectively.
Row side inner code RS (182, 172, 11)
Outer column code RS (208, 192, 17)
Is adopted. Here, if an “erasure correction” method is used that performs error correction in the PI series, attaches an error mark flag to an uncorrectable row, treats the error mark as an error position in the PO series, and calculates and extracts only the error pattern, a maximum of 16 Can correct row burst errors.
[0014]
In DVD, the recording density is data bit length = 0.267 nm.
0.000267 × 8 × 182 × 16 = 6.2 mm
It can be said that it has a burst error correction capability of about 6 mm.
[0015]
However, studies on large-capacity optical disks with higher density as next-generation DVDs have begun. In order to increase the capacity of current DVDs, the recording density must be increased. In order to meet these recent requirements, in order to improve the recording density, in order to reduce the laser spot size, it is necessary to shorten the wavelength λ or increase the aperture ratio NA as shown in the equation (1). I need it. Recently, studies are being conducted by companies to adopt a wavelength λ = 405 nm and an aperture ratio NA = 0.85. If such a method is used, it becomes possible to increase the density five times or more of the current DVD, and high-definition video such as Hi-vision can be recorded on one disc for two hours or more.
[0016]
However, when the aperture ratio NA is increased, the aberration increases and it becomes difficult to record and reproduce signals. As one of the solutions, a method of reducing the aberration by thinning the disk base is known.
[0017]
The problem of aberrations has been addressed by changing the thickness of the 1.2 mm substrate used for compact discs (CDs) to 0.6 mm. In the next generation, a thin cover method of about 0.1 mm is being studied.
[0018]
However, if the substrate thickness is reduced, the aberration is reduced, but a problem arises that a data error increases due to small dust or scratches on the disk substrate surface, which has been less problematic in the past.
[0019]
In increasing the density by such a method, if a conventional error correction method is introduced, it is not possible to provide the same defect handling capability as in the prior art. As an example, with the burst error correction capability, it is necessary to improve the correction capability by a factor of 2 to 3 in order to have the same capability as a DVD.
[0020]
Furthermore, as described above, the error correction code length is maximum of 255 bytes as long as a symbol length = 8-bit processing system is used, and the DVD standard is PO series 208 bytes. Is close to the limit and only slightly improved.
[0021]
In order to improve the correction capability, the correction code may be increased, but the data utilization rate deteriorates. Thus, conventionally, improvement of burst error correction capability has been considered using interleave technology.
[0022]
Based on the DVD error correction method, a conventionally considered capacity improvement method will be described.
[0023]
FIG. 9 shows a data sector structure adopted in the DVD. One data sector is 172 bytes × 12 rows. Only the first line is provided with a 4-byte ID, a 2-byte IED, and a 6-byte CPR-MAI, and the remaining 160 bytes are main data. From the second line, all 172 bytes are the main data. In the last line, 4-byte EDC is provided after 168-byte main data.
[0024]
As shown in FIG. 10, 16 data sectors are assembled to form one error correction (ECC) data block. One error correction data block includes a data block of 172 bytes × 192 rows. For each column of 192 rows, 16 error correction outer codes PO are generated, and then for each of 208 rows including the outer code PO rows, a 10-byte error correction inner code PI is generated, An error correction block of (172 + 10) bytes × (192 + 16) rows is constructed.
[0025]
Here, the error correction block (outer code PO) of 16 rows is actually subjected to row interleaving processing by being distributed in each sector constituted by 12 rows as shown in FIG. ) 16 sets of recording sectors 0 to 15 composed of +1 row (outer code PO) are configured. FIG. 11 shows an error correction data block after the PO code row has been interleaved.
[0026]
Although there is no difference in the random error correction capability in improving the correction capability in the product code block used in the DVD system, the burst error correction capability can be improved by introducing interleave processing.
[0027]
FIG. 12 shows one method for improving the burst error correction capability that has been conventionally considered. That is, the rows are replaced by interleaving between the ECC blocks A and B shown in FIG. 11 to which the correction code is added. In FIG. 12, a new recording ECC block A ′ is composed of even rows of ECC block A and odd rows of ECC block B, and a new recording ECC block B ′ is composed of odd rows of ECC block A and even rows of ECC block B. Constitute. In this way, the error correction codes viewed from the correction process are distributed at twice the distance, and as a result, have the ability to correct even double burst errors.
[0028]
As a method that does not use the product code as described above, a method using LDC (Long-Distance error correction) Code has been recently proposed. The LDC code method used in hard disk drives and MO disk drives generates and adds error correction codes only in the column direction, and does not use correction codes in the row direction, so that the column direction correction code can be increased.
[0029]
However, the conventional LDC code correction method cannot use the erasure correction technique using the error mark flag used in the product code. In other words, in the product code method, a row that cannot be corrected by the error correction processing in the row direction is marked with an error mark, and the error position information is substituted with the error mark in the error correction processing in the column direction. Operation extraction in processing can be performed only for error patterns, and as a result, correction ability can be improved.
[0030]
In recent years, technology development has been made to use the erasure correction technology used in such product codes also in the LDC code system. An example is described in “Error Modeling and Performance Analysis of Error-Correcting Codes for the Digital Video Recording System”, Kouhei Yamamoto et al., SPIE Journal No. 3864, pages 339 to 341.
[0031]
This document proposes an ECC block structure aiming at improving the capability against short burst errors of an optical disc system which is an open media and a removable media. As shown in FIG. 13, this block structure is an LDC code system in which no error correction code is provided in the data stream (data flow) direction, that is, the row direction, and an error correction code is added only in the column direction. Instead of providing an error correction code in the row direction, a burst error detection code (BEC) symbol for detecting a short burst error consisting of a control code and an error correction code is inserted in a certain unit (n bytes) in the row direction. If the BEC symbol is an error, the surrounding data symbol is regarded as an error, and the correction capability is improved by providing a function such as an error mark flag used for erasure correction with a product code.
[0032]
For 8n bytes × M rows of information data, a P-byte error correction code is generated and added in the row direction. For example, a Reed-Solomon code RS [M + P, M, P + 1] is used as a correction method for each column. The correction block of (8n) bytes × (M + P) rows with the correction code added is divided into 8 equal parts in the row direction, and a total of 6 on the other block boundaries except between the first and fourth blocks. BEC symbols are inserted. As shown in FIG. 14, the BEC symbol is a signal composed of code lengths of control information (S bytes) and a parity code (Q bytes (S + Q = k−1)).
[0033]
In addition to the BEC symbol, a synchronization signal (SYNC) is further inserted into the first, fourth and fifth blocks, and the reliability of the data sandwiched there is predicted from the detection result of the BEC symbol and the SYNC symbol. , LDC error correction processing is performed.
[0034]
When such a technique is used, there is a possibility that a further improvement in performance may be obtained by detecting an error situation in an optical disc having many short burst errors.
[0035]
[Problems to be solved by the invention]
Burst errors such as optical discs occur due to defects such as dust and scratches, but in the reproduction process when reading data, the signal read from the disc is generated in the “read channel” section and the clock generation circuit that reads it. It is supplied to a circuit that reads data with a read clock. The current recording / reproducing method is a self-clocking method. Once the read clock generation circuit in the read channel part goes out of channel synchronization, it is re-synchronized, and even if correct channel data is read out of sync → synchronization The bit shift of about 1 to several channel bits occurs in the conversion process. When a bit shift occurs, the symbol division point when 1 symbol = n channel bits changes, and the demodulator does not demodulate (synchronize) a correct data symbol. A data error caused by such a synchronization shift prevents an error chain in the next synchronization signal. However, in the efficiency improvement by the data structure, the distance between the synchronization signal becomes longer and the error increases. The above bit shift is a channel bit shift of the symbol division point due to increase / decrease of the channel bit clock generated by synchronization of the read clock → synchronization, etc., but such bit shift is scratched on an open media optical disc. In the case of a code length with a long sync signal period, which is generated frequently due to the data structure efficiency, there is a problem that error propagation becomes long.
[0036]
The bit shift indicates that the symbol division point shifts when the channel bit data read clock due to the defect increases or decreases during the defect period. There is also a phenomenon in which the area of the portion shifts back and forth, and such a phenomenon is also called bit shift in the present invention.
[0037]
As described above, the correction method based on the conventional LDC code method using the erasure correction technique used in the product code may not be demodulated into a correct data symbol (synchronization shift) due to bit shift.
[0038]
The present invention relates to data recording / reproduction on a recording medium having an improved recording density, and an object of the present invention is to prevent bit stream shifts caused by defects and the like in order to improve error correction capability against defects and the like.
[0039]
Another object of the present invention is to perform error propagation caused by a bit shift phenomenon by performing a confirmation process of a short burst error detection code in recording and reproduction using a predetermined symbol size and a large error correction block. It is to reduce.
[0040]
[Means for Solving the Problems]
In order to solve the above problems and achieve the object, the present invention uses the following means.
[0041]
  (1) In the signal reproduction method and apparatus of the present invention, an error correction code of P symbols is added in the column direction to an information data block of (M × N) data symbols composed of M rows × N columns, An error correction information block of ((M + P) × N) symbols is constructed, each row is composed of K frames, each frame is divided into L, burst error detection symbols are inserted at the division locations, and burst error detection symbols are An error correction code sequence is formed with (M + P) / J symbols in the direction, and a synchronization signal is added to the head of each frame.And symbol dataRecording media that are sequentially modulated and recorded (however, J,K, M, N is an integer) signal reproducing method and apparatus,
  Detecting the sync signal, demodulating the modulated signal by dividing it into symbols at regular intervals starting from the sync signal detection point, and generating data symbols
  Detect the distance between sync signals,
  If the distance between sync signals deviates from the specified distance, change the demodulation symbol division point.The original modulated signal is divided into symbolsData symbols are re-demodulated to generate data symbols, and data symbol error correction processing is performed.
[0042]
According to this method and apparatus, when a signal modulated by a specific method is transmitted or read from a recording medium and reproduced, the division point when demodulating channel bits into symbol data in the detection state of the synchronization signal is changed. Thus, if the channel bit data is read correctly, the symbol data that has been an error can be read with the correct symbol data.
[0043]
  (2) According to another signal reproduction method and apparatus of the present invention, an error correction code of P symbol is added in the column direction to an information data block of (M × N) data symbols composed of M rows × N columns. An error correction information block of ((M + P) × N) symbols is constructed, each row is composed of K frames, each frame is divided into L, burst error detection symbols are inserted at the division locations, and burst error detection symbols are An error correction code sequence is formed with (M + P) / J symbols in the column direction, and a synchronization signal is added to the head of each frame.And symbol dataRecording media that are sequentially modulated and recorded (however, J,K, M, N is an integer) reproducing method and apparatus,
  Detecting the sync signal, demodulating the modulated signal by dividing it into symbols at regular intervals starting from the sync signal detection point, generating data samples,
  Detect the distance between sync signals,
  When the distance between the synchronization signals is deviated from the predetermined distance, the presence or absence of an error in the burst error detection symbol inserted in the previous frame of the synchronization signal is detected,
  If it is detected that the burst error detection symbol is incorrect,The demodulated data between the burst error detection symbol detected as incorrect and the synchronization signal is modulated back to the original modulation signal, and the original modulation signalData symbol division points are changed and re-demodulated to generate data symbols, and error correction processing of the data symbols is performed.
[0044]
According to the other method and apparatus of the present invention, the range of occurrence of the shift phenomenon of the channel bit clock can be specified by taking the bit shift countermeasure together with the correction processing result of the short burst detection code.
[0045]
  (3) According to another signal reproduction method and apparatus of the present invention, an error correction code of P symbol is added in the column direction to an information data block of (M × N) data symbols composed of M rows × N columns. An error correction information block of ((M + P) × N) symbols is constructed, each row is composed of K frames, each frame is divided into L, burst error detection symbols are inserted at the division locations, and burst error detection symbols are An error correction code sequence is formed with (M + P) / J symbols in the column direction, and a synchronization signal is added to the head of each frame.And symbol dataRecording media that are sequentially modulated and recorded (however, J,K, M, N is an integer) signal reproducing method and apparatus,
  Detecting the sync signal, demodulating the modulated signal by dividing it into symbols at regular intervals starting from the sync signal detection point,
  Execute error correction processing for burst error detection symbols inserted in each frame,
  If it is detected that the burst error detection symbol is incorrect,Remodulate the demodulated data back to the original modulation signal, shift the original burst error detection symbol division point back and forth to the new symbol division point, demodulate the modulation data by the new symbol division point,
  When the error-corrected data of the burst error detection symbol is equal to the burst error detection symbol demodulated at the new symbol division point, the modulation signal is demodulated starting from the new symbol division point.
[0046]
According to another method and apparatus of the present invention, a bit shift is performed by introducing a processing step of comparing a short burst error detection code obtained by changing an error correction result of a short burst error detection code and a symbol division point. The reliability of whether or not a phenomenon has occurred can be greatly improved.
[0047]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of a signal reproduction method and apparatus according to the present invention will be described with reference to the drawings.
[0048]
First embodiment
FIG. 1 is a diagram showing a basic configuration of a DVD disk drive as a signal reproducing apparatus according to the first embodiment of the present invention. A process of reading data written on the disk will be described with reference to FIG.
[0049]
A signal read by the optical pickup head 12 from the optical disk 10 that is a recording medium is amplified by the preamplifier 14 and subjected to waveform equalization processing and the like, and is sent to the read channel unit 16. Here, the frequency component and phase component of the sent signal are extracted, and a channel bit clock (PLCK) for reading channel bit data is generated by a built-in PLL (Phase Locked Loop) circuit. A synchronization signal is detected by this clock PLCK, and channel bit data is read out from the subsequently transmitted signal. The read channel bit data is demodulated into symbol data by the demodulator 18, sector synchronization and error correction block synchronization are taken, and once recorded at a designated location in the buffer memory 20.
[0050]
When the data stored in the buffer memory 20 is stored in units of error correction blocks, the error correction unit 22 reads the sequentially stored data and executes error correction processing. In this process, the data in the buffer memory 20 is subjected to error correction processing, and sent to the outside through the interface (I / F) unit 24 in response to an external data request instruction. This is the process of the drive playback process.
[0051]
A control microcomputer 30 is connected to a servo circuit 28 for controlling the motor 26 of the optical disk 10, the error correction unit 22, and the I / F unit 24 via a microcomputer bus 32.
[0052]
Now, in such processing steps, a bit shift compensation technique based on correction processing with BEC symbols added will be described in order to improve the correction capability by adopting an LDC code without using a product code as an error correction method.
[0053]
FIG. 2 shows an ECC block in which a BEC symbol is inserted in this embodiment. In FIG. 2, 6 columns of BEC are inserted in one row, and 2 sets of BEC columns (for example, the 1st BEC sequence, for example, the first BEC sequence is 2 sets of BEC sequences of “B00 to B0k” and “B60 to B6k”). This is an example in which BEC symbols are arranged in every other line. Actually, it is not limited to two sets, and a plurality of sets of BEC symbols may be arranged in order. Alternatively, since 4 sets are arranged in a row and there are 6 BEC symbol sequences, 24 sets of BEC symbols may be inserted in total. In FIG. 2, there are 12 BEC sequences in total.
[0054]
In the main information data, a P-byte parity is generated and added to M-byte information. However, if the BEC symbol has a code length of (M + P) / J and the BEC correction code is a P symbol, the main information data Since it is the same as the parity number, the BEC data length is ((M + P) / J) -P, and since the ratio of the parity length to the data length is large, the error correction capability is strong for the main information data. Yes. When the BEC symbol is added, the information and the BEC symbol are modulated next time transmission or recording is performed, but the synchronization signal SYNC is added between the head and the 4th and 5th blocks of the above division. Is done. A configuration including the addition of the synchronization signal SYNC is shown in FIG.
[0055]
In such a configuration, information data and BEC symbols are modulated and transmitted or recorded. However, when a recorded or transmitted signal is reproduced, there are signal defects caused by scratches or other defects. This signal defect not only destroys data but also disturbs a PLL circuit for generating a read clock for reading data. Even when the PLL bit is out of sync and the channel bit is out of synchronization, and resynchronization is performed with the correct signal, a clock increase or decrease of about 1 to 2 clocks often occurs. Such an operation is an operation in which the distance between the synchronization signals is increased or decreased from the viewpoint of the signal processing system, resulting in erroneous channel bit division points for demodulating channel bit data into symbol data. When such a bit shift phenomenon occurs, all the symbol data becomes an error until the channel bit synchronization is resynchronized with the next synchronization signal.
[0056]
FIG. 3 shows a relationship in which a bit shift occurs due to a signal defect. Here, a defect occurs between the synchronization signals (SYNC), readout is out of synchronization, and bit shift occurs.
[0057]
FIG. 4 is a timing chart showing the relationship between signals generated by bit shift. Generally, the sync signal detection circuit in the playback device is considered to generate a false sync signal pattern due to scratches in the information area, so that the sync signal detection window circuit is operated by the previously detected sync signal, Actually, the channel bit clock is counted, a detection window is generated with a width before and after the timing at which the next synchronization signal is generated, and when it is detected, it is recognized as a correct synchronization signal.
[0058]
In FIG. 4, the sync signal detected with respect to the reproduction signal is “SYNC detection signal”, the signal to be generated at the just point of the specified SYNC distance from the previous sync signal is “SYNC detection TP”, and the SYNC detection window. Is displayed as “SYNC detection WD”.
[0059]
In the case of the “2cb shift example” in which the PLL is output in the defective area due to a shortage of two clocks due to the inter-SYNC defect as shown in FIG. 3, the “SYNC distance TP” is It occurs at a point delayed by 2 channel bits.
[0060]
When the last change point of the sync pattern is read out, the “SYNC detection signal” is generated, the distance is measured from the “SYNC detection signal” one frame before, and the “SYNC distance TP” signal is generated. If the rising edge of the “SYNC detection signal” occurs in the TP signal, it means that the SYNC has occurred at the correct distance. In the case of the 2CB shift in FIG. 4, the edge of the “SYNC detection signal” occurs 2 CB before the TP signal, and it is found that the CB shift is 2 CB forward. Actually, when shifting to the front side, the distance from the edge of the SYNC detection signal to TP is measured to determine the shift amount. When shifting to the rear side, the distance from TP to SYNC is measured. To detect the shift amount. As a result, the “SYNC detection signal” detected from the reproduction signal is detected two channel bits before the “SYNC distance TP”.
[0061]
In the reproduction operation, the information data before the synchronization signal detected here is symbol-divided on the basis of “distance TP between SYNCs” (actually, the previously detected synchronization signal is a reference) and demodulated by the demodulator. If a bit shift occurs in the channel, all the symbol data becomes an error even if the channel bit data is correctly read.
[0062]
If such a bit shift phenomenon can be detected from where it starts and how much it has been shifted, the symbol division point can be changed to prevent a shift error. Since the shift phenomenon does not always occur due to defects, it has been difficult to detect the shift phenomenon. In this embodiment, detection of shift amount is confirmed using a BEC symbol signal as shown in FIG. 2 to compensate for a short burst error.
[0063]
FIG. 5 is a block diagram of a first embodiment of a signal reproducing apparatus in which an error correction method with BEC symbols added is introduced.
[0064]
A signal read from the optical disk 40 which is a recording medium by the optical pickup head 42 is amplified by a preamplifier (not shown), waveform equivalent processing, and the like, and sent to the read channel unit 44. Here, the frequency component and phase component of the sent signal are extracted, and a channel bit clock (PLCK) for reading channel bit data is generated by a synchronization detection separation circuit 46 such as a PLL (Phase Locked Loop) circuit. A synchronization signal is detected by this clock PLCK, and channel bit data is read out from the subsequently transmitted signal. When the synchronization signal is detected from the read channel bit data and the symbol division point is determined, the demodulator 48 demodulates the symbol data, and sector synchronization and error correction block synchronization are obtained. Once recorded. At this time, the synchronization signal detection status is also stored in the data storage unit 50.
[0065]
When all data in the error correction block is stored in the data storage unit 50, the BEC correction / detection unit 52 first performs error correction processing of the BEC symbol. The error flag control unit 54 generates a flag indicating the reliability of the information data from the synchronization signal detection status and the BEC symbol correction processing result. In general, when neither the synchronization signal nor the BEC symbol code is correctly detected (there is no reliability of the information data), it is determined that the synchronization signal has a channel bit shift and the information data is incorrect, The re-modulation / re-division unit 62 re-modulates and re-demodulates the periphery of the target BEC symbol code to check the shift state, and re-modulates the information data between the synchronization signal and the BEC symbol code according to the result, and divides the symbol. Change the point. The re-demodulation unit 64 changes the data in the data storage unit 50 according to the changed symbol division point, and the error correction decoding unit 56 processes the data. If the error-corrected signal is scrambled data, it is descrambled by the descramble unit 58 and output through the data output control circuit 60. If the information data is reliable, it is processed by the error correction decoding unit 56 as it is.
[0066]
Next, the operation of the playback apparatus in FIG. 5 will be described.
[0067]
FIG. 6 shows the data relationship of the processing steps of this embodiment. The modulation scheme in FIG. 6 is an example shown for the case where the RLL (17) code is used.
[0068]
Information symbol data (hexadecimal) to be recorded is connected in a bit arrangement of symbol data and converted into channel bits according to the rules of the RLL (17) code. The channel bit stream is recorded on a recording medium by a recording driver by NRZI modulation. NRZI is a modulation method in which polarity inversion is performed at the position of 1 in the channel bit signal. As a result, the continuous distance of “1” or “0” of the recording signal is limited to a minimum of 2 channel bits and a maximum of 8 channel bits. When reproducing the signal recorded on the recording medium in this way, if the 2-channel bit shift phenomenon in FIG. 3 occurs and the symbol division point is reproduced with 2-channel bit shift, the reproduction data of FIG. It becomes a relationship.
[0069]
In the present embodiment, when it is detected by the synchronization signal detection that the channel is shifted by 2 channels, the information is stored in the data storage unit 50, and the symbol division point is corrected and demodulated.
[0070]
FIG. 7 is a flowchart of the reproduction process. First, a channel bit clock (PLCK) for reading channel bit data is generated by the synchronization detection separation circuit 46 from the signal read from the optical disk 40 by the optical pickup head 42 in step S12, and the synchronization signal is detected by this clock PLCK. The At this time, a bit shift is also detected. In step S14, a symbol division point is determined based on the synchronization signal, and demodulated by the demodulator 48 into symbol data. At the time of demodulation, sector synchronization and error correction block synchronization are also taken and once recorded at a designated location in the data storage unit 50.
[0071]
In step S16, the error flag generation unit 54 determines whether or not there is a bit shift from the synchronization signal detection status. If there is a bit shift, a shift flag indicating the reliability of the information data is set in step S18, and the bit shift amount Record.
[0072]
After all the error correction block data is stored in the data storage unit 50, the BEC symbol error correction processing is performed by the BEC correction / detection unit 52 in step S20. In the example of FIG. 6, it is assumed that the BEC symbol “93” is corrected to “58” by error correction. In general, the information between the synchronization signal and the BEC symbol and the BEC symbol code is determined to be erroneous if the synchronization signal and the BEC symbol code are not correctly detected, and error correction decoding is performed. Processed by the unit 56. In step S22, a flag indicating that the BEC symbol is corrected is set, and the corrected BEC symbol “58” is recorded.
[0073]
In order to perform error correction processing of the information data, the information data is read from the data storage unit 50, and it is checked in step S24 whether the BEC symbol preceding the target information data is corrected. If the target BEC symbol is corrected, there is a high possibility that the surrounding symbol data is not correct. Therefore, in step S26, the demodulated symbol data array is remodulated to obtain channel bits. In step S28, it is determined whether or not the corrected BEC symbol “58” is included in the remodulated channel bits. Specifically, the symbol division point is shifted by the bit shift amount recorded in step S18 to determine whether there is a pattern that matches the symbol “58”. If there is no matching pattern, the symbol division point is further shifted. If the corrected BEC symbol “58” is included in the remodulated channel bits, in step S30, the channel bits are subdivided based on the symbol division point corresponding to the symbol “58” and re-demodulated. To do. As a result of this re-demodulation, the same symbols as the recorded symbols are demodulated.
[0074]
As described above, according to this embodiment, even if channel bit loss of synchronization (bit shift phenomenon) occurs and the channel bit division point for demodulating channel bit data into symbol data is incorrect, bit shift is performed. By detecting where the phenomenon starts and how much it has shifted, it is possible to compensate for the short burst error by changing the symbol division point and performing maximum demodulation. Therefore, an error correction code is generated and added only in the column direction, and the correction code in the row direction is not used, so that the column direction correction code can be increased (the erasure correction technique using the error mark flag used in the product code). Can be compensated for short burst errors. Recently, in order to increase the efficiency of the data structure, the synchronization signal distance has been increased, and the shift phenomenon of the channel bit clock due to defects or the like has become a big problem, so this embodiment has a great effect.
[0075]
Further, when the burst error detection symbol is incorrect, the data obtained by performing error correction processing on the burst error detection symbol and the data re-demodulated by changing the division point of the channel bit string generated by remodulating the intra-frame data symbol are described above. By setting a division point equal to the error-corrected burst error detection symbol data as a subsequent data symbol division point, it is possible to correctly detect the shift amount between the short burst error codes.
[0076]
In addition, burst error detection symbols are inserted at a plurality of positions in the frame, and when data symbols are remodulated and re-demodulated by changing the division points, the correct division point deviation amount of each burst error detection symbol is detected. By determining the change division point for data symbol demodulation based on the deviation amount, it is possible to derive the shift amount setting in the detection state of the preceding and following short burst error detection codes.
[0077]
In addition, when a plurality of burst error detection symbols are inserted in a frame and the detection distance of the synchronization signal is shifted, an error of the burst error detection symbol closest to the synchronization signal in the previous frame of the synchronization signal is first detected. By examining the detection and correction result, then examining the error detection and correction result of the previous burst error detection symbol, examining the detection and correction result sequentially up to the synchronization signal before the synchronization signal, and adjusting the re-dividing point of the data symbol When the detection point of the synchronization signal is shifted, the short burst error detection process is performed by tracing back to the previous signal from the side close to the synchronization signal, so that detection can be performed even when there are a plurality of shifts.
[0078]
Also, when a deviation is detected in the detection timing of the synchronization signal, depending on the state of the burst error detection symbol, the data symbol is returned to the channel bit signal, the symbol division point is changed, and re-demodulation is performed. Even when the symbol division point is changed by setting a flag as an error and performing error correction processing of the data symbol, it is possible to prevent erroneous correction by adding a flag indicating that the information data is in error.
[0079]
Hereinafter, other embodiments of the signal reproduction method and apparatus according to the present invention will be described. In the description of the other embodiments, the same parts as those of the first embodiment are denoted by the same reference numerals, and the detailed description thereof is omitted.
[0080]
Second embodiment
FIG. 8 is a block diagram of the signal reproducing apparatus of the second embodiment. In the first embodiment, as shown in FIG. 5, in the reproduction processing step, channel bit data is read, symbol division points are set by detecting a synchronization signal, and symbol data is demodulated and demodulated by the demodulation circuit 48. After the symbol data is stored in the data storage unit 50 and the error correction block data is stored, it is read from the data storage unit 50 and subjected to error correction processing. This process includes re-modulation / re-demodulation, but the re-modulation process is deleted in the second embodiment. That is, the channel bit data is stored in the data storage unit 50 without being demodulated, and when performing error correction processing, the channel bit data is read from the data storage unit 50 and demodulated into symbol data on the spot for correction processing. When bit shift is detected, the symbol division point changed by the re-division unit 62 is instructed to the demodulation unit 48. In this configuration, re-modulation is not required, but when the modulation method is the RLL method, the data storage unit 50 is required more than in the first embodiment. The modulation method 2/3 RLL (1, 7) requires 1.5 times the capacity, and the 8/16 RLL (2, 10) used in the DVD standard requires twice the capacity. In this method, since it is necessary to perform error correction after reading and demodulating the channel bits stored in the data storage unit 50, first, in the error correction processing of the BEC code, the modulation method is 2/3 RLL ( In 1 and 7), there is a possibility that the demodulation is connected to the adjacent symbol signal. If channel bit data for three symbols before and after is not read, the BEC code cannot be demodulated. However, it is to provide a technique for compensating the read data from the bit shift phenomenon by the channel bit clock from the detection state of the synchronization signal and the error correction state of the BEC code, and the processing and demodulation of the data storage unit 50 There is no difference in the processing of the unit 48.
[0081]
As described above, according to the present invention, by using the LDC method without using a product code, a large error correction block with a predetermined symbol size can be obtained without increasing the redundancy rate of the parity amount with respect to the entire data amount. In the data reproduction from the high-density recording medium to be employed, the error propagation caused by the bit shift phenomenon can be reduced by performing the confirmation process of the short burst error detection code.
[0082]
【The invention's effect】
As described above, according to the present invention, a short burst error detection code is related to data recording / reproduction on a recording medium with improved recording density, in recording / reproduction having a predetermined symbol size and employing a large error correction block. By performing the confirmation process, it is possible to reduce the error propagation caused by the bit shift phenomenon and to improve the error correction capability for defects and the like.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a DVD drive to which a signal reproducing apparatus according to the present invention is applied.
FIG. 2 is a diagram showing an example of arrangement of BECs in an ECC block according to an embodiment of the present invention.
FIG. 3 is a diagram for explaining a factor causing a bit shift;
FIG. 4 is a diagram showing bit shift detection timing.
FIG. 5 is a block diagram showing a configuration of a first embodiment of a signal reproduction device according to the present invention.
FIG. 6 is a diagram showing a data conversion process according to the first embodiment.
FIG. 7 is a flowchart showing the reproduction operation of the first embodiment.
FIG. 8 is a block diagram showing a configuration of a second embodiment of a signal reproduction device according to the present invention.
FIG. 9 is a diagram showing a configuration example of a data sector in a conventional DVD.
FIG. 10 is a diagram showing a configuration example of an ECC block in a conventional DVD.
FIG. 11 is a diagram showing a configuration example of an ECC block after row interleaving in a conventional DVD.
FIG. 12 is a diagram showing an example of interleaving between different ECC blocks.
FIG. 13 is a diagram showing a configuration example of an ECC block of an LDC code method in which an error correction code is added only in a conventional column direction.
14 is a diagram showing the configuration of the BEC symbol in FIG.
[Explanation of symbols]
40 ... Optical disc
46. Synchronization detection separation unit
48. Demodulator
50: Data storage unit
52 ... BEC correction / detection unit
54 ... Error control section
56: Error correction decoding unit
58 ... Descramble club
60: Data output control section
62 ... Remodulation / re-division unit
64... Re-demodulation unit

Claims (16)

An error correction code of P symbols is added in the column direction to an information data block of (M × N) data symbols composed of M rows × N columns, and an error correction information block of ((M + P) × N) symbols. And each row is composed of K frames, each frame is divided into L, burst error detection symbols are inserted at the division locations, and the burst error detection symbols are error correction code sequences of (M + P) / J symbols in the column direction. A signal of a recording medium (where J, K, M , and N are integers) in which a synchronization signal is added to the head of each frame and the synchronization signal and symbol data of each frame are sequentially modulated and recorded In the playback method,
Detecting a synchronization signal, demodulating the modulation signal by dividing the modulation signal at regular intervals starting from the synchronization signal detection point, and generating data symbols;
Detecting a distance between synchronization signals ;
If the synchronization signal distance is deviated from a predetermined distance, to generate data symbols and re demodulated symbols dividing the original modulated signal by changing the demodulation symbol dividing point, the step of performing error correction processing of the data symbols When,
A signal reproduction method comprising : The error correction processing step includes
  If the distance between the sync signals deviates from the specified distance, remodulate the demodulated data and restore the original modulation signal, then change the demodulation symbol division point and divide the original modulation signal into symbols and re-demodulate 2. The signal reproduction method according to claim 1, wherein data symbols are generated and error correction processing of the data symbols is performed. An error correction code of P symbols is added in the column direction to an information data block of (M × N) data symbols composed of M rows × N columns, and an error correction information block of ((M + P) × N) symbols. And each row is composed of K frames, each frame is divided into L, burst error detection symbols are inserted at the division locations, and the burst error detection symbols are error correction code sequences of (M + P) / J symbols in the column direction. Reproduction of a recording medium (where J, K, M , and N are integers) in which a synchronization signal is added to the head of each frame and the synchronization signal and symbol data of each frame are sequentially modulated and recorded In the method
Detecting the sync signal, demodulating the modulated signal by dividing it into symbols at regular intervals starting from the sync signal detection point, generating data samples,
Detect the distance between sync signals,
When the distance between the synchronization signals is deviated from the predetermined distance, the presence or absence of an error in the burst error detection symbol inserted in the previous frame of the synchronization signal is detected,
If it is detected that the burst error detection symbol is incorrect, the demodulated data between the burst error detection symbol detected as incorrect and the synchronization signal is modulated back to the original modulation signal, and the original modulation signal is divided into data symbols. A signal reproduction method in which data symbols are generated by changing the points and re-demodulated to perform error correction processing on the data symbols. An error correction code of P symbols is added in the column direction to an information data block of (M × N) data symbols composed of M rows × N columns, and an error correction information block of ((M + P) × N) symbols. And each row is composed of K frames, each frame is divided into L, burst error detection symbols are inserted at the division locations, and the burst error detection symbols are error correction code sequences of (M + P) / J symbols in the column direction. A signal is reproduced from a recording medium (where J, K, M , and N are integers) in which a synchronization signal is added to the head of each frame, and the synchronization signal and symbol data of each frame are sequentially modulated and recorded. In the method
Detecting the sync signal, demodulating the modulated signal by dividing it into symbols at regular intervals starting from the sync signal detection point,
Execute error correction processing for burst error detection symbols inserted in each frame,
If it is detected that the burst error detection symbol is incorrect, the demodulated data of that frame is re-modulated and returned to the original modulation signal, and the original burst error detection symbol division point is shifted back and forth to the new symbol division point. And demodulate the modulated data using the new symbol division point,
A signal reproduction method for demodulating a modulated signal starting from a new symbol division point when the error-corrected data of the burst error detection symbol is equal to a burst error detection symbol demodulated by the new symbol division point . If incorrect further burst error detection symbols, burst error detection symbols and error correction processing data, the division point by changing remodulated data is the error of the channel bit string generated by re-modulating the frame data symbols 2. The signal reproduction method according to claim 1, wherein a division point equal to the corrected burst error detection symbol data is set as a division point of the subsequent data symbol. In the reproduction of a recording medium in which the data in which a plurality of burst error detection symbols are inserted in a frame is modulated and recorded, the data symbol that has been symbol-divided and demodulated from the detected synchronization signal is remodulated, When the division point is changed again and re-demodulation is performed, if a correct division point shift amount of each burst error detection symbol is detected, a change division point for data symbol demodulation is determined based on the shift amount. 4. The signal reproduction method according to any one of 3.   When a plurality of burst error detection symbols are inserted in a frame and the synchronization signal detection distance is detected to be shifted, error detection is first performed on the burst error detection symbol closest to the synchronization signal in the previous frame of the synchronization signal. 2. The correction result is checked, then the error detection correction result of the previous burst error detection symbol is checked, the detection correction result is sequentially checked up to the synchronization signal before the synchronization signal, and the subdivision point of the data symbol is adjusted. The signal reproduction method according to claim 3. When a deviation is detected in the synchronization signal detection timing, if the burst error detection symbol is incorrect , the data symbol is returned to the channel bit signal and the symbol division point is changed according to the synchronization signal detection timing deviation. 4. The signal reproduction method according to claim 1, wherein the data symbol that has been re-demodulated and changed is flagged as an error and an error correction process is performed on the data symbol. An error correction code of P symbols is added in the column direction to an information data block of (M × N) data symbols composed of M rows × N columns, and an error correction information block of ((M + P) × N) symbols. And each row is composed of K frames, each frame is divided into L, burst error detection symbols are inserted at the division locations, and the burst error detection symbols are error correction code sequences of (M + P) / J symbols in the column direction. A signal of a recording medium (where J, K, M, and N are integers) in which a synchronization signal is added to the head of each frame and the synchronization signal and symbol data of each frame are sequentially modulated and recorded In the playback device ,
Means for detecting a synchronization signal, demodulating the modulation signal by dividing it into symbols at regular intervals starting from the synchronization signal detection point, and generating data symbols;
Means for detecting the distance between the synchronization signals;
Means for changing data symbol error by changing demodulated symbol division point, dividing original modulation signal into symbols, and re-demodulating to generate data symbols when distance between synchronization signals deviates from predetermined distance When,
A signal reproducing apparatus comprising: The error correction processing means includes
  If the distance between the sync signals deviates from the specified distance, remodulate the demodulated data and restore the original modulation signal, then change the demodulation symbol division point and divide the original modulation signal into symbols and re-demodulate 10. The signal reproducing apparatus according to claim 9, wherein a data symbol is generated and error correction processing of the data symbol is performed. An error correction code of P symbols is added in the column direction to an information data block of (M × N) data symbols composed of M rows × N columns, and an error correction information block of ((M + P) × N) symbols. And each row is composed of K frames, each frame is divided into L, burst error detection symbols are inserted at the division locations, and the burst error detection symbols are error correction code sequences of (M + P) / J symbols in the column direction. Reproduction of a recording medium (where J, K, M , and N are integers) in which a synchronization signal is added to the head of each frame and the synchronization signal and symbol data of each frame are sequentially modulated and recorded In the device
Means for detecting a synchronization signal, demodulating the modulation signal by dividing it into symbols at regular intervals starting from the synchronization signal detection point, and generating data samples;
Means for detecting the distance between the synchronization signals;
Means for detecting the presence or absence of an error in a burst error detection symbol inserted in the previous frame of the synchronization signal when the distance between the synchronization signals is deviated from a predetermined distance;
If it is detected that the burst error detection symbol is incorrect, the demodulated data between the burst error detection symbol detected as incorrect and the synchronization signal is modulated back to the original modulation signal, and the original modulation signal is divided into data symbols. Means for changing points and re-demodulating to generate data symbols, and performing error correction processing of the data symbols;
A signal reproducing apparatus comprising: An error correction code of P symbols is added in the column direction to an information data block of (M × N) data symbols composed of M rows × N columns, and an error correction information block of ((M + P) × N) symbols. And each row is composed of K frames, each frame is divided into L, burst error detection symbols are inserted at the division locations, and the burst error detection symbols are error correction code sequences of (M + P) / J symbols in the column direction. A signal is reproduced from a recording medium (where J, K, M , and N are integers) in which a synchronization signal is added to the head of each frame, and the synchronization signal and symbol data of each frame are sequentially modulated and recorded. In the device
Means for detecting a synchronization signal, demodulating the modulation signal by dividing it into symbols at regular intervals starting from the synchronization signal detection point;
Means for performing error correction processing of a burst error detection symbol inserted in each frame;
If it is detected that the burst error detection symbol is incorrect, the demodulated data of that frame is re-modulated and returned to the original modulation signal, and the original burst error detection symbol division point is shifted back and forth to the new symbol division point. Means for demodulating the modulated data with the new symbol division point;
Means for demodulating a modulated signal starting from a new symbol division point when the error-corrected data of the burst error detection symbol is equal to a burst error detection symbol demodulated by a new symbol division point;
A signal reproducing apparatus comprising: If incorrect further burst error detection symbols, burst error detection symbols and error correction processing data, the division point by changing remodulated data is the error of the channel bit string generated by re-modulating the frame data symbols 10. The signal reproducing apparatus according to claim 9, wherein a division point equal to the corrected burst error detection symbol data is used as a division point of the subsequent data symbol. In the reproduction of a recording medium in which the data in which a plurality of burst error detection symbols are inserted in a frame is modulated and recorded, the data symbol that has been symbol-divided and demodulated from the detected synchronization signal is remodulated, 10. When the division point is changed again and re-demodulation is performed, if a correct division point shift amount of each burst error detection symbol is detected, a change division point for data symbol demodulation is determined based on the shift amount. The signal reproduction device according to any one of 12.   When a plurality of burst error detection symbols are inserted in a frame and the synchronization signal detection distance is detected to be shifted, error detection is first performed on the burst error detection symbol closest to the synchronization signal in the previous frame of the synchronization signal. 10. The correction result is checked, then the error detection correction result of the previous burst error detection symbol is checked, the detection correction result is sequentially checked up to the synchronization signal before the synchronization signal, and the subdivision point of the data symbol is adjusted. The signal reproducing device according to claim 12. When a deviation is detected in the synchronization signal detection timing, if the burst error detection symbol is incorrect , the data symbol is returned to the channel bit signal and the symbol division point is changed according to the synchronization signal detection timing deviation. 13. The signal reproduction device according to claim 9, wherein the data symbol that has been re-demodulated and changed is flagged as an error and an error correction process is performed on the data symbol.

Images