JP5365683B2 - Error correction device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an error correction device capable of reducing an access amount to an external memory even while suppressing an increase in a circuit scale due to an increase in a built-in memory capacity. <P>SOLUTION: A descramble circuit 20 includes an EOR circuit 21 for receiving scramble data sDu from a built-in memory in which an ECC cluster is divided from a demodulator circuit and stored. The descramble circuit 20 includes: a one-shift computing element 24 for shifting an inputted scramble value for one byte according to a generating function &Phi;(x) to generate a new scramble value, and a 209 shift computing element 25 for shifting the inputted scramble value for 209 bytes according to the generating function &Phi;(x) to generate a new scramble value. The descramble circuit 20 includes a selection circuit 28 and a selector 27 for selecting a scramble value to be outputted to the EOR circuit 21 in accordance with a processing order during scramble processing of the scramble data sDu to be inputted to the EOR circuit 21. <P>COPYRIGHT: (C)2012,JPO&amp;INPIT

Description

The present invention relates to an error correction apparatus.
In recent years, with an increase in storage capacity in a recording medium such as an optical disk, it is also required to increase the data reading time. However, it is difficult to correctly read data from the optical disk at high speed due to defects during manufacture of the optical disk or dirt attached to the surface of the optical disk. Therefore, the recorded data of such an optical disc is recorded with an error correction code using a Reed-Solomon code together with the data, and correct data is restored by performing error correction processing based on the error correction code. I am doing so. Since the processing time of this error correction processing is long, in order to realize a high data reading time, it is necessary to increase the speed of the error correction processing.

  In recent years, development of a new high-density optical disc capable of recording high-quality video data and high-quality audio data for a long time, for example, a Blu-ray Disc (BD) (for example, see Patent Document 1). ). The data recorded on the BD includes an error detection code (EDC) for detecting an error that occurs when the data is read and an error for correcting an error that occurs when the data is read. An error correcting code (ECC) is added in advance.

  Here, the format of the data stored in the BD will be described in detail. As shown in FIG. 34, the user data UD is composed of 32 data frames DF0 to DF31 (collectively referred to as data frames DF). One data frame DF includes 2048 bytes (2 kbytes) of User Data, and a 4-byte error detection code EDC calculated by the following equation is added to the end of one data frame DF.

When the data is recorded on the BD, each data frame DF is scrambled, and each data frame DF is arranged in one column as shown in FIG. It is converted into a scrambled data frame SDF consisting of d (u, v). Note that u of the data d (u, v) indicates a row number in the scrambled data frame SDF, and v indicates a column number in the scrambled data frame SDF.

Here, as is well known, the scrambled data is generated by taking the exclusive OR of each byte of the original data before the scramble process and the lower 8 bits of the scramble value SC. For the generation of such a scramble value SC, a scramble value generation circuit 100 shown in FIG. 35 is used. As shown in FIG. 35, the scramble value generation circuit 100 is configured according to a predetermined generation polynomial Φ (x) (in the example of FIG. 35, Φ (x) = x ¹⁶ + x ¹⁵ + x ¹³ + x ⁴ +1), A feedback shift register 101 composed of bits b15 to b0 and EOR circuits 102, 103, and 104 are provided.

  The EOR circuit 102 takes an exclusive OR of the bits b15 and b14 and outputs the operation result to the EOR circuit 103. The EOR circuit 103 takes an exclusive OR of the operation result from the EOR circuit 102 and the bit b12 and outputs the operation result to the EOR circuit 104. The EOR circuit 104 takes an exclusive OR of the operation result from the EOR circuit 103 and the bit b3, and outputs the operation result to the bit b0.

  The shift register 101 takes in an initial value S0 of a preset scramble value and outputs the lower 8 bits (bits b7 to b0) of the initial value S0 to an EOR circuit to which original data is input. In the EOR circuit, the exclusive OR of the lower 8 bits of the scramble value SC (initial value S0) and the data D0 of the first byte of each data frame (hereinafter referred to as an exclusive OR Scramble processing is performed on the data D0. That is, the scrambled data sD0 after the data D0 is scrambled is generated according to the following equation.

In one user data UD, the same initial value S0 is used for all data frames DF0 to DF31.

Next, every time a pulse of the clock signal CLK is input, the shift register 101 shifts the values of the bits b14 to b0 to the bits b15 to b1, respectively, and takes in the output of the EOR circuit 104 as the bit b0. Then, the scramble value generation circuit 100 outputs the bits b7 to b0 at that time to the EOR circuit to which the original data is input after 8 pulses of the clock signal CLK, that is, after shifting by 8 bits (1 byte). Here, representing the scramble value SC of the time and S0 · γ ^1. Γ is a 1-byte shift operator given by the generator polynomial Φ (x), and S0 · γ ¹ is obtained by shifting the initial value S0 by 1 byte according to the generator polynomial Φ (x). The scramble value SC is indicated. The data D1 is scrambled by calculating an exclusive OR between the lower 8 bits of the scramble value SC (= S0 · γ ¹ ) and the data D1 of the second byte of each data frame DF. That is, the scramble data sD1 for the data D1 is generated according to the following equation.

Thereafter, similarly, scrambled data sD2 to sD2051 are generated according to the following equation.

S0 · γ ³ indicates a scramble value SC obtained by shifting the initial value S0 by 3 bytes according to the generator polynomial Φ (x), and S0 · γ ²⁰⁵¹ is generated from the initial value S0. The scramble value SC obtained when shifted by 2051 bytes according to the polynomial Φ (x) is shown.

  When the scrambled data sD0 to sD2051 of each data frame DF generated in this way is arranged in order from the top in the column direction, the scrambled data frame SDF is generated.

  Next, as shown in FIG. 36, the scrambled data frame SDF (see FIG. 36A) is converted into a data block DB of 216 rows × 304 columns (see FIG. 36B). Specifically, each data d (u, v) in the column direction of one data frame DF of the scrambled data frame SDF is stored in the data block DB in the order of the solid line arrows shown in FIG. That is, data d (0, 0) to data d (215, 0) of the data frame DF0 are stored in order from the top in the first column of the data block DB. The data d (216, 0) to d (431, 0) of the data frame DF0 are stored in order from the top in the second column of the data block DB. Data d (1944, 0) to data d (2051, 0) of the data frame DF0 are sequentially stored from the top to the 10th column in the 10th column of the data block DB. Next, data d (0, 1) to data (107, 1) of the data frame DF1 are the next positions where the last data d (2051, 0) of the data frame DF0 of the data block DB is stored, that is, 10 The 109th to 216th rows in the column are stored in order from the top. The data d (108, 1) to d (323, 1) of the data frame DF1 are sequentially stored from the top in the 11th column of the data block DB. Data d (1836, 1) to data d (2051, 1) of the data frame DF1 are sequentially stored from the top in the 19th column of the data block DB. The data frames DF2 to DF31 are also stored in the data block DB in the same manner as the data frames DF0 and DF1. When the data d (u, v) of each data frame DF is thus stored in the data block DB, even data frames (for example, data frames DF0.DF2, DF4) and odd data frames (for example, data frames DF1, for example) are stored. When combined with DF3 and DF5), a two-dimensionally arranged block of 216 rows × 19 columns is obtained. Therefore, the data frame array in one data block DB is a repeated array every two data frames DF.

  Next, as shown in FIG. 37, an ECC unit LDCp consisting of parity bits p (i, j) of 32 rows × 304 columns is added to the data block DB consisting of data d (u, v) of 216 rows × 304 columns. (ECC parity) is added. That is, a 32-byte parity bit p (i, j) is added to each column in the data block DB. As a result, a parity addition block PB (LDC (Long-distance code) block) composed of data components e (i, j) of 216 rows × 304 columns and parity bits p (i, j) of 32 rows × 304 columns is formed. Is done. Note that i of the data component e (i, j) and the parity bit p (i, j) indicates a row number in the parity addition block PB, and j indicates a column number in the parity addition block PB. The parity bit p is calculated and added by the following equation.

The ECC unit LDCp is an error correction code for data in the column direction in the parity addition block PB, and is added to generate an LDC syndrome (parity calculation result) for performing error correction for each column of the parity addition block PB. ing.

  Next, the parity added block PB made up of 248 rows × 304 columns of data is converted into 496 (= 248 × 2) rows × 152 (= 304/2) columns of LDC clusters LDCc (see FIG. 39).

  First, the data component e (i, j) and the parity bit p (i, j) are first interleaved, and the data component e (i, j) and the parity bit p (i, j) are located at the position f in FIG. (M, n) is interleaved according to a predetermined rule. Here, m at the position f (m, n) indicates a row number in the LDC cluster LDCc, and n indicates a column number in the LDC cluster LDCc.

  The first interleaving process will be described in detail with reference to FIGS. 37 and 38. Data components e (i, j) of j = 0, 2, 4, 6, 8,..., 302 (even columns) and parity bits p ( i, j) is interleaved at position f (m (= 2i), n (= j / 2)). Further, the data component e (i, j) and the parity bit p (i, j) of j = 1, 3, 5, 7, 9,..., 303 (odd column) are located at the position f (m (= 2i + 1). ), N (= (j−1) / 2)). For example, the data component e (1,2) in the even column is interleaved at the position f (2,1) because m = 2 × 1 = 2 and n = 2/2 = 1. Further, the data component e (1, 3) in the odd-numbered column is interleaved at the position f (3, 1) because m = 2 × 1 + 1 = 3 and n = (3-1) / 2 = 1. As a result, the block is converted into a block that has been subjected to the first interleave processing and includes 496 rows × 152 columns of data shown in FIG.

  Next, the first interleaved data (data component e and parity bit p) shown in FIG. 38 is second interleaved to form the LDC cluster LDCc shown in FIG. The second interleaving process will be described in detail. The data component e (i, j) and the parity bit p (i, j) arranged at the position f (m, n) are the same by the shift amount SI bytes in the following equation. Shift left in line.

That is, the data component e (i, j) and the parity bit p (i, j) arranged at the position f (m, n (= n1)) are the position f (m, n (n2 = n1-SI)). Is interleaved. However, since this shift direction is cyclic, when n2 becomes a negative value, n2 = −1 corresponds to n = 151, and n2 = −2 corresponds to n = 150. As the absolute value of n2, which is a negative value, increases, the actual column number n of the shift destination decreases sequentially from 151.

  For example, each data component e arranged at the position f where m = 2, 3 is shifted to the left in the same row by SI = mod (1 × 3, 152) = 3 bytes. Therefore, the data component e (1, 0) and the data component e (1, 1) respectively arranged at the position f (2, 0) and the position f (3, 0) in FIG. 38 are as shown in FIG. Are interleaved at position f (2,149) and position f (3,149) of column number n = 149 corresponding to n2 = 0-3 = -3. Similarly, each data component e arranged at the position f where m = 2, 3 is all interleaved at the position f shifted by 3 bytes to the left in the same row.

As described above, when the first interleaving process and the second interleaving process are completed for the parity added block PB, the LDC cluster LDCc shown in FIG. 39 is formed.
Next, as shown in FIG. 40, a B96 (Burst Indicator Subcode) cluster BISc of 496 rows × 3 columns composed of a data block (720 bytes) including address information, user control data, and the like and parity (768 bytes) Interleaved with the LDC cluster LDCc. Specifically, as shown in FIG. 40B, the BIS cluster BISc is divided for each column, and the divided BIS cluster BISc (B1 to B3 in FIG. 40A) is divided into 38 columns of the LDC cluster LDCc. Are inserted one column at a time.

  Then, an ECC cluster ECCc (see FIG. 40B) composed of the LDC cluster LDCc and the BIS cluster BISc is modulated by the 17PP modulation method or the like and written onto the BD. With such a format, high error detection / correction capability is ensured in error correction processing during data reproduction.

  Next, an operation of reading data written on the BD in this way will be described. As shown in FIG. 41, in the controller 110 that inputs / outputs data to / from a BD as a recording medium, first, the demodulation circuit 111 reads data from the BD, and buffers the data obtained by demodulating the data, that is, the ECC cluster ECCc. Store in memory 120. Next, the LDC syndrome generation circuit 112 rearranges the data so that the LDC cluster LDCc in the ECC cluster ECCc stored in the buffer memory 120 is in the arrangement order before being rearranged by the first and second interleaving processes. (Read each data while performing the first and second deinterleave processing). That is, the LDC syndrome generation circuit 112 receives the parity addition block PB (see FIG. 37) before the first and second interleaving processes are performed. Then, the LDC syndrome generation circuit 112 generates an LDC syndrome based on the input parity addition block PB and outputs the LDC syndrome to the error correction circuit 114.

  Further, the BIS syndrome generation circuit 113 reads out the BIS cluster BISc in the ECC cluster ECCc stored in the buffer memory 120. The BIS syndrome generation circuit 113 generates a BIS syndrome based on the read BIS cluster BISc, and outputs the BIS syndrome to the error correction circuit 114. The error correction circuit 114 corrects errors such as user data UD in the ECC cluster ECCc stored in the buffer memory 120 based on the input LDC syndrome and BIS syndrome.

  Next, the descrambling circuit 115 reads the data component e corresponding to the user data UD in the error-corrected LDC cluster LDCc from the buffer memory 120 while performing the first and second deinterleaving processes. That is, the data block DB (see FIG. 36B) subjected to error correction processing is input to the descrambling circuit 115. Then, the descrambling circuit 115 generates a scramble value SC used at the time of the scramble processing for each data frame DF, and performs exclusive logic between the lower 8 bits of the scramble value SC and each byte of the read data block DB. The sum is taken in order and the descrambling process is executed. Accordingly, the descrambling data dD0 to dD2051 is generated by the following equation.

Next, the EDC check circuit 116 transfers the descrambled data block DB (user data UD, etc.) input from the descrambling circuit 115 to an external device such as a host computer and the descrambled data. EDC check is performed based on the block DB. The EDC check in the EDC check circuit 116 first generates an EDC syndrome EDCS for each data frame DF (2052 bytes). Specifically, as shown in FIG. 42, descrambling data is input from descrambling circuit 115 to EDC check circuit 116 in the order of dD0 → dD1 → dD2 →... → dD2049 → dD2050 → dD2051. A polynomial expression (input polynomial) representing the order weight of this descrambling data dDu (2052 bytes) is:

It becomes. Here, since the EDC operation is an operation in which each order of the input polynomial is weighted by the vector β, the EDC syndrome EDCS is calculated by the following equation.

That is, the EDC syndrome EDCS can be obtained by performing a byte of data (e.g., descramble data DD0, DD2050, etc.) each time for calculating a primary weighting vectors beta ¹ to the data. Therefore, the order of the vector β to be weighted becomes higher as the data is input earlier, and the order of the vector β to be weighted becomes lower as the data is input later. For example, the order of the vector β corresponding to the data dD0 having the earliest input order is the highest order 2051 (β ²⁰⁵¹ ), and the order of the vector β corresponding to the data dD2051 having the latest input order is the lowest order. 0th order (β ⁰ = 1). The weighting of the 1-byte data X ⁿ with the vector β ¹ means that the weighted data X ⁿ is a predetermined generator polynomial G (x) (in this example, G (x) = x ³² + x ³¹ + x ⁴ + 1) to shift by 1 byte. Then, the EDC check circuit 116 confirms that the error correction is normally completed when the calculated EDC syndromes are all “0”. After this confirmation, the EDC check is terminated.

Incidentally, in recent years, with the increase in processing speed of a host computer, there has been a demand for an increase in data reading / writing speed with respect to a storage device. However, the increase in reading speed is hindered by the limit of the access capability of the buffer memory 120. That is, since the following accesses (1) to (5) are performed on the buffer memory 120, it is difficult to increase the access speed and the error correction processing cannot be increased.
(1) Input of the ECC cluster ECCc from the demodulation circuit 111.
(2) Reading of the LDC cluster LDCc by the LDC syndrome generation circuit 112.
(3) Reading of the BIS cluster BISc by the BIS syndrome generation circuit 113.
(4) Error correction processing of the error correction circuit.
(5) Reading the error-corrected LDC cluster LDCc by the descrambling circuit 115.

  Further, the LDC syndrome generation circuit 112 (descrambling circuit 115) having the above configuration adds parity to the LDC cluster LDCc (data component e in the LDC cluster LDCc) from the buffer memory 120 while performing the first and second deinterleaving processes. It is necessary to read as a block PB (data block DB). Here, since the external buffer memory 120 is generally composed of an SDRAM that is inexpensive and has a low processing speed, when the LDC cluster LDCc is read out while being rearranged, each data of the parity addition block PB is read out in pieces. It will be. As described above, there is a problem that the buffer memory 120 is accessed with a small data access size and inefficient, and the access bandwidth of the buffer memory 120 is reduced.

  Therefore, in order to reduce the access amount to the buffer memory 120, as shown in FIG. 43, one ECC cluster ECCc is temporarily stored in the built-in memory 131, and the LDC syndrome generation circuit 112 and the BIS syndrome are stored in the built-in memory 131. A controller 130 is proposed in which the generation circuit 113 is directly accessed. As a result, the amount of access to the buffer memory 120 by the LDC syndrome generation circuit 112 and the BIS syndrome generation circuit 113 is eliminated, so that the total access amount to the buffer memory 120 can be reduced and the speed of error correction processing can be increased. Can do.

JP2003-242720A.

  However, in order to perform error correction processing by the error correction circuit 114, data in ECC cluster ECCc units is required. Therefore, in the controller 130 shown in FIG. 43, the built-in memory 131 is a large-capacity memory that can store data for one ECC cluster. Therefore, a buffer memory corresponding to the capacity of one ECC cluster ECCc is built in, and a new problem arises that the circuit scale of the controller increases.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to reduce the access amount to the external memory while suppressing an increase in circuit scale due to an increase in the built-in memory capacity. An object of the present invention is to provide an error correction apparatus.

To achieve the above object, a first aspect of the present invention, the scrambling process is performed on a predetermined number of data frames, scrambled data frame is converted into data blocks of a predetermined size, to each column of the data block parity is generated and added parity added block, the parity adding block interleaving against there is subjected clusters are generated, the cluster is a error correction device for correcting errors in data modulated, the modulation A demodulating circuit that demodulates the generated data to generate a cluster as demodulated data, a built-in memory unit that stores a divided cluster obtained by dividing the cluster from the demodulating circuit every predetermined number of rows, and the built-in memory unit Through the parity added block generated by deinterleaving the divided cluster. That each data is input in the column direction, the parity calculation result generating circuit for generating a parity operation results in each column of the parity addition block based on the data, via the internal memory unit, deinterleaving processing on the divided clusters is performed, the scramble data in the data block is inputted as a read block in a column direction, said the EDC syndrome generation circuit for generating EDC syndrome based on the scramble data, the EDC syndrome EDC syndrome generated by the generating circuit An EDC memory unit that stores an intermediate result for each data frame; and an error correction circuit that performs an error correction process on the scrambled data based on the parity calculation result. The EDC syndrome generation circuit includes: Built-in memo And the scrambling data input as byte data from the parts, taking an exclusive OR of the shift operation value input from the channel selection択回passage, a first exclusive OR circuit which generates an EDC calculated value, the first against EDC operation value input from the 1 XOR generates the shift operation value by weighting for one byte shift in accordance with a predetermined generate polynomial and stores the shift operation value to the EDC memory unit For the EDC operation value input from the first weighting circuit and the first exclusive OR,
{(Total number of bytes in the column direction of the data block) + 1− (number of bytes in the column direction of each read block)}
Bytes, relative to the front and a second weighting circuit for generating a shift operation value by weighting shifting according Kisei formed polynomial, EDC computation value input from the first exclusive OR circuit, before Kisei formed by weighting to 108 bytes shifted according polynomial to generate a shift operation value, a third weighting circuit for storing the shift operation value to the EDC memory unit, and the shift operation value from said second weighting circuit, the EDC taking the exclusive OR of the intermediate result of EDC syndrome from the memory unit, and a second exclusive OR circuit for generating a shift operation value comprises, before hexene択回path, the first exclusive depending on the position in the data block of the byte data input to the oR circuit, said first weighting circuit, or the second weighting circuit and said second exclusive oR circuit Processing one of the shift operation value is output to the first XOR circuit, wherein the data block from the first data frame to be processed the data frame every other one of the shift operation value input When the data frame to be processed is an even-numbered data frame and the data frame processed next to the even-numbered data frame is an odd-numbered data frame, the EDC memory unit stores the final data frame in each data frame. An EDC operation value generated when byte data in the last row of rows is input to the first exclusive OR circuit is stored as an intermediate result of the EDC syndrome through the first weighting circuit, and the EDC syndrome generation circuit To each even data frame of the last read block that is input last An EDC operation value generated when byte data in the last row and last column is input to the first exclusive OR circuit is stored as the EDC syndrome through the third weighting circuit. An EDC operation value generated when byte data in the last row and last column in each odd data frame is input to the first exclusive OR circuit is stored as the EDC syndrome .

  According to the above configuration, since the EDC syndrome generation circuit and the parity calculation result generation circuit can access the built-in memory unit independently, it is possible to execute EDC syndrome generation and parity calculation processing in parallel. The time required for error correction processing can be shortened. In addition, since the EDC syndrome generation circuit is provided with the first to third weighting circuits, the third exclusive OR circuit, and the second selection circuit, even for byte data that is input in a different order from when the EDC is added, The same weighting as the number of bytes can be performed when the EDC calculation processing is performed in the same data processing order as when EDC is added.

  As described above, according to the present invention, it is possible to provide an error correction device capable of reducing the access amount to the external memory while suppressing an increase in circuit scale due to an increase in the built-in memory capacity. .

The block diagram which shows an optical disk control apparatus. 1 is a block diagram showing an optical disk controller in a first embodiment. Explanatory drawing which shows the division | segmentation ECC block stored in a built-in memory. Explanatory drawing which shows the read block read by the LDC syndrome production | generation circuit. The block diagram which shows a built-in memory part. The block diagram which shows the descrambling circuit in 1st Embodiment. Explanatory drawing which shows the read-out block read by a descrambling circuit. The block diagram which shows the EDC syndrome production | generation circuit in 1st Embodiment. Explanatory drawing which shows an external buffer memory. The flowchart which shows the operation | movement of a descrambling process. The flowchart which shows the operation | movement of a descrambling process. The flowchart which shows the operation | movement of a descrambling process. Explanatory drawing which shows the input polynomial input into an EDC syndrome production | generation circuit. The flowchart which shows operation | movement of the EDC arithmetic processing in 1st Embodiment. Explanatory drawing which shows the logical expression of the input / output in 1 shift arithmetic unit of a descrambling circuit. Explanatory drawing which shows the logical expression of the input / output in the 209 shift calculator of a descrambling circuit. Explanatory drawing which shows the logical expression of the input / output in the 108 shift calculator of a descrambling circuit. Explanatory drawing which shows the logical expression of the input / output in 1 shift arithmetic unit of an EDC syndrome production | generation circuit. Explanatory drawing which shows the logical expression of the input / output in the 209 shift computing unit of an EDC syndrome production | generation circuit. Explanatory drawing which shows the logical expression of the input / output in the 108 shift arithmetic unit of an EDC syndrome production | generation circuit. Explanatory drawing which shows the logical expression of the input / output in the -211 shift computing unit of an EDC syndrome production | generation circuit. The block diagram which shows the descrambling circuit in 2nd Embodiment. The block diagram which shows the EDC syndrome production | generation circuit in 2nd Embodiment. (A)-(d) is explanatory drawing which shows the data processing order in 2nd Embodiment. Explanatory drawing which shows the data storage method to the external buffer memory in 2nd Embodiment. (A), (b) is explanatory drawing which shows the input polynomial input into an EDC syndrome production | generation circuit. The block diagram which shows the EDC syndrome production | generation circuit in 3rd Embodiment. The flowchart which shows operation | movement of the EDC arithmetic processing in 3rd Embodiment. Explanatory drawing which shows the logical expression of the input / output in the beta- ¹ shift arithmetic unit of an EDC syndrome production | generation circuit. Explanatory drawing which shows the input / output logic type ^| formula in the (beta) ^-1664 shift arithmetic unit of an EDC syndrome production ^| generation circuit. The block diagram which shows the optical disk controller in 4th Embodiment. The block diagram which shows the optical disk controller in 5th Embodiment. The block diagram which shows the optical disk controller in another example. Explanatory drawing which shows user data. The block diagram which shows the conventional scramble value generation circuit. (A), (b) is explanatory drawing which shows a data block. Explanatory drawing which shows a LDC block. Explanatory drawing which shows a 1st interleave process. Explanatory drawing which shows a LDC cluster. (A), (b) is explanatory drawing which shows an ECC cluster. The block diagram which shows the conventional optical disk controller. Explanatory drawing which shows the conventional descrambling process. The block diagram which shows the conventional optical disk controller.

(First embodiment)
A first embodiment embodying the present invention will be described below with reference to FIGS.
As shown in FIG. 1, an optical disk control device 1 as a data reading device is connected to a computer 2 via a predetermined interface such as ATAPI (AT attachment packet interface). The optical disk control device 1 is connected to the optical disk drive device 3 through an interface.

  The optical disk drive 3 rotates and drives the Blu-ray disk (BD) 4 at a predetermined speed, and reads the disk data stored in the BD 4 with an optical pickup (not shown). Then, the optical disk drive device 3 outputs the disk data to the optical disk control device 1.

  The disk data is input from the optical disk driving device 3 to the input / output drive circuit 5 of the optical disk control device 1, and the disk data is output to an optical disk controller (controller) 10 as an error correction device.

  The controller 10 transmits instructions to the optical disk drive 3 and receives status, decodes the read format from the BD 4 that is an optical disk, corrects errors, transfers data between the optical disk drive 3 and the external buffer memory 6, and an interface circuit. Each process such as data transfer between 60 and the external buffer memory 6 is performed. That is, the controller 10 performs descrambling processing on the disk data input from the input / output drive circuit 5 to generate descrambling data, and generates various syndromes such as EDC syndrome, LDC syndrome, and BIS syndrome. The controller 10 stores these descrambling data and various syndromes in the external buffer memory 6. Further, the controller 10 performs error correction processing on the descrambling data stored in the external buffer memory 6 based on various syndromes stored in the external buffer memory 6. Then, the controller 10 transfers the user data UD after error correction stored in the external buffer memory 6 to the computer 2 via the interface circuit 60 based on the instruction of the microprocessor 8.

Next, the internal configuration of the optical disk controller 10 shown in FIG. 1 will be described with reference to FIG.
Disc data read from the BD 4 is sequentially input to the demodulation circuit 11 of the optical disc controller 10 through the input / output drive circuit 5. The demodulating circuit 11 converts the input disk data into digital data and generates a clock signal CLK synchronized with the digital data. The demodulation circuit 11 further demodulates the digital data to generate the demodulated data, that is, the ECC cluster ECCc (see FIG. 40B), and outputs the ECC cluster ECCc and the clock signal CLK.

  The built-in memory unit 12 to which data of the ECC cluster ECCc (cluster) is input from the demodulation circuit 11 includes two buffer memories M1 and M2. Each of the buffer memories M1 and M2 has a memory capacity capable of storing data of a predetermined number of bytes (2480 bytes in this embodiment). In the present embodiment, the ECC cluster ECCc (76880 bytes) is divided into 31 (= 76880/2480) pieces and stored in the buffer memories M1 and M2 of the built-in memory unit 12 every 2480 bytes. That is, the ECC cluster ECCc (see FIG. 40B) is divided every 16 rows and stored in each buffer memory M1, M2 in block units (divided ECC cluster DECC) of 16 rows × 155 columns (= 2480 bytes). The FIG. 3 shows a divided ECC cluster DECC (divided cluster) of 16 rows × 155 columns first stored in the built-in memory unit 12 from the demodulation circuit 11. As shown in FIG. 3, each data from the first row to the 16th row in each column of the ECC cluster ECCc is first stored in the built-in memory unit 12 from the demodulation circuit 11.

  The built-in memory unit 12 uses one of the buffer memory M1 and the buffer memory M2 for data storage for storing data from the demodulation circuit 11, and the other is used for the LDC syndrome generation circuit 13 to be described later, etc. It is used for access to be accessed. That is, when the buffer memory M1 is used for data storage, the buffer memory M2 is used for access, and when the buffer memory M2 is used for data storage, the buffer memory M1 is used for access. .

  FIG. 5 shows a circuit configuration example of the built-in memory unit 12. As shown in FIG. 5, the data input from the demodulation circuit 11 is input to the first selector 12a. A selection signal from the selection circuit 12c is input to the first selector 12a. The selection circuit 12c counts the clock signal CLK input from the demodulation circuit 11, and switches the signal level of the selection signal at a predetermined timing. In this embodiment, the selection circuit 12c switches the signal level of the selection signal every time it counts 2480 bytes of clock. This selection signal is also input to the second selector 12b via an inverter.

  The first selector 12a outputs the data from the demodulation circuit 11 to the buffer memory M1 or the buffer memory M2 based on the signal level of the selection signal from the selection circuit 12c. Further, the second selector 12b enables access to either the buffer memory M1 or the buffer memory M2 based on the signal level of the selection signal from the selection circuit 12c input via the inverter.

  As shown in FIG. 2, the LDC syndrome generation circuit 13 (parity calculation result generation circuit) rearranges the LDC cluster LDCc in the divided ECC cluster DECC stored in the buffer memory of the internal memory unit 12 selected to be accessible. Read while. As described above, the LDC cluster LDCc (see FIG. 39) in the ECC cluster ECCc read from the BD 4 is rearranged by the data of the parity addition block PB (see FIG. 37) being subjected to the first interleave process and the second interleave process. It is a block. Further, since the ECC unit LDCp in the LDC cluster LDCc is a parity added to the data block DB before the first and second interleaving processes (see FIG. 36B), the data arrangement order in the LDC cluster LDCc As it is, the LDC syndrome (parity calculation result) based on the ECC unit LDCp cannot be generated accurately.

  Therefore, the LDC syndrome generation circuit 13 according to the present embodiment performs the first and second deinterleave processing on the LDC cluster LDCc in the divided ECC cluster DECC stored in the built-in memory unit 12, and arranges the parity added block PB. Each data is read out while rearranging so that The reading order will be described by taking the data block first stored in the built-in memory unit 12 as an example. First, the LDC syndrome generation circuit 13 first outputs the data component e (at the position f (0, 0) shown in FIG. 0, 0) is read, then the data component e (1, 0) at position f (2, 149) is read, and then the data component e (2, 0) at position f (4, 146) is read. In the same manner, the LDC syndrome generation circuit 13 performs the data component e (3, 0), the data component e (4, 0),..., The data component e (7, 0) as indicated by the solid arrows in FIG. The data components in the first column in the parity added block PB are read out in the order of). Subsequently, the LDC syndrome generation circuit 13 performs the data components e (0, 1),..., Data component e (7, 1) in order of the data from the first row to the eighth row in the second column in the parity addition block PB. Read the component. Similarly, the LDC syndrome generation circuit 13 sequentially reads out the data components e of the respective columns in the parity addition block PB. As described above, when the LDC cluster LDCc in the 31 divided ECC clusters DECC is read by performing the first and second deinterleaving processes, as shown in FIG. 4, it is divided into 31 divided 8 rows × 304 columns. The parity addition block PB is read as a read block RB (first to 31st read blocks RB1 to RB31).

  The LDC syndrome generation circuit 13 reads out the data in the reading order, and generates an intermediate value of the LDC syndrome of each column for each number of bytes in the column direction (8 bytes in the present embodiment) in each read block RB. The intermediate value of the LDC syndrome is stored in the LDC memory unit 14. Further, the LDC syndrome generation circuit 13 reads the intermediate value of the LDC syndrome in the same column stored in the previous read block RB (for example, the first read block RB1 in the case of the second read block RB2) from the LDC memory unit 14. . Then, the LDC syndrome generation circuit 13 updates the intermediate value of the LDC syndrome based on the intermediate value of the read LDC syndrome and the data in the column direction of the current read block RB (for example, the second read block RB2). Then, the updated LDC syndrome is stored in the LDC memory unit 14.

  As shown in FIG. 2, the LDC memory unit 14 includes two LDC buffer memories M3 and M4. Each of the buffer memories M3 and M4 has a memory capacity capable of storing LDC syndromes of all columns in one parity addition block PB, that is, a memory capacity capable of storing 9728 (= 32 × 304) bytes. In the LDC memory unit 14, the intermediate value of the LDC syndrome stored after the end of data reading of the entire one parity addition block PB (31 times: first to 31st read blocks RB1 to RB31) becomes the final LDC syndrome. Therefore, the final LDC syndrome for each column is stored in the external buffer memory 6. The LDC memory unit 14 uses one of the buffer memory M3 and the buffer memory M4 for the syndrome storage for storing the intermediate value of the LDC syndrome from the LDC syndrome generation circuit 13 at the same time. A new LDC syndrome is used for outputting a syndrome to the external buffer memory 6. Note that the LDC memory unit 14 has substantially the same configuration as the built-in memory unit 12.

  In the BIS memory unit 15 connected to the BIS syndrome generation circuit 16, the BIS cluster BISc in the divided ECC cluster DECC stored in the buffer memory of the internal memory unit 12 selected to be accessible is sequentially stored in the column direction. Is done. For example, in the case of the data block shown in FIG. 3, B (0,0) to B (15,0) are stored in the BIS memory unit 15 in the order indicated by solid line arrows in FIG. When the data of the entire BIS cluster BISc is stored, the BIS memory unit 15 outputs the BIS cluster BISc to the BIS syndrome generation circuit 16.

  As shown in FIG. 2, the BIS memory unit 15 is composed of two BIS buffer memories M5 and M6. Each of the buffer memories M5 and M6 has a memory capacity capable of storing one BIS cluster BISc, that is, a memory capacity capable of storing 1488 (= 496 × 3) bytes. The BIS memory unit 15 uses one of the buffer memory M5 and the buffer memory M6 for data storage for storing the data of the BIS cluster BISc from the built-in memory unit 12, and the other is used as the stored BIS at the same time. The cluster BISc is used for data output to be output to the BIS syndrome generation circuit 16. The BIS memory unit 15 has substantially the same configuration as the built-in memory unit 12.

  The BIS syndrome generation circuit 16 reads the BIS cluster BISc from the BIS memory unit 15 and generates a BIS syndrome based on the BIS cluster BISc. The BIS syndrome generation circuit 16 stores the generated BIS syndrome in the external buffer memory 6.

  Similar to the LDC syndrome generation circuit 13, the descrambling circuit 20 reads the divided LDC cluster LDCc stored in the buffer memory of the built-in memory unit 12 selected to be accessible as a read block RB. Here, among the 1st to 31st read blocks RB1 to RB31, each data of the ECC unit LDCp constituting the 28th to 31st read blocks RB28 to RB31 (data on the 217th to 248th rows) is scrambled. It has not been processed. Therefore, as shown in FIG. 7, the descrambling circuit 20 includes the first to 27th read blocks RB1 to RB27 (the data block DB shown in FIG. 36B) excluding the 28th to 31st read blocks RB28 to RB31. Read only. The descrambling circuit 20 performs descrambling processing on each scrambled data for each read block RB, and releases a predetermined scramble applied to the data. The descrambling circuit 20 outputs descrambling data dDu subjected to descrambling processing to the external buffer memory 6 and the EDC syndrome generation circuit 30. FIG. 7 shows the reading order of the scrambled data sDu of the first to 27th read blocks RB1 to RB27 read by the descrambling circuit 20 by the row number u in the scrambled data frame SDF (see FIG. 36A). It was. The row number u is a number indicating the processing order for each data frame DF at the time of scramble processing, and 0 ≦ u ≦ 2051.

  FIG. 6 shows a circuit configuration example of the descrambling circuit 20. As shown in FIG. 6, the scrambled data sDu read from the built-in memory unit 12 is input to the EOR circuit 21. The EOR circuit 21 receives the scramble value SC generated by the scramble value generation circuit 22. The EOR circuit 21 generates the descrambling data dDu by taking the exclusive OR of the scrambled data sDu from the built-in memory unit 12 and the lower 8 bits of the predetermined scramble value SC, and the descrambling data dDu is stored in the external buffer. The data is output to the memory 6 and the EDC syndrome generation circuit 30.

  Here, the descrambling process will be described. As described above, when the initial value of the scramble value SC is S0, the descramble data dDu for the scramble data sDu is generated by the following equation.

In this way, the descrambling process can be executed by taking the exclusive OR of the same scramble value SC and the scrambled data sDu as in the scramble process. That is, by taking an exclusive OR of the scrambled data sDu read in the same order as in the scramble process and the scramble value SC generated in the same order as in the scramble process, all the descrambled data dD0 ~ DD2051 can be generated accurately. However, in the case of this embodiment, the data block DB is divided into 31 parts and read out, so the data reading order is different from that in the scramble process. Therefore, as in the scramble process, when the scramble value SC is generated by shifting one byte at a time, the exclusive OR of the scramble value SC and each data different from the scramble process is taken, and the descramble process is accurately performed. Can not run.

Therefore, in this embodiment, 1-shift arithmetic units 24 and 209 for shifting the shift operator γ by the desired number of bytes in the scramble value generation circuit 22 in accordance with the scramble data sDu input to the EOR circuit 21. The shift calculator 25 and the 108 shift calculator 26 are provided. The scramble value SC is input from the flip-flop (FF) circuit 23 to the 1-shift calculator 24 and the 209-shift calculator 25. The 1 shift computing unit 24 is a computing unit that outputs S0 · γ ^{n + 1} as the scramble value SC to the selector 27 when S0 · γ ⁿ is inputted as the scramble value SC from the FF circuit 23. That is, the 1-shift calculator 24 is a calculator that shifts the inputted scramble value by 1 byte according to the generator polynomial Φ (x) to generate a new scramble value SC. FIG. 15 shows a logical expression of O [15: 0] output from the 1-shift calculator 24 when I [15: 0] is input to the 1-shift calculator 24.

The 209 shift computing unit 25 is a computing unit that outputs S0 · γ ^{n + 209} as the scramble value SC to the selector 27 when S0 · γ ⁿ is input from the FF circuit 23 as the scramble value SC. In other words, the 209 shift computing unit 25 is a computing unit that shifts the inputted scramble value by 209 bytes according to the generator polynomial Φ (x) to generate a new scramble value SC. FIG. 16 shows a logical expression of O [15: 0] output from the 209 shift calculator 25 when I [15: 0] is input to the 209 shift calculator 25.

The 108 shift calculator 26 is a calculator that outputs S0 · γ ¹⁰⁸ to the odd register R3 as the scramble value SC when the initial value S0 of the scramble value SC is input from the initial value register R1. That is, the 108 shift calculator 26 is a calculator that shifts the inputted scramble value by 108 bytes according to the generator polynomial Φ (x) to generate a new scramble value SC. FIG. 17 shows a logical expression of O [15: 0] output from the 108 shift calculator 26 when I [15: 0] is input to the 108 shift calculator 26. Note that O [7] to O [0] in FIGS. 15 to 17 are data that are actually exclusive-ORed with the scrambled data sDu in the EOR circuit 21.

The selector 27 receives the scramble value SC from the initial value register R1, the even number register R2, and the odd number register R3 together with the various arithmetic units 24 and 25. Here, the initial value S0 of the scramble value SC input from the initial value register R1 is stored in the even register R2 at the time of initial setting. Further, the even register R2 stores the scramble value SC for the scrambled data in the last row of the first column in the last even data frame (data frame DF30) in each read block RB through the 1 shift calculator 24. On the other hand, the scramble value SC (= S0 · γ ¹⁰⁸ ) input from the initial value register R1 through the 108 shift calculator 26 is stored in the odd-numbered register R3 at the time of initial setting. The odd register R3 stores the scramble value SC for the scrambled data in the last row of the first column in the last odd data frame (data frame DF31) in each read block RB through the 1-shift calculator 24.

  The selector 27 receives a selection signal from the selection circuit 28. Based on the selection signal from the selection circuit 28, the selector 27 outputs one scramble value SC among the scramble values input from the various arithmetic units 24 and 25 and the registers R 1 to R 3 to the FF circuit 23. The selection circuit 28 generates a selection signal according to the number of pulses of the input clock signal CLK, that is, according to the position of data to be read (position in the data block DB), and outputs the selection signal to the selector 27. . For example, when the first scrambled data in the first row and the first column in each even data frame in each read block RB is input to the EOR circuit 21, the selection circuit 28 supplies the scramble value SC from the even register 28 to the FF circuit 23. A selection signal to be output is generated. The selection circuit 28 causes the FF circuit 23 to output the scramble value SC from the odd register 29 when the first scramble data in the first row and the first column in each odd data frame in each read block RB is input to the EOR circuit 21. A selection signal is generated. The selection circuit 28 generates a selection signal that causes the FF circuit 23 to output the scramble value SC input from the 1-shift arithmetic unit 24 when the scramble data of the second and subsequent rows of each read block is input to the EOR circuit 21. . When the scramble data of the first row of each column after the second column in each data frame DF of each read block is input to the EOR circuit 21, the selection circuit 28 receives the scramble value SC input from the 209 shift computing unit 25. Is generated by the FF circuit 23.

  The FF circuit 23 latches the scramble value SC input from the selector 27 to the D terminal based on the pulse of the clock signal CLK input to the CK terminal, and the latched scramble value SC from the Q terminal to the EOR circuit 21. And output to various arithmetic units 24 and 25.

  As shown in FIG. 2, the EDC syndrome generation circuit 30 performs EDC calculation based on the descrambling data dDu input from the descrambling circuit 20 as byte data, and generates an EDC syndrome. That is, the EDC syndrome generation circuit 30 generates an EDC syndrome for the descrambling data dDu for each data frame DF in the read block RB, and stores the EDC syndrome in the EDC memory unit 40.

  FIG. 8 shows a circuit configuration example of the EDC syndrome generation circuit 30. As shown in FIG. 8, the descrambling data dDu input from the descrambling circuit 20 is input to the EOR circuit 31. The EOR circuit 31 receives the shift operation value Es from the selector 32. The EOR circuit 31 calculates the exclusive OR of the descrambling data dDu and the lower 8 bits of the shift operation value Es and outputs the EDC operation value Er (c) to the FF circuit 33. Note that r in the EDC operation value Er (c) is a read block RB number, 1 ≦ r ≦ 27, and c represents the number of pulses of the clock signal CLK in each data frame DF.

  Here, the EDC calculation will be described. As described above, when descrambling data is input in the order of dD0 → dD1 →... → dD2050 → dD2051, similarly to the data arrangement (data processing order) when adding 4-byte EDC, EDC The syndrome EDCS is generated by the following equation.

Thus, operation of the EDC syndrome EDCS is weighted by the primary vector beta ¹ for each data every time 1-byte data is arithmetic processing is performed. Therefore, the order of the vector β to be weighted becomes higher as the data is input earlier, and the order of the vector β to be weighted becomes lower as the data is input later. For example, the input data (byte data) X ²⁰⁵¹ input earliest in each data frame DF is weighted 2051 times by the primary vector β ¹ . However, in the present embodiment, as shown in FIG. 7, the descrambling data from the descrambling circuit 20 to the EDC syndrome generation circuit 30 is in the order of dD0 → dD1 →... → dD7 → dD216 → dD217 →. It is input with. Thus, when descrambling data is input in an order different from the data processing order at the time of EDC addition, weighting by the vector β cannot be performed accurately, so that a normal EDC syndrome cannot be generated.

Therefore, as shown in FIG. 8, the EDC syndrome generation circuit 30 of the present embodiment includes a 1-shift calculator 34 and a 209-shift calculator 35 for performing weighting with a vector β ⁿ of a desired order n according to byte data. , 108 shift computing unit 36 and -211 shift computing unit 37. The EDC calculation value Er (c) is input from the FF circuit 33 to these various calculators 34 to 37. That is, the FF circuit 33 latches the EDC operation value Er (c) input to the D terminal based on the clock signal CLK input to the CK terminal, and the EDC operation value Er (c) from the Q terminal. The data is output to various arithmetic units 34 to 37 and the EDC memory unit 40.

The 1-shift computing unit 34 is a computing unit that outputs Er (c) · β ¹ to the selector 32 as a shift computation value when the EDC computation value Er (c) is input from the FF circuit 33. That is, the 1-shift computing unit 34 is a computing unit that generates a shift computation value Es by weighting an input EDC computation value Er (c) by one byte according to the generator polynomial G (x). . FIG. 18 shows a logical expression of O [31: 0] that is output from the 1-shift calculator 34 when I [31: 0] is input to the 1-shift calculator 34.

The 209 shift arithmetic unit 35 is an arithmetic unit that outputs Er (c) × β ²⁰⁹ as a shift arithmetic value to the selector 32 and the EOR circuit 38 when the EDC arithmetic value Er (c) is input from the FF circuit 33. . In other words, the 209 shift computing unit 35 is a computing unit that generates the shift computation value Es by weighting the input EDC computation value Er (c) by 209 bytes according to the generator polynomial G (x). . FIG. 19 shows a logical expression of O [31: 0] outputted from the 209 shift computing unit 35 when I [31: 0] is inputted to the 209 shift computing unit 35.

The 108 shift computing unit 36 is a computing unit that outputs Er (c) × β ¹⁰⁸ to the selector 32 as the intermediate value EDCs of the EDC syndrome when the EDC computed value Er (c) is input from the FF circuit 33. That is, the 108 shift computing unit 36 is a computing unit that generates a shift computation value Es by weighting the input EDC computation value Er (c) by 108 bytes according to the generator polynomial G (x). . FIG. 20 shows a logical expression of O [31: 0] output from the 108 shift calculator 36 when I [31: 0] is input to the 108 shift calculator 36.

The −211 shift arithmetic unit 37 is an arithmetic unit that outputs Er (c) × β− ²¹¹ to the selector 32 as the intermediate value EDCs of the EDC syndrome when the EDC arithmetic value Er (c) is input from the FF circuit 33. is there. In other words, the -211 shift computing unit 37 performs a weighting that reversely shifts the input EDC computation value Er (c) by 211 bytes according to the generator polynomial G (x) to generate the shift computation value Es. It is a vessel. FIG. 21 shows a logical expression of O [31: 0] output from the −211 shift calculator 37 when I [31: 0] is input to the −211 shift calculator 37.

  In the EDC memory unit 40, an EDC operation value Er (c) for the byte data of the last row and last column in each data frame of each read block RB is stored as an intermediate value EDCs through the 1-shift operation unit 34. In the EDC memory unit 40, the EDC operation value Er (c) for the last byte data in each even data frame of the 14th read block RB is stored as an intermediate value EDCs through the -211 shift operation unit 37. In the EDC memory unit 40, the EDC operation value Er (c) for the byte data of the last row and the last column in each even data frame of the 27th read block RB is stored as the EDC syndrome EDCS through the 108 shift calculator 36. The EDC memory unit 40 stores the EDC operation value Er (c) for the last byte data in each odd data frame of the 27th read block RB from the FF circuit 33 as the EDC syndrome EDCS.

  As shown in FIG. 8, the EDC memory unit 40 includes two EDC buffer memories M7 and M8. Each of the buffer memories M7 and M8 has a memory capacity capable of storing EDC syndrome in all the data frames DF0 to DF31, that is, a memory capacity capable of storing 128 (= 4 × 32) bytes. The EDC memory unit 40 outputs the intermediate value EDCs stored through the 1 shift computing unit 34 and the −211 shift computing unit 37 to the EOR circuit 38. The EDC memory unit 40 outputs the EDC syndrome EDCS from the FF circuit 33 and the 108 shift calculator 36 to the external buffer memory 6. The EDC memory unit 40 uses one of the buffer memory M7 and the buffer memory M8 for the syndrome storage for storing the intermediate value EDCs of the EDC syndrome from the EDC syndrome generation circuit 30 at the same time, and the other is used as the final A typical EDC syndrome EDCS is used for syndrome output to be output to the external buffer memory 6. The EDC memory unit 40 has substantially the same configuration as the built-in memory unit 12.

The EOR circuit 38 obtains the exclusive OR of the shift operation value Es (= Er (c) · β ²⁰⁹ ) from the 209 shift operation unit 35 and the intermediate value EDCs from the EDC memory unit 40, and the operation result is obtained. The shift operation value Es is output to the selector 32.

  Based on the selection signal from the selection circuit 39, the selector 32 outputs one shift calculation value Es among the shift calculation values input from the various calculators 34 and 35 and the EOR circuit 38 to the EOR circuit 31. Further, the selector 32 prevents the shift operation value Es from being output based on the selection signal from the selection circuit 39.

  The selection circuit 39 generates various selection signals according to the number of pulses of the input clock signal CLK, that is, according to the byte data position (position in the data block DB), and outputs the selection signal to the selector 32. For example, the selection circuit 39 outputs a selection signal that causes the EOR circuit 31 to output the shift operation value Es from the 1-shift operation unit 34 when the byte data of the second and subsequent rows of each read block RB is input to the EOR circuit 31. Generate. When the byte data of the first row of each column excluding the first column and the last column of each data frame DF in each read block RB is input to the selection circuit 39, the selection circuit 39 receives the shift calculation value Es from the 209 shift calculator 35. A selection signal to be output to the EOR circuit 31 is generated. The selection circuit 39 generates a selection signal that causes the EOR circuit 31 to output the shift operation value Es from the EOR circuit 38 when the byte data of the first row of the last column in each data frame DF of each read block RB is input. To do. The selection circuit 39 generates a selection signal that prevents the selector 32 from outputting the shift operation value Es when the first byte data in the first row and the first column in each data frame DF of each read block RB is input.

  As shown in FIG. 2, the correction circuit 51 in the error correction circuit 50 reads the LDC syndrome and BIS syndrome from the external buffer memory 6, and based on these LDC syndrome and BIS syndrome, stores the data stored in the external buffer memory 6. Error correction processing is performed on the scrambled data frame DF. The correction circuit 51 outputs an error position and an error value calculated based on various syndromes to the EDC correction circuit 52.

  The EDC correction circuit 52 corrects the EDC syndrome stored in the correction memory unit 53 from the external buffer memory 6 based on the error position and error value input from the correction circuit 51. The EDC check circuit 54 reads the corrected EDC syndrome stored in the correction memory unit 53 and performs an EDC check. Here, the EDC check circuit 116 in the controllers 110 and 130 shown in FIGS. 41 and 43 performs the EDC check at the same time as transferring the data after error correction to the host computer. Therefore, in the conventional controllers 110 and 130, the data after error correction may be transferred to the host computer even when error correction is not normally completed. On the other hand, in the controller 10 of this embodiment, since the EDC check is performed immediately after the error correction processing is completed, it is possible to suppress transfer of erroneously corrected data to the computer 2.

The interface circuit 60 reads out the error-corrected data stored in the external buffer memory 6 based on an instruction from the microprocessor 8 and outputs it to the computer 2.
As shown in FIG. 9, the external buffer memory 6 is composed of a plurality of storage blocks. One storage block is composed of 13000 H bytes. Specifically, in one storage block, the user data UD portion (65536 bytes) of all data frames DF0 to DF31 is stored at addresses 000000H to 00FFFFH, and BIS data (720 bytes) is stored at addresses 010000H to 0103FFH. The In one storage block, the LDC syndrome (9728 bytes) is stored at addresses 010400H to 0129FFH, the BIS syndrome (768 bytes) is stored at addresses 012A00H to 012DFFH, and the EDC syndrome (128 bytes) is stored at addresses 012E00H to 12FFFH. ) Is stored. Therefore, data for one ECC cluster ECCc can be stored in one storage block. As is well known, “H” indicates that the value is a hexadecimal number.

Next, the operation of the optical disk controller 10 configured as described above will be described.
The demodulating circuit 11 reads the disk data from the BD 4 and demodulates it, and the demodulated data (ECC cluster ECCc) is selected as data storage among the first and second buffer memories M1 and M2 of the built-in memory unit 12. Stored in the existing buffer memory. One buffer memory stores a divided ECC cluster DECC of 2480 bytes obtained by dividing the ECC cluster ECCc into 31 pieces.

  The LDC syndrome generation circuit 13 reads out the read blocks RB from the first buffer memory M1 or the second buffer memory M2 selected for access while rearranging the LDC clusters LDCc in the divided ECC cluster DECC. At this time, as shown in FIG. 4, since the read block RB is composed of data of an arbitrary 2432 (8 rows × 304 columns) bytes, the LDC syndrome generation circuit 13 performs every 8 bytes of data in the column direction. That is, the LDC syndrome is generated for each column of the read block RB. The LDC syndrome generation circuit 13 stores the generated LDC syndrome in the LDC memory unit 14 as an intermediate value of the LDC syndrome.

  Next, when the LDC syndrome generation circuit 13 reads the read blocks RB after the second read block RB2 (second to 31st read blocks RB2 to RB31), the LDC syndrome of the same column generated by the previous read block RB is read. The intermediate value is read from the LDC memory unit 14. Then, the LDC syndrome generation circuit 13 generates an LDC syndrome from the read intermediate value of the LDC syndrome and data of 8 bytes in the column direction of the current read block RB, and the LDC syndrome is generated as an intermediate value of the new LDC syndrome. Is stored in the LDC memory unit 14.

  In the LDC memory unit 14, when all the data in one parity addition block PB (first to 31st read blocks RB1 to RB31) has been read, selection for saving syndromes and outputting syndromes in the buffer memory M3 and the buffer memory M4 is performed. Is switched. As a result, the final LDC syndrome for each column in one parity addition block PB is stored in the buffer memory switched from the syndrome storage to the syndrome output. The buffer memory switched for the syndrome output stores the final LDC syndrome for each column in the external buffer memory 6.

  On the other hand, the buffer memory selected for data storage in the BIS memory unit 15 includes the BIS cluster BISc in the divided ECC cluster DECC from the first buffer memory M1 or the second buffer memory M2 selected for access. Stored. In the BIS memory unit 15, when the storage of all data in one BIS cluster BISc is completed, the selection of data storage and data output in the buffer memory M5 and the buffer memory M6 is switched. Next, the BIS syndrome generation circuit 16 reads the BIS cluster BISc from the buffer memory M5 or the buffer memory M6 of the BIS memory unit 15 switched for data output. The BIS syndrome generation circuit 16 generates a BIS syndrome based on the read BIS cluster BISc, and stores the BIS syndrome in the external buffer memory 6.

  The descrambling circuit 20 reads the scrambled data sDu from the first buffer memory M1 or the buffer memory M2 selected for access simultaneously with the data reading by the LDC syndrome generation circuit 13, and the first to 27th reading blocks RB1 to RB27. Read in the column direction.

Next, the descrambling process in the descrambling circuit 20 will be described with reference to FIGS. 7 and 10 to 12.
In step S1 of FIG. 10, the initial value S0 of the scramble value SC is stored in the even register R2, and S0 · γ ¹⁰⁸ is stored as the scramble value SC in the odd register R3. Here, the reason why not the initial value S0 but S0 · γ ¹⁰⁸ is stored in the odd register R3 will be described. As shown in FIG. 7, since the row number u of the scrambled data sDu read first in the odd data frame DF (for example, the data frame DF1) in the first read block RB1 is 108, S0 · γ ¹⁰⁸ is stored. I did it. That is, the scrambled data sD108 is data that is processed by 108 bytes later in the scramble process than the start data sD0 in the scramble process of each data frame. Therefore, S0 · γ108 obtained by shifting by 108 bytes from the initial value S0 is stored in the odd register R3 as the scramble value SC. The initial value S0 is a value set in advance according to the physical sector number in the ECC cluster ECCc in the disk data.

  Next, the descrambling circuit 20 reads the scrambled data sD0 in the data frame DF0 (even data frame) of the first read block RB1, and inputs the scrambled data sD0 to the EOR circuit 21 (step S2). At this time, the selection circuit 28 outputs to the selector 27 a selection signal for causing the EOR circuit 21 to output the scramble value SC from the even register R2. Accordingly, the initial value S0 is input to the EOR circuit 21 as the scramble value SC. The EOR circuit 21 generates the descrambling data dD0 by taking the exclusive OR of the scrambled data sD0 and the lower 8 bits of the initial value S0, and supplies the descrambling data dD0 to the external buffer memory 6 and the EDC syndrome generation circuit 30. Output (step S3). The scramble value SC used here is also input to the 1-shift calculator 24 and the 209-shift calculator 25. In the following description, the lower 8 bits of the scramble value SC in the exclusive OR operation of the EOR circuit 21 are simply referred to as “scramble value SC”. An exclusive OR with the lower 8 bits of the scramble value SC is taken.

  Next, in step S4, the selection circuit 28 determines whether the descrambling process for one column in each read block RB is completed based on the number of pulses of the input clock signal CLK. If selection circuit 28 determines that the descrambling process for one column has not been completed (NO in step S4), it proceeds to step S5. At this time, if the scramble data read by the descrambling circuit 20 in step S3 is sDu, the next read scramble data is sDu + 1. That is, when moving from step S4 to step S5, the scrambled data is read in the column direction in each read block RB, and the read order is the same as the data processing order during the scramble process.

Therefore, the selection circuit 28 causes the EOR circuit 21 to output the scramble value SC from the 1 shift computing unit 24 in step S5, that is, when the data of the second and subsequent rows of each read block are input to the EOR circuit 21. A selection signal for output is output to the selector 27. As a result, the EOR circuit 21 shifts the scramble value SC (=) obtained by shifting the scramble value used at the time of the previously read scramble data sDu by 1 byte according to the generator polynomial Φ (x). S0 · γ ^u · γ ¹ ) is input. Then, the EOR circuit 21 obtains an exclusive OR of the scrambled data sD1 read next by the descrambling circuit 20 and the scramble value SC (= S0 · γ ¹ ) from the 1 shift computing unit 24, and descrambles Data dD1 is generated (step S3). Thus, steps S3 to S5 are repeatedly executed until the descrambling process for one column in each read block RB is completed. In this embodiment, the descrambling circuit 20 reads the scrambled data in the order of sD2->sD3->sD4->sD5->sD6-> sD7 (see the solid line arrow in FIG. 7). Is generated as shown in the following equation.

When the descrambling data dD7 is generated in this way, in step S4, the selection circuit 28 determines that the descrambling process for one column in each read block RB is completed, and proceeds to step S6. Next, the selection circuit 28 determines whether the descrambling process for one data frame in each read block RB is completed (step S6). If the selection circuit 28 determines that the descrambling process for one data frame has not been completed (NO in step S6), the selection circuit 28 proceeds to step S7. At this time, if the scramble data read by the descrambling circuit 20 in step S3 is sDu, the next read scramble data is sDu + 209. That is, when moving from step S6 to step S7, the scrambled data in the first row of the next column is read in the same data frame of each read block RB, and the read order is different from the data processing order during the scramble process. It becomes like this. Specifically, the data D2 in the first row of the next column read when moving to step S7 is {[(the number of rows in the data block DB] at the time of the scramble processing than the data D1 read out immediately before. ) +1] − (number of rows in one read block RB) = 209} data that is processed later by bytes. Therefore, the descrambling process is accurately performed by shifting the scramble value used at the time of the data D1 by 209 bytes according to the generator polynomial Φ (x) so that the scramble value SC matches the processing order at the time of the scramble process. be able to.

Therefore, the selection circuit 28 determines the scramble value from the 209 shift computing unit 25 in step S7, that is, when the first row data after the second column in each data frame of each read block is input to the EOR circuit 21. A selection signal for outputting the SC to the EOR circuit 21 is output to the selector 27. Thus, the EOR circuit 21, scrambling scrambled value S0 · gamma ^7, which is used when the scramble data sD7 read in the previous, obtained by 209 bytes shifted in accordance with the above generator polynomial [Phi (x) The value SC (= S0 · γ ⁷ · γ ²⁰⁹ ) is input. Subsequently, the EOR circuit 21 obtains an exclusive OR of the scramble data sD216 read next by the descramble circuit 20 and the scramble value SC (= S0 × γ ²¹⁶ ) from the 209 shift computing unit 25, Scramble data dD216 is generated (step S3).

  Next, similarly to the processing of the first column in the data frame DF0 described above, steps S3 to S5 are repeatedly executed, and the scramble data is scrambled by the descrambling circuit 20 in the order of sD217 → sD218 → sD219 → sD220 → sD221 → sD222 → sD223. Since it is read, descrambling data is generated as shown in the following equation.

Thereafter, similarly, Steps S3 to S7 are repeatedly executed, and the descrambling process for each scrambled data in the data frame DF0 of the first read block RB1 is executed. When descrambling data dD1951 (= sD1951 ^ S0 · γ ¹⁹⁵¹ ) for the scrambled data sD1951 in the last row and last column in the data frame DF0 of the first read block RB1 is generated, the selection circuit 28 outputs the data for one data frame. Is determined to have been completed (YES in step S7), and the process proceeds to step S8.

  Next, the selection circuit 28 determines whether the data frame DF that has been descrambled is the final data frame, that is, the data frame DF31 (step S8). When the selection circuit 28 determines that the data frame for which the descrambling process has been completed is not the data frame DF31 (NO in step S8), the selection circuit 28 determines whether the data frame DF to be read next is an even data frame (step S9). . If the selection circuit 28 determines that the next data frame DF is an even data frame (YES in step S9), the selection circuit 28 proceeds to step S10. If the selection circuit 28 determines that the next data frame DF is an odd data frame (NO in step S9), the selection circuit 28 proceeds to step S11. Here, since the next data frame is the data frame DF1 (odd data frame), the process proceeds to step S11.

Subsequently, since the first data in the first row and the first column in each odd-numbered data frame of each read block RB is read next, the selection circuit 28 scrambles the SC from the odd-numbered register R3 (here, S0 · γ ¹⁰⁸ ). Is generated by the EOR circuit 21 (step S11).

Next, in step S12 shown in FIG. 11, the selection circuit 28 determines whether the next data frame DF is the last odd data frame, that is, the data frame DF31. If selection circuit 28 determines that next data frame DF is not final odd data frame DF31 (NO in step S12), it returns to step S3. In step S3 at this time, since the selection signal generated in step 11 is input to the selector 27, S0 · γ ¹⁰⁸ stored in the odd register R3 is input to the EOR circuit 21 as the scramble value SC. The Further, the EOR circuit 21 receives the scrambled data sD108 in the first row and the first column in the data frame DF1 of the first read block RB1. The EOR circuit 21 generates the descrambling data dD108 (= sD108 ^ S0 · γ ¹⁰⁸ ) by taking the exclusive OR of the scrambled data sD108 and the scramble value S0 · γ ¹⁰⁸ of the data frame DF1.

Thereafter, as in the case of the data frame DF0, when data from the second row onward in each column of each read block RB (for example, scrambled data sD109) is input to the EOR circuit 21, the previous scramble value is set to 1 byte. The descramble process is performed using the shifted scramble value SC. When data in the first row of each column after the second column in each data frame of each read block RB (for example, scrambled data sD324) is input to the EOR circuit 21, the previous scramble value SC is 209 bytes. The descramble processing is performed using the shifted scramble value. Then, Steps S3 to S7 shown in FIG. 10 are repeatedly executed, and the descrambling data dD1843 (= sD1843 ^ S0 · γ ¹⁸⁴³ ) for the last scrambled data sD1843 (see FIG. 7) in the data frame DF1 of the first read block RB1. ) Is generated, the process proceeds from step S6 to step S8, and then to step S9. At this time, since the next data frame is the data frame DF2 (even data frame) (YES in step S10), the process proceeds to step S10.

  Subsequently, since the first data in the first row and the first column in each even data frame of each read block RB is read out next, the selection circuit 28 uses the scramble value SC (here, S0) from the even register R2 as an EOR circuit. A selection signal to be output to 21 is generated (step S10). Next, in step S13 shown in FIG. 11, the selection circuit 28 determines whether the next data frame is the last even data frame, that is, the data frame DF30. If selection circuit 28 determines that the next data frame is not the last even data frame (NO in step S13), it returns to step S3. Thereafter, the descrambling process is performed for the data frames DF2 to DF29 in the same manner as the process for the data frame DF0 and the data frame DF1. In the same read block RB, the order of the scramble values SC used in each of the even data frames (data frames DF0, DF2,..., DF30) is the same. In the same read block RB, the order of the scramble values SC used in each of the odd data frames (data frames DF1, DF3,..., DF31) is the same.

Next, when the descrambling process of the final even data frame DF30 is started (YES in step S13), the process proceeds to step S14. In step S14, steps S3 to S5 are repeatedly executed, and when the descrambling process of the first column in the data frame DF30 is completed, the scramble value SC used for the last data sD7 of the first column is calculated by one shift operation. It is stored in the even register R2 through the device 24. Therefore, in the first read block RB1, the scramble value SC (= S0 · γ ⁷ · γ ¹ = S0 · γ ⁸ ) is stored in the even-numbered register R2. Note that when the second read block RB2 is scramble value SC (= S0 · ^{S 16)} is stored in the even register R2. After step S14, the selection circuit 28 generates a selection signal for causing the EOR circuit 21 to output the scramble value SC (in this case, S0 · γ ²¹⁶ ) from the 209 shift computing unit 25 (step S16). Return to S3. Thereafter, the descrambling process of the second and subsequent columns of the data frame DF31 is executed in the same manner as the second and subsequent columns of the data frame DF0.

Next, when the descrambling process of the last odd data frame DF31 is started (YES in step S12), the process proceeds to step S15. In step S15, steps S3 to S5 are repeatedly executed, and when the descrambling process for the first column in the data frame DF31 is completed, the scramble value SC used for the last data sD115 in the first column is calculated by one shift operation. It is stored in the odd register R3 through the device 24. Accordingly, in the first read block RB1, the scramble value SC (= S0 · γ ¹¹⁵ · γ ¹ = S0 · γ ¹¹⁶ ) is stored in the odd-numbered register R3. In the case of the second read block RB2, a scramble value SC (= S0 × S ¹²⁴ ) is stored in the odd register R3. The descrambling process for the second and subsequent columns of the data frame DF31 is executed in the same manner as for the data frame DF1. When descrambling data dD1843 (= sD1843 ^ S0 · γ ¹⁸⁴³ ) for scrambled data sD1843 in the last row and last column of data frame DF31 is generated (YES in step S8), first read block RB1 (rth read block) ) Is descrambled.

When the descrambling process of the first read block RB1 is completed, the descrambling process of the second read block RB2 is started. Specifically, the processing is started from step S2 shown in FIG. 10, and the selection circuit 28 gives the selector 27 a selection signal for causing the EOR circuit 21 to output the scramble value SC (here, S0 · γ ⁸ ) from the even-numbered register R2. Output. That is, the selection circuit 28 causes the EOR circuit 21 to output the scramble value SC from the even register R2 when the first data in the first row and the first column in each even data frame of each read block is input to the EOR circuit 21. A selection signal for output is output to the selector 27. As a result, the EOR circuit 21 shifts the scramble value used for the data sD7 in the last row of the first column of the last even data frame DF30 in the previous first read block RB1 by 1 byte according to the generator polynomial Φ (x). The scramble value SC (= S0 · γ ⁸ ) obtained as described above is input. Thus, in step S3, the scramble value S0 · gamma ⁸ from even register R2, the exclusive OR of the first data sD8 in the first row and the first column of the data frame DF0 the second read block RB2 take the EOR circuit 21 Thus, descrambling data dD8 (= sD8 ^ S0 · γ ⁸ ) is generated. Incidentally, the odd case of the data frame also, S0 · gamma ¹¹⁶ as the scramble value in the odd register ^R3, that is, the scramble value is stored for the leading data sD116 in the first row and the first column in an odd-numbered data frames of the second read block RB2 Therefore, the descrambling process is also accurately performed on the top data sD116.

Thereafter, similarly, the descrambling process in the second to thirteenth read blocks RB2 to RB13 is executed according to the flowcharts of FIGS.
Next, the descrambling process in the fourteenth read block RB14 will be described. The data frames DF0 and DF1 shown in FIG. 7 will be described as an example. In the first to thirteenth read blocks RB1 to RB13, the 1st to 10th columns become the data frame DF0, and the 11th to 19th columns become the data frame DF1. Become. In the 15th to 27th read blocks RB15 to RB27, the 1st to 9th columns become the data frame DF0, and the 10th to 19th columns become the data frame DF0. However, in the fourteenth read block RB14, the tenth column becomes the last column of the data frame DF0 (even data frame) and the first column of the data frame DF1 (odd data frame). Further, the data frame DF1 is the first data sD0 in the scramble process in the data frame DF1 from the fifth row of the tenth column. At this time, the scramble value SC stored in the odd register R3 during the descrambling process of the last odd data frame in the thirteenth read block RB13 is S0 · γ ²¹² . Therefore, when the first data sD0 of the data frame DF1 of the fourteenth read block RB14 is input to the EOR circuit 21, the scramble value S0 · γ ²¹² is input to the EOR circuit 21 from the odd register R3 according to the flowcharts of FIGS. If this happens, the descrambling process cannot be performed accurately.

  Therefore, in the present embodiment, the descrambling process of the last column of each even data frame and the first column of each odd data frame in the fourteenth read block RB14 is executed according to the flowchart of FIG.

In step S20 shown in FIG. 12, the descrambling for the last data sD1839 in the column (here, the ninth column) one column before the last column in the even-numbered data frame (here, the data frame DF0) of the fourteenth read block RB14. When data dD1839 (= sD1839 ^ S0 · γ ¹⁸³⁹ ) is generated, the process proceeds to step S21. In step S 21, the selection circuit 28 outputs to the selector 27 a selection signal that causes the FF circuit 23 to output the scramble value SC (= S 0 · γ ¹⁸³⁹ · γ ²⁰⁹ = S 0 · γ ²⁰⁴⁸ ) from the 209 shift computing unit 25.

Next, the EOR circuit 21 generates the scrambled data (here, scrambled data sD2048) to be read next by the descrambling circuit 20 and the scramble value SC (here, S0 · γ ²⁰⁴⁸ ) from the 209 shift computing unit 25. The exclusive OR is taken to generate descrambling data dD2048 (step S22).

  Next, in step S23, the selection circuit 28 determines whether the descrambling process of the last column in the even-numbered data frame of the fourteenth read block RB14 is completed. If the selection circuit 28 determines that the descrambling process of the even data frame has not been completed (NO in step S23), the selection circuit 28 proceeds to step S24. Subsequently, steps S22 to S24 are repeatedly executed in the same manner as steps S3 to S5 shown in FIG. 10 until the descrambling process of the even data frame is completed. Then, as shown in the following equation, when descrambling data for the final data sD2051 in each even data frame of the fourteenth read block RB14 is generated (YES in step S23), the process proceeds to step S25.

Next, the selection circuit 28 outputs to the selector 27 a selection signal for causing the EOR circuit 21 to output the scramble value SC from the initial value register R1, that is, the initial value S0 (step S25). Subsequently, the EOR circuit 21 obtains an exclusive OR of the head data sD0 at the time of the scramble processing in each odd data frame read out by the descrambling circuit 20 and the initial value S0 from the initial value register R1, and performs descrambling. Data dD0 (= sD0 ^ S0) is generated (step S26).

  Next, in step S27, the selection circuit 28 determines whether the descrambling process for the first column in each odd data frame of the fourteenth read block RB14 is completed. If the selection circuit 28 determines that the descrambling process for the first column of the odd data frame has not been completed (NO in step S27), the selection circuit 28 proceeds to step S28. Subsequently, steps S26 to S28 are repeatedly executed in the same manner as steps S3 to S5 shown in FIG. 10 until the descrambling process of the first column in each odd data frame of the fourteenth read block RB14 is completed. Note that the descrambling process is performed on the second and subsequent columns in each odd data frame of the fourteenth read block RB14 by repeatedly executing steps S3 to S7 shown in FIG.

When descrambling data dD4 is generated for scrambled data sD4 in the last row of the first column in each odd-numbered data frame of 14th read block RB14 (YES in step S27), the process proceeds to step S29. In this step S29, if the current data frame DF is the last odd data frame (data frame DF31), the scramble value SC used for the scrambled data sD4 is stored in the odd register R3 through the 1 shift calculator 24. That is, the scramble value SC stored in the odd register R3 during the descrambling process of the last odd data frame in the fourteenth read block RB14 is S0 · γ ⁴ · γ ¹ = S0 · γ ⁵ .

  Thereafter, the descrambling process in the 15th to 27th read blocks RB15 to RB27 is executed according to the flowcharts of FIGS. As described above, when the descrambling process in the first to twenty-seventh read blocks RB1 to RB27 is completed according to the flowcharts of FIGS. 10 to 12, descrambling data dDu can be generated as shown in the following equation.

Next, an EDC syndrome generation method in the EDC syndrome generation circuit 30 will be described with reference to FIGS. The descrambling data dDu is input from the descrambling circuit 20 to the EDC syndrome generation circuit 30 in the descrambling process order. That is, descrambling data dDu is input to the EDC syndrome generation circuit 30 for each of the read blocks RB1 to RB27. Specifically, in the even-numbered data frame of the first read block RB1, the descrambling data is generated as byte data in the order of dD0 → dD1 → ... → dD7 → dD216 → dD217 → ... → dD223 → dD432 → ... → dD1951. Input to the circuit 30. In the odd data frame of the first read block RB1, the descrambling data is input as byte data to the EDC syndrome generation circuit 30 in the order of dD108 → dD109 →... DD115 → dD324 → dD325 →... → dD331 → dD540 →. Is done. FIG. 13 shows the arrangement of the byte data input to the EDC syndrome generation circuit 30 by the order n when the byte data is represented by a polynomial expression ^Xn . In the following description, byte data is expressed by a polynomial. In the same read block RB, even-numbered data frame calculation methods are common to each other, and the data frame DF0 will be described as an example here. In the same read block RB, the odd-numbered data frame calculation method is common to each other, and the data frame DF1 will be described as an example here.

In the initial value setting of step S30 shown in FIG.
Subsequently, in step S31, byte data X ⁿ (here, X ²⁰⁵¹ ) is input from the descramble circuit 20 to the EOR circuit 31. At this time, since the shift operation value Es is not input from the selector 32 to the EOR circuit 31, the EOR circuit 31 outputs the byte data X ²⁰⁵¹ as the EDC operation value E1 (1) to the various arithmetic units 34 to 37 through the FF circuit 33. .

Next, in step S32, the selection circuit 39 determines whether the EDC calculation processing for one column (for example, X ^{2051 to} X ²⁰⁴⁴ ) in each read block RB is completed based on the number of pulses of the input clock signal CLK. Judging. When selecting circuit 39 determines that the EDC operation processing for one column has not been completed (NO in step S32), the process proceeds to step 33. At this time, if the byte data input to the EOR circuit 31 in step S31 is ^Xn , the next byte data input is ^Xn-1 . That is, when moving from step S32 to step S33, the next byte data is input in the column direction in each read block RB, and the data input order is the same as the data processing order when EDC is added.

Therefore, the selection circuit 39 EORs the scramble value SC from the 1-shift arithmetic unit 24 in step S33, that is, when the byte data from the second row onward in each column in each read block RB is input to the EOR circuit 31. A selection signal for outputting to the circuit 31 is output to the selector 32. Thereby, the EOR circuit 31 is obtained by weighting the EDC operation value E1 (1) calculated for the byte data X ²⁰⁵¹ input immediately before by the primary vector β ^1. The shift operation value Es (= β ¹ · E1 (1)) is input. Then, the EOR circuit 31 performs exclusive OR operation between the byte data X ²⁰⁵⁰ inputted next and the lower 8 bits of the shift operation value Es (= β ¹ · E1 (1)) from the 1-shift operation unit 34. Then, a new EDC calculation value E1 (2) is calculated. In the following description, the lower 8 bits of the shift operation value Es in the exclusive OR operation of the EOR circuit 31 are simply referred to as “shift operation value Es”. The exclusive OR of ^Xn and the lower 8 bits of the shift operation value Es is taken.

Subsequently, these steps S31 to S33 are repeatedly executed until the EDC operation processing for one column is completed. In this embodiment, since the byte data is input to the EOR circuit 31 in the order of X ²⁰⁴⁹ → X ²⁰⁴⁸ →... → X ²⁰⁴⁵ → X ²⁰⁴⁴ , the EDC operation value Er (c) is in the order shown in the following formula. Calculated.

As described above, when the EDC operation processing for the byte data X ^{2044 in} the last row of the first column in the data frame DF0 of the first read block RB1 is completed (YES in step S32), the process proceeds to step S34. Next, the selection circuit 39 generates a selection signal for causing the EOR circuit 31 to output the shift calculation value Es (= ^β209 · E1 (8)) from the 209 shift calculator 35 (step S34). Subsequently, in step S35, the selection circuit 39 determines whether or not the EDC calculation process for one data frame DF in each read block RB1 is completed. If the selection circuit 39 determines that the EDC calculation processing for one data frame DF has not been completed (NO in step S35), the selection circuit 39 proceeds to step S36. Subsequently, the selection circuit 39 determines whether the next column is the last column of each data frame (for example, in the case of the data frame DF0 of the first read block RB1, the next column is the tenth column). If selection circuit 39 determines that the next column is not the last column (NO in step S36), it returns to step S31. Subsequently, the EOR circuit 31 calculates the exclusive OR of the byte data X ¹⁸³⁵ inputted next and the shift operation value Es (= β ²⁰⁹ · E1 (8)) from the 209 shift operation unit 35, An EDC operation value Er (9) shown in the following equation is calculated.

As described above, when the byte data ^Xn of the first row of each column excluding the first column and the last column in each data frame DF of each read block is input, the byte data ^{Xn + 209} input before that The calculated EDC operation value Er (c) is weighted by the 209th-order vector β ²⁰⁹ through the 209 shift operation unit 35. Here, the reason why the 209th-order vector β ²⁰⁹ is weighted will be described. For example, the data X ¹⁸³⁵ in the first row in the second column of the data frame DF0 of the first read block RB1 is greater than the data X ^{2044 in} the last row in the first column when EDC is added (all bytes in the data block DB). Number) + 1− (number of bytes in the column direction of one read block) = data that is processed later by 209 bytes. That is, the byte data X ^{2051 to} X ²⁰⁴⁴ are weighted 209 times by the primary vector β ¹ until the byte data X ¹⁸³⁵ is input when the EDC operation is performed in the same data processing order as when EDC is added. Done. Therefore, in the present embodiment, descrambling is performed by weighting the EDC operation value Er (c) calculated for the last row data of the previous column with the vector β (vector β ²⁰⁹ ) for 209 bytes. It was made to process.

Thereafter, similarly, Steps S31 to S36 are repeatedly executed, and when the EDC arithmetic processing for the last byte data X ³¹⁶ (the 72nd clock) of the ninth column in the data frame DF0 of the first read block RB1 is completed, the next column is final. Since it becomes a column (10th column) (YES in step S36), the process moves to step S37. Subsequently, the selection circuit 39 outputs to the selector 32 a selection signal for causing the EOR circuit 31 to output the shift operation value Es (= β ²⁰⁹ · Er (c) ^ EDCs) from the EOR circuit 38. However, since the intermediate value EDCs of the EDC syndrome input from the EDC memory unit 40 to the EOR circuit 38 is “0” here, the shift operation value Es from the EOR circuit 38 is β ²⁰⁹ · E1 (72). Therefore, in step S31, the EOR circuit 31 calculates the exclusive OR of the byte data X ¹⁰⁷ inputted next and the shift operation value Es (= β ²⁰⁹ · E1 (72)) from the EOR circuit 38. The EDC calculation value E1 (73) shown in the following equation is calculated.

Hereinafter, steps S31~S33 are repeated, as shown in the following formula, EDC calculated value for the byte data ^X 100 of the last row of the last column of the data frame DF0 the first read block RB1 (80 th clock) E1 (80 ) Is calculated (YES in step S35), the process proceeds to step S38.

Subsequently, in step S38, the selection circuit 39 determines whether the current data frame is an even data frame of the 14th read block RB14, and determines whether it is an even data frame of the 27th read block RB27 in step S39. In step S40, it is determined whether it is an odd data frame of the 27th read block RB27. Here, since the data frame DF currently undergoing EDC calculation processing is the data frame DF0 (even data frame) of the first read block RB1 (NO in steps S38 to S40), the process proceeds to step S41. Next, the EDC operation value Er (c) (here, E1 (80)) calculated in step S34 is weighted by the vector β1 through the 1-shift operation unit 34, and the intermediate value EDCs of the EDC syndrome is obtained. (= Β ¹ · E1 (80)) is calculated (step S41). Subsequently, the calculated intermediate value EDCs is stored in the EDC memory unit 40 as an intermediate result of the EDC syndrome in the data frame DF0 (step S42).

  Next, in step S43, the selection circuit 39 determines whether the data frame DF currently undergoing EDC calculation processing is the final data frame DF31 of the final read block RB27. Here, since it is the data frame DF0 of the first read block RB1 that is currently being processed, the process returns to step S30 and moves to the EDC calculation process of the next data frame DF1.

The EDC calculation process for the data frame DF1 which is an odd data frame is basically performed in the same manner as the EDC calculation process for the data frame DF0 (even data frame) described above. That is, first, the byte data X ¹⁹⁴³ input at the beginning of the data frame DF1 in each read block RB becomes the EDC operation value E1 (1) of the first clock. Next, when the byte data ^{Xn of the} second and subsequent rows of each column in each data frame of each read block RB is input to the EOR circuit 31, the EOR circuit 31 has the byte data ^{Xn + 1} input immediately before. A shift operation value Es obtained by weighting the calculated EDC operation value Er (c) with the primary vector β ¹ is input to the EOR circuit 31. Further, when the byte data X ⁿ of the first row of each column excluding the first column and the last column in each data frame DF of each read block RB is input to the EOR circuit 31, the byte data X ^{n + 209} input immediately before is input ^. A shift operation value Es obtained by weighting the EDC operation value Er (c) calculated at this time by the 209th-order vector β ²⁰⁹ is input to the EOR circuit 31. By these EDC calculation processes, EDC calculation values Er (c) shown in the following formulas are calculated in order.

Thus, when the EDC operation value E1 (72) for the byte data X ²⁰⁸ (72nd clock) of the last column in the data frame DF1 of the first read block RB1 is calculated, the EDC operation value E1 (72 relative), weighting throughout the shift calculator 34 by a vector beta ¹ is performed, the intermediate value EDCs is generated (step S41). Then, the generated intermediate value EDCs (= β ¹ · E1 (72)) is stored in the EDC memory unit 40 as an intermediate result of the EDC syndrome in the data frame DF1 (step S42).

Thereafter, similarly, Steps S30 to S43 are repeatedly executed, and the EDC calculation processing of the data frames DF2 to DF31 in the first read block RB1 is executed.
Next, an EDC calculation process in the data frame DF0 of the second read block RB2 will be described. From the first clock to the 72nd clock, steps S30 to S36 are repeatedly executed, and EDC operation values E2 (1) to E2 (72) shown in the following equations are calculated.

When the EDC arithmetic processing for the last byte data X ^{308 in} the last column, that is, the ninth column preceding the tenth column is completed, in step S37, the selection circuit 39 outputs the shift operation value Es (= A selection signal for causing the EOR circuit 31 to output β ²⁰⁹ · E2 (72)） EDCs) is output to the selector 32. At this time, the intermediate value EDCs input from the EDC memory unit 40 to the EOR circuit 38 is a value (= β ¹⁾ stored in the EDC memory unit 40 during the EDC operation processing of the same data frame DF0 in the previous first read block RB1. E1 (80)). Accordingly, the 73rd clock EDC operation value E2 (73) in the data frame DF0 of the second read block RB2 is calculated by the following equation.

Thereafter, steps S31 to S33 are repeatedly executed, and the EDC operation value E2 (80) of the byte data X ⁹² (80th clock) of the last row and last column in the data frame DF0 of the second read block RB2 as shown in the following equation. Is calculated, the process proceeds to step S41.

Next, in step S41, with respect to the calculated EDC calculated value E2 (80), weighting by the vector beta ¹ through 1 shift calculator 34 is performed, the intermediate value EDCs are calculated. Then, the calculated intermediate value EDCs (= β ¹ · E2 (80)) is stored in the EDC memory unit 40 as a new intermediate result of the data frame DF0 (step S42).

  Similarly, steps S30 to S43 are repeatedly executed, and the EDC arithmetic processing of the data frames DF0 to DF31 in the second to thirteenth read blocks RB2 to RB13 is executed.

Then, in the even data frame of the 14 read block RB14, 76 when the EDC calculated value E14 for byte data X ⁰ the lowest orders of the input polynomial input to th clock (76) is calculated as the following equation (YES in step S38), the process proceeds to step S44.

Note here, the even data frames DF0, the lowest byte data X ⁰ degrees is input polynomial, when the EDC operation in the same data processing order and time of EDC addition, slowest input within said data frames DF0 Data.

Next, in step S44, a first correction operation for weighting the calculated EDC operation value E14 (76) by the vector β (vector β ⁻²¹¹ ) for −211 bytes through the −211 shift operation unit 37 is performed. The intermediate value EDCs (= β− ²¹¹ · E14 (76)) of the EDC syndrome is calculated. Subsequently, the calculated intermediate value EDCs is stored in the EDC memory unit 40 as a new intermediate result of the new data frame DF0 (step S42).

  Similarly, steps S30 to S44 are repeatedly executed, and the EDC calculation processing of the data frames DF0 to DF31 in the fourteenth to twenty-sixth read blocks RB14 to RB26 is executed.

Next, an EDC operation value E27 (72) for the byte data X ¹⁰⁸ (72nd clock) in the last row and last column in the even data frame (for example, data frame DF0) of the 27th read block RB27 is calculated as follows: Then (YES in step S39), the process proceeds to step S45.

Next, in step S45, a second correction calculation is performed on the calculated EDC calculation value E27 (72) by weighting the vector β ¹ (vector β ¹⁰⁸ ) for 108 bytes through the 108 shift calculator 36. The final EDC syndrome EDCS (= β ¹⁰⁸ · E27 (72)) in the data frame DF0 expressed by the following equation (1) is calculated.

Then, the calculated EDC syndrome EDCS is stored in the EDC memory unit 40 (step S42).

Here, the first correction calculation and the second correction calculation will be described. First, the second correction calculation will be described.
In this embodiment, the byte data input to the last in the even data frame of the 27 read block RB27 is X ^108. The byte data X ¹⁰⁸ is input by 108 bytes earlier than the byte data X ⁰ (final byte data) input last in the calculation processing when the EDC calculation is performed in the same data processing order as when EDC is added. Data. In other words, the byte data X ¹⁰⁸ is weighted 108 times by the primary vector β ¹ until the last byte data X ⁰ is input when the EDC operation is performed in the same data processing order as when EDC is added. Data to be performed. However, in the present embodiment, since the byte data ^{X 108} in the even data frame is input to the end, the EDC calculation value E27 is calculated (72) when the byte data ^{X 108} is input, the byte data ^{X 108} weighting by the vector β ¹ for has not been performed even once. Therefore, in the present embodiment, the EDC calculation value when the data ^{X 108} of the last row last column is entered in each even data frame of the 27 read block RB27 E26 (72), of 108 bytes vector beta (Vector Weighting by β ¹⁰⁸ ) was performed.

Next, the first correction calculation will be described. In this embodiment, the 14 read block EDC calculated value last byte data X ⁰ is calculated when it is entered in the even data frame RB14 E14 (76) is as shown in the following equation.

As is clear from the above equation, the byte data input until the EDC operation value E14 (76) is calculated is the vector β having a desired order when the EDC operation value E14 (76) is calculated. Is weighted. For example, the byte data X ²⁰⁵¹ input first in the even data frame is weighted by the 2051-order vector β ²⁰⁵¹ . However, in the present embodiment, the EDC operation in the fourteenth read block RB14 fifteenth to 27 read block RB15~RB27 later, to the EDC calculated value E14 (76), weighting by the primary vector beta ¹ of 103 Performed once. That is, the relative EDC calculated value E14 (76), the weighting by {(remaining read block total number of bytes in the column direction in RB) -1} times Vector beta ¹ is performed. Further, the EDC operation value E14 (76) is weighted by the ^108th vector β ¹⁰⁸ by the second correction operation described above. Therefore, when the EDC operation and the second correction operation in the 15 to 27 read block RB15~RB27 ends, to the byte data input to to the EDC calculated value E14 (76) is calculated, according to the vector beta ¹ Weighting is performed 211 times more than when EDC calculation is performed in the same processing order as when EDC is added. Therefore, in the present embodiment, −211 bytes worth of the EDC operation value E14 (76) when the last data X ⁰ (final descrambling data) in each even data frame of the fourteenth read block RB14 is input. Weighting with vector β (vector β- ²¹¹ ) was performed.

For odd data frames, special correction calculations (first and second correction calculations) such as step S44 and step S45 are performed in the same manner as the EDC calculation process of the data frame DF1 in the first read block RB1 described above. Without application, an EDC syndrome similar to the above equation (1) can be generated. Therefore, in the case of the data frame DF1 (odd data frames), the EDC calculated value of the byte data ^{X 0} of the last row the last row in the 27 read block RB27 E27 (80) is calculated (YES in step S40), The calculated EDC operation value E27 (80) is stored as it is in the EDC memory unit 40 as an EDC syndrome (step S42).

  Similarly, steps S30 to S45 are repeatedly executed to generate a final EDC syndrome EDCS of each data frame DF0 to DF31, and the EDC syndrome EDCS is stored in the EDC memory unit 40. In the EDC memory unit 40, when the final EDC syndrome EDCS of all the data frames DF0 to DF31 is stored in the buffer memory selected for syndrome storage, the syndrome storage and syndrome output in the buffer memory M7 and the buffer memory M8 are performed. The selection for is switched. As a result, the EDC syndrome EDCS of all the data frames DF0 to DF31 is stored in the buffer memory switched from the syndrome storage to the syndrome output. The buffer memory switched to the syndrome output stores the EDC syndrome EDCS of all the data frames DF0 to DF31 in the external buffer memory 6.

  When the descrambling data, the LDC syndrome, the BIS syndrome, and the EDC syndrome for one parity added block PB are stored in the external buffer memory 6, the error correction circuit 50 receives the LDC syndrome and BIS from the external buffer memory 6. Read out the syndrome. The correction circuit 51 in the error correction circuit 50 calculates an error position polynomial and a numerical polynomial by Euclidean mutual division based on the read LDC syndrome and BIS syndrome. Next, the correction circuit 51 calculates an error position and an error value based on the calculated error position polynomial and numerical polynomial. Subsequently, the correction circuit 51 performs error correction processing on the descrambled data frame DF stored in the external buffer memory 6 based on the calculated error position and error value, and calculates the calculated error position and The error numerical value is output to the EDC correction circuit 52.

  Next, the EDC correction circuit 52 corrects the EDC syndrome stored in the correction memory unit 53 from the external buffer memory 6 based on the error position and error value input from the correction circuit 51. Then, the EDC check circuit 54 reads the corrected EDC syndrome from the correction memory unit 53 and checks whether the error correction has been normally completed. When it is confirmed in the EDC check circuit 54 that the error correction has been normally completed, the controller 10 converts the error-corrected data stored in the external buffer memory 6 into the interface circuit based on the instruction of the microprocessor 8. The data is output to the computer 2 via 60.

According to this embodiment described above, the following effects can be obtained.
(1) The built-in memory unit 12 is accessed by the LDC syndrome generation circuit 13, the BIS syndrome generation circuit 16, and the descrambling circuit 20, respectively. The circuits 13, 16, and 20 each generate the generated LDC syndrome, BIS syndrome, and data. The scrambled data dDu is stored in the external buffer memory 6. Thereby, the access amount to the external buffer memory 6 can be reduced as compared with the controller 110 shown in FIG.

  Specifically, in the controller 110 shown in FIG. 41, (a) input of the ECC cluster ECCc from the demodulation circuit 111, (b) reading of the LDC cluster LDCc by the LDC syndrome generation circuit 112, (c) There are five types of access: reading of the BIS cluster BISc by the BIS syndrome generation circuit 113, (d) error correction processing from the error correction circuit, and (e) reading of the LDC cluster LDCc after error correction processing by the descramble circuit 115. It was. That is, there was an access to the buffer memory 120 having a data amount (total 219299 bytes) substantially corresponding to three ECC clusters ECCc (76880 bytes).

  In contrast, in the controller 10 of the present embodiment, (A) input of descrambling data dDu (65536 bytes), (B) input of LDC syndrome (9728 bytes), (C) BIS syndrome (1488 bytes) input, (D) EDC syndrome (128 bytes) input, (E) LDC syndrome read, (F) BIS syndrome read, (G) Descramble data read after error correction, There are seven types of access. Although there are many types of access, the amount of data in each access is small. In other words, as a whole, the external buffer memory 6 is accessed for a data amount (total 152272 bytes) substantially corresponding to two ECC clusters ECCc. Therefore, the total access amount to the external buffer memory 6 is greatly reduced as compared with the conventional controller 110. As a result, the use bandwidth of the external buffer memory 6 can be reduced, so that it becomes easy to cope with an increase in transfer speed. As a result, even if an inexpensive, large-capacity, low-speed memory is used as the external buffer memory 6, data transfer at a desired speed can be realized.

  Further, at this time, the LDC syndrome generation circuit 13 and the descrambling circuit 20 have a performance capable of high-speed access to the built-in memory (built-in memory unit 12), so that the first and second deinterleaving processes are performed. Even when the LDC cluster is read, each data can be read from the built-in memory unit 12 efficiently and at high speed. As a result, the time required for error correction processing can be shortened compared to the controller 110.

  (2) The built-in memory unit 12 stores a divided ECC cluster DECC obtained by dividing one ECC cluster ECCc into a plurality of pieces. The descrambling circuit 20 is provided with a 1-shift computing unit 24 and a 209-shift computing unit 25 so that the descrambling process can be accurately performed on the data input for each data block DB in the divided ECC cluster DECC. I did it. As a result, the buffer memories M1 and M2 of the built-in memory unit 12 can have a memory capacity capable of storing one divided ECC cluster DECC (2480 bytes). Therefore, unlike the controller 130 shown in FIG. 43, there is no need to provide a built-in large-capacity (76880 bytes) memory. Therefore, it is possible to suppress an increase in circuit scale that occurs due to an increase in the memory capacity of the built-in memory unit.

  Further, since the descrambling process by the descrambling circuit 20 can be executed in parallel with the LDC syndrome generation by the LDC syndrome generation circuit 13 and the BIS syndrome generation by the BIS syndrome generation circuit 16, the time required for the error correction process is reduced. can do. As a result, the access amount to the external buffer memory 6 can be reduced while suppressing an increase in circuit scale.

  (3) 1 shift arithmetic unit 34, 209 shift arithmetic unit 35, 108 shift so that the EDC syndrome generation circuit 30 can accurately perform EDC arithmetic processing (weighting) on the data input for each divided ECC cluster DECC. An arithmetic unit 36 and an EOR circuit 38 are provided. As a result, the EDC syndrome generation circuit 30 can generate the EDC syndrome EDCS based on the descrambling data dDu directly input from the descrambling circuit 20. Therefore, the EDC syndrome generation by the EDC syndrome generation circuit 30 can be executed in parallel with the LDC syndrome generation and the BIS syndrome generation by the LDC syndrome generation circuit 13 and the BIS syndrome generation circuit 16. Therefore, the time required for error correction processing can be shortened.

  (4) The descrambling circuit 20 includes an even register R2 that stores a scramble value SC for the first scrambled data in the first column and the first row in each even data frame of each read block. The descrambling circuit 20 includes an odd register R3 that stores a scramble value SC for the first scrambled data in the first column and the first row in each odd data frame of each read block. This eliminates the need to previously calculate and store the scramble value for the top scramble data in each even data frame and each odd data frame of each read block in a memory or the like.

(5) The descrambling circuit 20 includes a 108 shift calculator 26. By inputting the initial value S0 to the 108 shift computing unit 26, the scramble value SC (= S0 · γ ¹⁰⁸ ) for the first scrambled data in the first row and the first column in each odd data frame of the first read block RB1 can be simply obtained. Can be generated.

  (6) An EDC memory unit 40 for storing intermediate values EDCs of EDC syndrome for each data frame generated by the EDC syndrome generation circuit 30 is provided. As a result, since a common circuit can be used for EDC syndrome generation for each data frame, it is not necessary to provide a syndrome generation circuit for each data frame, and an increase in circuit scale can be suppressed. Note that the memory capacity of each buffer memory of the EDC memory unit 40 is a capacity capable of storing EDC syndrome (128 bytes) for each data frame, which is smaller than the capacity of the conventional built-in memory (76880 bytes). Therefore, an increase in circuit scale is suppressed.

  (7) An LDC memory unit 14 for storing an intermediate value of the LDC syndrome for each column generated by the LDC syndrome generation circuit 13 is provided. As a result, since a common circuit can be used for LDC syndrome generation for each column, it is not necessary to provide a syndrome generation circuit for each column, and an increase in circuit scale can be suppressed. The memory capacity of each buffer memory of the LDC memory unit 14 is a capacity capable of storing the LDC syndrome (9728 bytes) for each column, and is smaller than the capacity of the conventional built-in memory (76880 bytes). Therefore, an increase in circuit scale is suppressed.

  (8) A BIS memory unit 15 for storing all BIS clusters BISc is provided. Thereby, the BIS syndrome generation circuit 16 can continuously read all the BIS clusters BISc from the BIS memory unit 15 and generate a BIS syndrome based on the BIS clusters BISc. Therefore, the BIS syndrome generation circuit 16 can efficiently generate the BIS syndrome. Note that the memory capacity of each buffer memory of the BIS memory unit 15 is a capacity capable of storing all BIS clusters (1488 bytes), and is smaller than the capacity of a conventional built-in memory (76880 bytes). Increase is suppressed.

  (9) The built-in memory unit 12, the LDC memory unit 14, the BIS memory unit 15, and the EDC memory unit 40 are composed of two buffer memories. For example, in the case of the built-in memory unit 12, one buffer memory is used for data storage and the other buffer memory is used for access. Thereby, data storage from the demodulation circuit 11 and data reading from the built-in memory unit 12 can be performed simultaneously. Accordingly, it is possible to continuously read the disk data from the BD 4. Similarly, the intermediate value of the LDC syndrome from the LDC syndrome generation circuit 13 can be continuously stored, the BIS cluster BISc can be continuously read from the built-in memory unit 12, and the EDC syndrome generation circuit The intermediate value EDCs of the EDC syndrome from 30 can be continuously stored.

  In the case of the built-in memory unit 12, the effect of suppressing an increase in circuit scale becomes more remarkable when the built-in memory unit 12 is constituted by two buffer memories M1 and M2. Specifically, in the controller 130 shown in FIG. 43, when a plurality of ECC clusters ECCc are continuously read from the BD, in order to read the plurality of ECC clusters ECCc continuously, a large capacity (76880 bytes) is required. It is necessary to provide two internal memories 131. On the other hand, the built-in memory unit 12 of the present embodiment has a total memory capacity of 4960 bytes, and therefore can effectively suppress an increase in circuit scale.

  (10) After all the EDC syndrome EDCS generated by the EDC syndrome generation circuit 30 is stored in the correction memory unit 53, the EDC syndrome stored in the correction memory unit 53 is corrected by the EDC correction circuit 52. Thereby, since it can correct | amend continuously with respect to all the EDC syndrome EDCS stored in the correction memory part 53, processing efficiency can be improved.

(Second Embodiment)
Hereinafter, a second embodiment of the present invention will be described with reference to FIGS. The circuit configuration of the descrambling circuit and EDC syndrome generation circuit of this embodiment and the data processing order in those circuits are different from those of the first embodiment. Hereinafter, the difference from the first embodiment will be mainly described.

  First, the circuit configuration of the descrambling circuit 70 will be described with reference to FIG. The scramble value generation circuit 71 of the descrambling circuit 70 according to the present embodiment newly includes an even work register R4 and an odd work register R5 in the scramble value generation circuit 22 of the descramble circuit 20 according to the first embodiment.

  The scramble value SC for the data in the last row of each column in the final even data frame DF30 in each read block RB is stored in the even work register R4 through the 209 shift calculator 25. On the other hand, the scramble value SC for the last row data of each column in the last odd data frame DF31 in each read block RB is stored in the odd work register R5 through the 209 shift computing unit 25. The odd work register R4 and the even work register R5 output the stored scramble value SC to the selector 27.

  The selector 27 receives a selection signal from the selection circuit 72. Based on the selection signal from the selection circuit 72, the selector 27 outputs one scramble value SC among the scramble values input from the 1-shift calculator 24 and the registers R 1 to R 5 to the FF circuit 23. Note that the selection circuit 28 generates a selection signal in accordance with the number of pulses of the input clock signal CLK and outputs the selection signal to the selector 27. For example, when the first row data of each column after the second column in each even data frame of each read block RB is input to the EOR circuit 21, the selection circuit 72 uses the scramble value SC from the even work register R4. A selection signal to be output to the EOR circuit 21 is generated. When the data in the first row of each column after the second column in each odd data frame of each read block RB is input to the EOR circuit 21, the selection circuit 72 uses the scramble value SC from the odd work register R5 as the EOR circuit. A selection signal to be output to 21 is generated.

  Next, the circuit configuration of the EDC syndrome generation circuit 80 will be described with reference to FIG. In the EDC syndrome generation circuit 80 of the present embodiment, a work memory 81 is newly provided in the EDC syndrome generation circuit 30 of the first embodiment.

  In the working memory 81, the EDC operation value Er (c) calculated at the time of the last row data of each column in each data frame of each read block RB is stored as the shift operation value Es through the 209 shift operation unit 25. . The working memory 81 outputs the stored shift calculation value Es to the EOR circuit 38 and the selector 32.

  The selector 32 receives a selection signal from the selection circuit 82. The selector 27 outputs one scramble value SC among the scramble values input from the 1-shift calculator 34, the EOR circuit 38, and the work memory 81 to the EOR circuit 31 based on the selection signal from the selection circuit 82. Alternatively, the shift operation value Es is not output to the EOR circuit 31. Note that the selection circuit 82 generates a selection signal according to the number of pulses of the input clock signal CLK, and outputs the selection signal to the selector 32. For example, the selection circuit 82 shifts from the work memory 81 when data in the first row of each column excluding the first column and the last column in each data frame DF of each read block RB is input to the EOR circuit 31. A selection signal that causes the EOR circuit 31 to output the calculated value Es is generated. The selection circuit 82 outputs a selection signal for preventing the selector 32 from outputting the shift operation value Es when the first data in the first row and the first column in each data frame of each read block RB is input to the EOR circuit 31. Generate. Further, the selection circuit 82 causes the EOR circuit 31 to output the shift operation value Es from the EOR circuit 38 when the first row data of the last column in each data frame of each read block RB is input to the EOR circuit 31. A selection signal is generated.

Next, the storage arrangement of the descrambling data dDu of each data frame DF0 to DF31 stored in the external buffer memory 6 will be described with reference to FIG.
As shown in FIG. 9, the descrambling data dDu of each of the data frames DF0 to DF31 that has been descrambled by the descrambling circuit 70 is a block of 32 rows × 64 columns and is divided into four data frames DF in the column direction. Stored. Specifically, data frames DF0, DF1, DF2, and DF3 are stored side by side in the column direction, and data frames DF4, DF5, DF6, and DF7 are stored side by side in the column direction. Thereby, in the row direction, the data frames DF0, DF4, DF8, DF12, DF16, DF20, DF24, DF28 are stored side by side, and the data frames DF1, DF5, DF9, DF13, DF17, DF21, DF25, DF29 are stored. Are stored side by side. In the row direction, DF2, DF6, DF10, DF14, DF18, DF22, DF26, and DF30 are stored side by side, and DF3, DF7, DF11, DF15, DF19, DF23, DF27, and DF31 are stored side by side. The

  Next, the operation of the optical disk controller 10 configured as described above will be described. Here, the data processing order in the descrambling circuit 70 and the EDC syndrome generation circuit 80 will be described in detail.

  As shown in FIG. 24A, the descrambling circuit 70 reads the first column of scrambled data sDu in the data frame DF0 of the first read block RB1 from the internal memory unit 12 in the column direction, and then in the data frame DF4. Read scrambled data sDu in the first column in the column direction. Subsequently, the descrambling circuit 70 sequentially reads the scrambled data sDu of the first column in the data frames DF8, DF12, DF16, DF20, DF24, and DF28 in the column direction. The descrambling circuit 70 sequentially performs descrambling processing on the read scrambled data sDu, stores the descrambling data dDu in the external buffer memory 6, and outputs the descrambling data dDu to the EDC syndrome generation circuit 80. As for the descrambling process, as in the first embodiment, the initial value S0 stored in the even register R2 is used for the first data sD0 in the first row and the first column of each even data frame DF. Data dD0 is generated. Subsequently, for the first column of scrambled data sD1 to sD7 in each even-numbered data frame DF of the first read block RB1, descrambling data using a scramble value SC obtained by shifting the previous scramble value by 1 byte. dD1 to dD7 are generated.

  On the other hand, regarding the method of storing data from the descrambling circuit 70 to the external buffer memory 6, as shown in FIG. 25, the descrambling of the first column in the data frames DF0, DF4, DF8, DF12, DF16, DF20, DF24, DF28 is performed. Data dDu is stored in the same row (first row) of the external buffer memory 6. That is, in the present embodiment, data in the same column in the eight data frames DF is stored in the same row by continuously reading out data in the same column of the data frame stored in the row direction in the external buffer memory 6. I can do it.

Next, as shown in FIG. 24B, the descrambling circuit 70 receives the scrambled data sDu in the first column in the data frames DF2, DF6, DF10, DF14, DF18, DF22, DF26, and DF30 from the built-in memory unit 12. Read sequentially in the column direction. The descrambling circuit 70 sequentially performs descrambling processing on the read scrambled data sDu, stores the descrambling data dDu in the same row of the external buffer memory 6, and stores the descrambling data dDu in the EDC syndrome generation circuit 80. Output. The scramble value SC used for the descrambling process of the last row data of each column of the last even data frame DF30 (here, the scrambled data sD7 of the last row of the first column) in each read block RB is shifted by 209. It is stored in the even work register R4 through the device 25. Therefore, here, the scramble value SC (= S0 · γ ⁷ · γ ²⁰⁹ = S0 · γ ²¹⁶ ) is stored in the even work register R4. Similarly to the first embodiment, the scramble value SC used for the descrambling process of the data in the last column of the first column (here, the scrambled data sD7) in the last even data frame DF30 of each read block RB is It is stored in the even-numbered register R2 through the 1 shift calculator 24.

Next, as shown in FIG. 24C, the descrambling circuit 70 receives the scrambled data sDu in the first column in the data frames DF1, DF5, DF9, DF13, DF17, DF21, DF25, and DF29 from the built-in memory unit 12. Read sequentially in the column direction. The descrambling circuit 70 sequentially performs descrambling processing on the read scrambled data sDu, stores the descrambling data dDu in the same row of the external buffer memory 6, and stores the descrambling data dDu in the EDC syndrome generation circuit 80. Output. As for the descrambling process, as in the first embodiment, the scramble value SC stored in the odd register R3 for the first data sD108 in the first column of each odd data frame DF in the first read block RB1. generating a descrambling data dD108 using (= S0 · γ ^108). Subsequently, for the first column of scrambled data sD109 to sD115 in each odd-numbered data frame DF of the first read block RB1, descrambling data using a scramble value SC obtained by shifting the previous scramble value by 1 byte. dD109 to dD115 are generated.

Next, as shown in FIG. 24D, the descrambling circuit 70 receives the scrambled data sDu of the first column in the data frames DF3, DF7, DF11, DF15, DF19, DF23, DF27, and DF31 from the built-in memory unit 12. Read sequentially in the column direction. The descrambling circuit 70 sequentially performs descrambling processing on the read scrambled data sDu, stores the descrambling data dDu in the same row of the external buffer memory 6, and stores the descrambling data dDu in the EDC syndrome generation circuit 80. Output. Note that the scramble value SC used for the descrambling process of the last row data (here, the scrambled data sD115 of the last row of the first column) of each column of the last odd data frame DF31 in each read block RB is shifted by 209. It is stored in the odd work register R5 through the device 25. Accordingly, here, the scramble value SC (= S0 · γ ¹¹⁵ · γ ²⁰⁹ = S0 · γ ³²⁴ ) is stored in the odd-numbered work register R5. Similarly to the first embodiment, the scramble value SC used for the descrambling process of the data of the last row of the first column (here, the scrambled data sD115) in the last odd data frame of each read block RB is 1. It is stored in the odd register R3 through the shift calculator 24.

Next, the descrambling circuit 70 sequentially reads the scrambled data sDu of the second column in the data frames DF0, DF4, DF8, DF12, DF16, DF20, DF24, and DF28 from the built-in memory unit 12 in the column direction. The descrambling circuit 70 sequentially performs descrambling processing on the read scrambled data sDu. It should be noted that the scramble value SC (stored in the even work register R4) (the scrambled data sD216 in this case) for the first row data in each column after the second column in each even data frame of each read block RB ( Here, descrambling data dDu is generated using S0 · γ ²¹⁶ ). Further, for the data in the first row of the second column in each odd data frame of each read block RB (for example, scrambled data sD324), the scramble value SC (for example, S0 · γ, for example) stored in the odd work register R5. ³²⁴ ) is used to generate descrambling data dDu. Here, the scramble value SC stored in the even work register R5 is shifted by 209 bytes from the scramble value used for the last row data in the previous column in the even data frame (odd data frame) in the same read block RB. The scramble value.

  As described above, the descrambling process of the present embodiment is different from the first embodiment in the data processing order, and the scramble value used for the data in the last row of each column is 209 shift operations in accordance with the processing order. The data is stored in the work registers R4 and R5 through the device 25, and the stored scramble value is used for processing the data in the first row of the next column.

Next, EDC calculation processing in the EDC syndrome generation circuit 80 will be described. The descrambling data dDu is input from the descrambling circuit 70 to the EDC syndrome generation circuit 80 in the order of generation. That is, the descrambling data dDu is input to the EDC syndrome generation circuit 80 in the order of FIGS. Therefore, data in each data frame DF of each read block RB is input to the EDC syndrome generation circuit 80 for each column (8 bytes). When the data in each data frame DF of each read block RB is input for each column, the EDC syndrome generation circuit 80 sequentially calculates the EDC operation value Er (c) as in the first embodiment. For example, when the byte data X ^{2051 to} X ²⁰⁴⁴ in the first column in the data frame DF0 (even data frame) of the first read block RB1 are sequentially input to the EOR circuit 31, the EDC operation value Er (c ) Are calculated sequentially.

Then, when the EDC operation value Er (c) (here, the EDC operation value E1 (8)) for the data of the last row of each column in each data frame DF (here, the data frame DF0) is calculated, the EDC The calculated value E1 (8) is stored in the work memory 81 as a shift calculated value Es (= ^β209 · E1 (8)) through the 209 shift calculator 35. The work memory 81 stores a shift operation value Es for each data frame DF.

Next, an EDC operation when data in the first row of each column excluding the first column and the last column in each data frame of each read block RB is input will be described. For example, when data X ^{1835 in} the first row and second column of the data frame DF0 of the first read block RB1 is input to the EOR circuit 31, the selection circuit 82 supplies the shift operation value Es from the work memory 81 to the EOR circuit 31. A selection signal to be output is output to the selector 32. As a result, the EOR circuit 31 receives the shift operation value Es (= β) stored in the work memory 81 at the first column of the data frame DF0 in the first read block RB1 (the previous column of the same data frame in the same read block). ²⁰⁹ · E1 (8)) is input. Therefore, the EDC operation value E1 (9) for the data X ¹⁸³⁵ is calculated by the following equation.

Thus, the data in each data frame of each read block RB is input to the EDC syndrome generation circuit 80 of this embodiment for each column. For this reason, the shift calculation value Es obtained by weighting the EDC calculation value Er (c) calculated when the last data of one column is input with the 209th-order vector β ²⁰⁹ is obtained for each data frame. The data is stored in the work memory 81. When the first row data of the next column of each data frame in the same read block RB is input to the EOR circuit 31, the shift operation value Es of the same data frame stored in the work memory 81 is read and EDC operation processing is performed. To do. Other calculation methods are the same as those in the first embodiment.

According to the embodiment described above, the following effects are obtained in addition to the effects (1) to (10) of the first embodiment.
(11) The descrambling circuit 70 includes an even work register R4 that stores a scramble value SC for the first row of scrambled data in each column after the second column in each even data frame of each read block. The descrambling circuit 20 also includes an odd work register R5 that stores a scramble value SC for the first row of scrambled data in each column after the second column in each odd data frame of each read block. Thus, even if the data processing order is as shown in FIG. 24, that is, the data processing order is such that the processing moves to a different data frame every time one column of descrambling processing within each data frame is completed, By using the scramble value SC stored in the registers R4 and R5, the descrambling process can be executed continuously.

  (12) The EDC syndrome generation circuit 80 includes a working memory 81 in which the EDC operation value Er (c) after completion of the EDC operation processing for each column in each data frame of each read block is stored through the 209 shift operation unit 35. . Accordingly, even in the data processing order as shown in FIG. 24, the descrambling process can be continuously executed by using the scramble value SC stored in the working memory 81.

  (13) The descrambling circuit 70 continuously processes data in the same column of the data frame stored side by side in the row direction of the external buffer memory 6. As a result, the descrambling circuit 70 stores the descrambling data dDu in the same column in the eight data frames DF (for example, the data frames DF0, DF4, DF8, DF12, DF16, DF20, DF24, DF28) in the external buffer memory 6. Can be stored in the same row. At this time, the external buffer memory 6 has a function of allowing continuous access within the same row. Therefore, the access efficiency when the descrambling circuit 70 stores the descrambling data dDu in the external buffer memory 6 can be improved.

(Third embodiment)
Hereinafter, a third embodiment of the present invention will be described with reference to FIGS. The circuit configuration of the EDC syndrome generation circuit of this embodiment is different from that of the second embodiment. Hereinafter, the difference from the second embodiment will be mainly described. The order of reading data from the built-in memory unit 12 by the descrambling circuit 70 is the same as in the second embodiment.

As shown in FIG. 27, EDC syndrome generation circuit 90 includes 16 AND circuits A1~A16 the input data 1 bit from the descramble circuit 70 (bit data) ^{X k} is input.

  Here, the bit data input to the AND circuits A1 to A16 will be described. First, since one data frame DF is 2052 bytes, one data frame DF is composed of 16416 (= 2052 × 8) bits. A polynomial expression (input polynomial) representing the order weights of 16416-bit descrambled data dDu when EDC operation is performed in the same data processing order as when EDC is added,

It becomes. Here, as described above, since the EDC operation is an operation for weighting each degree of the input polynomial by the vector β ^k , the EDC syndrome EDCS is calculated by the following equation. Note that the order k of the vector β ^k in the present embodiment and the order n of the vector β ⁿ in the first and second embodiments have a relationship of k = 8n.

Thus, in the calculation of the EDC syndrome EDCS, each time 1-bit data is processed, each bit data is weighted by the first-order vector β ^k (= β ¹ ). Here, the 1-bit data X ^k is weighted with the vector β ^{1. The} weighted bit data X ^k is weighted with a predetermined generator polynomial G (x) (in this example, G (x) = x ³² + x ³¹ + X ⁴ +1) to shift by one bit. For example, the bit data X ¹⁶⁴¹⁵ input first in each data frame DF is weighted 16415 times with the primary vector β ¹ . However, in the case of the present embodiment, as in the second embodiment, the data block DB divided into 27 pieces is input from the descramble circuit 70 as the first to 27th read blocks RB1 to RB27 for each read block. Is done. Here, in FIG. 26 (a), showing the input order of the byte data X ⁿ and the bit data X ^k and first and second columns of each even-numbered data frame of the first read block RB1 order n, with k . Further, in FIG. 26 (b), representing the input order of the byte data X ⁿ and the bit data X ^k and first and second columns of each odd data frame of the first read block RB1 order n, with k.

As shown in FIG. 26A, the bit data X ^k is input to the EDC syndrome generation circuit 80 in the order of X ¹⁶⁴¹⁵ → X ¹⁶⁴¹⁴ →... → X ¹⁶³⁵³ → X ¹⁶³⁵² → X ¹⁴⁶⁸⁷ → X ¹⁴⁶⁸⁶ →. The That is, the AND circuit: A1 to A16 (see FIG. 27), is first ^X 16415 ^{to X} 16400 data frames ^{DF0 (X k ~X k-15} ) is input as bit data, then ^{X 16399} to X as bit data ¹⁶³⁸⁴ is input, then X ^{16387 to} X ¹⁶³⁶⁸ are input as bit data, and then X ^{16367 to} X ¹⁶³⁵² are input as bit data. When the input of bit data of the first column in all data frames DF0 to DF31 of the first read block RB1 is completed, the AND circuit A1 to A16 then receives the second column of the data frame DF0 of the first read block RB1. Bit data X ^{14687 to} X ¹⁴⁶⁷² are input. Thus, when data is input in an order different from the data input order at the time of EDC addition, weighting by the vector β cannot be performed accurately, so that a normal EDC syndrome cannot be generated.

Therefore, the EDC syndrome generation circuit 90 of the present embodiment includes a vector generation circuit 91 that generates a vector β ^k of a desired order k according to the bit data X ^k . The vector generation circuit 91 includes 16 −1 bit shift (β ⁻¹ ) arithmetic units B1 to B16 (first inverse shift arithmetic unit group) and −1664 bit shift (β ⁻¹⁶⁶⁴ ) arithmetic unit B17. Based on the vector β ^k input from the register 92, 18 vectors β are generated.

The β ^-1 calculators B1 to B16 are calculators that generate a vector β ^k-1 when the vector β ^k is input from the vector register 92 or the preceding stage β ^-1 calculator. In other words, the β ⁻¹ arithmetic units B1 to B16 are arithmetic units that generate a new vector β ^k by reversely shifting the input vector by one bit according to the generator polynomial G (x). Figure 29, beta ^-1 shift calculator B1 to B16 to I: when [31 0] is input, O output from the beta ^-1 shift calculator B1 to B16 [31: 0] logical expression showed that.

For example, in the case where vector beta ^k from vector register 92 beta ^-1 calculator B1 is input, beta ^-1 calculator B1~B16 to generate respective vectors ^{β k-1 ~β k-16} . Further, the vector β ^k input to the β ^-1 calculator B1 is input to the AND circuit A1, and the vector β ^k-1 output from the β ^-1 calculator B1 is input to the AND circuit A2, and the β ^-1 calculation is performed. The vector β ^k-2 output from the device B2 is input to the AND circuit A3. Similarly, vectors [beta] ^{k-3 to} [beta] ^k-15 output from [beta] ^-1 calculators B3 to B15 are input to AND circuits A4 to A16, respectively. The β- ¹ calculator B16 outputs the generated vector ^βk-16 to the β- ¹⁶⁶⁴ calculator B17, the even register R20, the odd register R21, and the selector 96.

The β- ¹⁶⁶⁴ computing unit B17 generates a vector β ^k-16-1664 when the vector β ^k-16 is input from the β- ¹ computing unit B16, and generates the vector β ^k-16-1664 as an even line. This is an arithmetic unit that outputs to the register R22, the odd line register R23, and the selector 96. That is, the β- ¹⁶⁶⁴ calculator B17 is a calculator that reversely shifts the input vector by 1664 bits according to the generator polynomial G (x) to generate a new vector ^βk . FIG. 30 shows a logical expression of O [31: 0] output from the β- ¹⁶⁶⁴ shift calculator B17 when I [31: 0] is input to the β- ¹⁶⁶⁴ shift calculator B17.

AND circuit A1~A16 the bit data is inputted from the descramble circuit 70 ^{X k} ~X ^k-15 and a value that extends to each 32-bit, vectors β ^k ~β ^k respectively input from the vector generating circuit 91 The logical product with ^-15 is taken and the operation result is output to the EOR circuit 94. Here, the value expanded to 32 bits is data in which all values from bit 0 to bit 31 are set to the value of ^Xn .

  The EOR circuit 94 calculates the exclusive OR of the 16 operation results input from the AND circuits A1 to A16 and outputs the operation result to the syndrome register 95. The syndrome register 95 outputs the calculation result to the EOR circuit 94 and the EDC memory unit 40 as an intermediate value EDCs of EDC syndrome based on the enable signal ENS at the EN terminal. The EDC memory unit 40 has the same configuration as that of the first and second embodiments, and stores an EDC syndrome intermediate value EDCs and a final EDC syndrome EDCS for each data frame DF0 to DF31.

Based on the selection signal from the selection circuit 97, the selector 96 outputs one vector ^βk among the vectors input from the β- ¹ calculator B16 and the various registers R20 to R23 to the vector register 92. The vector register 92 outputs the vector β ^k input from the selector 96 to the vector generation circuit 91 based on the enable signal ENS input to the EN terminal.

The even register R20 at the time of initialization stores a vector β ¹⁶⁴¹⁵ calculated in advance from a predetermined generator polynomial G (x). The even register R20 receives the vector β input from the β- ¹ calculator B16 when the calculation processing of the bit data X in the last row of the first column in the final even data frame of each read block RB is completed. Stored. On the other hand, the odd-numbered register R21 at the time of initialization stores a vector β ¹⁵⁵⁵¹ calculated in advance from a predetermined generator polynomial G (x). The odd-numbered register R21 receives the vector β input from the β- ¹ calculator B16 when the calculation processing of the bit data X in the last row of the first column in the last odd data frame of each read block RB is completed. Stored.

The even line register R22 stores the vector β input from the β- ¹⁶⁶⁴ arithmetic unit B17 when the EDC arithmetic processing of the bit data X of the last row of each column in the final even data frame of each read block RB is completed. Is done. The odd line register R23 stores the vector β input from the β- ¹⁶⁶⁴ arithmetic unit B17 when the EDC arithmetic processing of the last row of bit data in each column in the final odd data frame of each read block RB is completed. The

The selection circuit 97 generates various selection signals according to the number of pulses of the input clock signal CLK and outputs the selection signals to the selector 96. For example, when 16 bits of data from the first bit data in the first row and first column in each even data frame of each read block RB is input to the AND circuits A1 to A16, the selection circuit 97 receives the vector from the even register R20. A selection signal for generating the vector β ^k as the vector β ^k is generated. When 16 bits of data from the first bit data in the first row and first column in each odd data frame of each read block RB is input to the AND circuits A1 to A16, the selection circuit 97 receives the vector β from the odd register R21. A selection signal to be output to the vector register 92 as the vector β ^k is generated. The selection circuit 97 receives the vector β from the even line register R22 when 16 bits of data are input from the first row of bit data in each column after the second column in each even data frame of each read block RB. A selection signal to be output to the vector register 92 as the vector β ^k is generated. The selection circuit 97 receives the vector β from the odd line register R23 when 16 bits of data are input from the first row of bit data in each column after the second column in each odd data frame of each read block RB. A selection signal to be output to the vector register 92 as the vector β ^k is generated. When 16 bits of data are input from the next bit data in the column direction (bit data after the 17th row) in each read block RB, the selection circuit 97 uses the vector β from the β ⁻¹ arithmetic unit B16 as a vector. A selection signal to be output to the vector register 92 as β ^k is generated.

Next, the operation of the EDC syndrome generation circuit 90 configured as described above will be described with reference to FIG.
First, in step S50, initial values are set in the even register R20 and the odd register R21. That is, the even number register R20 stores the vector β ¹⁶⁴¹⁵ having the same order as the order of the first bit data X ¹⁶⁴¹⁵ of each even data frame in the first read block RB1. The odd number register R21 stores a vector β ¹⁵⁵⁵¹ having the same order as the order of the first bit data X ¹⁵⁵⁵¹ of each odd data frame in the first read block RB1.

Next, the selection circuit 97 outputs a selection signal for outputting the vector β from the even register R20 or the odd register R21 to the vector register 92 to the selector 96 (step S51). Specifically, the selection circuit 97 vectorizes the vector β from the even register R20 when performing the EDC operation processing of the even data frame, and vector vector from the odd register R21 when performing the EDC operation processing of the odd data frame. A selection signal to be output to the register 92 is generated. Here, since the data frame DF to be processed is the data frame DF0 (even data frame), a selection signal for outputting the vector β ¹⁶⁴¹⁵ from the even register R20 to the vector register 92 is output to the selector 96.

Next, in step S52, 16 bits worth of data X ^{16415 to} X ¹⁶⁴⁰⁰ are input from the first bit data X ¹⁶⁴¹⁵ of the data frame DF0 to the AND circuits A1 to A16 from the descrambling circuit 70, respectively. At this time, vectors β ^{16415 to} β ¹⁶⁴⁰⁰ are input from the vector generation circuit 91 to the AND circuits A1 to A16, respectively. As a result, the EOR circuit 94 calculates the intermediate value EDCs of the EDC syndrome represented by the following equation by taking the exclusive OR of the operation results from the AND circuits A1 to A16.

The calculated intermediate value EDCs is stored in the EDC memory unit 40 via the syndrome register 95 and input to the EOR circuit 94.

Next, in step S53, the selection circuit 97 determines whether or not the EDC operation processing for one column (64 bits) of data in each read block RB is completed according to the number of pulses of the input clock signal. If selection circuit 97 determines that the EDC operation processing for one column of data has not been completed (NO in step S53), selector circuit 97 selects a selection signal for outputting vector β from β- ¹ calculator B16 to vector register 92. It outputs to 96 (step S54). In this case, beta ^-1 for vector input from the arithmetic unit B16 in the vector register 92 via the selector 96 is beta ^16399, to the AND circuit A1~A16 from the vector generating circuit 91 vector β ¹⁶³⁹⁹ ~β ¹⁶³⁸⁴ respectively Entered. Further, bit data X ^{16399 to} X ¹⁶³⁸⁴ are input from the descramble circuit 70 to the AND circuits A1 to A16, respectively. Therefore, the EOR circuit 94 calculates an intermediate value EDCs represented by the following equation by taking an exclusive OR of the calculation results from the AND circuits A1 to A16 and the intermediate value EDCs calculated immediately before.

These steps S52 to S54 are repeatedly executed until the EDC operation processing for the bit data for one column is completed. In this way, when bit data is input in the same order as the data processing order when EDC is added, that is, when bit data arranged in the column direction is input in order, the vector β from the β ⁻¹ computing unit B16 Is input to the vector generation circuit 91.

Then, an intermediate value EDCs when the first read first column of the 16 bits including bit data ^{X 16352} the last line data ^X of the data frame DF0 blocks ^RB1 16367 ~X ¹⁶³⁵² is input to the AND circuit A1~A16 Once calculated, the intermediate value EDCs is stored in the EDC memory unit 40, and the process proceeds to step S55. Next, in step S55, the selection circuit 97 determines whether the data frame DF currently processed is the final even data frame DF30 or the final odd data frame DF31. Here, since the data frame DF currently processed is the data frame DF0 and the data in the first column, the process proceeds to step S51, and the same column of the next data frame (here, one column of the data frame DF4). Eye) EDC calculation processing is performed in the same manner as the data frame DF0.

Thereafter, steps S51 to S55 are repeatedly executed, and the data frame DF4 → DF8 → DF12 → DF16 → DF20 → DF24 → DF28 → DF2 → DF6 → DF10 → DF14 → DF18 → DF22 → DF26 → DF30 (FIG. 24A, ( The EDC calculation processing of the first column in (b)) is sequentially performed. When the intermediate value EDCs is calculated when the first read first column of the 16 bits including bit data ^{X 16352} the last line data ^X of the data frame DF30 blocks ^RB1 16367 ~X ¹⁶³⁵² is input (step (YES in S55), the process proceeds to step S56.

Next, the selection circuit 97 determines which of the conditions C1 to C3 the data currently being subjected to EDC calculation processing corresponds to. When the EDC calculation process for the first column of the final even data frame DF30 is completed (condition C1), the process proceeds to step S57, and when the EDC calculation process for the first column of the final odd data frame DF31 is completed (condition C2). In step S58, if the EDC calculation processing for the second and subsequent columns is completed (condition C3), the process proceeds to step S59. Here, since the EDC arithmetic processing of the first column of the data frame DF30 has been completed (condition C1), the vector β from the β ⁻¹ arithmetic unit B16 is stored in the even register R20 and the β ⁻¹⁶⁶⁴ arithmetic unit. The vector β from B17 is stored in the even line register R22 (step S57). That is, when the first read block RB1, along with vectors beta ¹⁶³⁵¹ is stored in the even register R20, vector beta ¹⁴⁶⁸⁷ is stored in the even-numbered line register R22. Subsequently, when returning to step S51, since the EDC calculation process for the first column of the data frame DF1 (odd data frame) in the first read block RB1 is started, the initial value vector β ¹⁵⁵⁵¹ from the odd register R21 is It is output to the vector register 92. Thereafter, as in the case of the even data frame, steps S52 to S54 are repeatedly executed until the EDC calculation process for the first column of the data frame DF1 in the first read block RB1 is completed. Then, an intermediate value EDCs when the first read first column of the 16 bits including bit data ^{X 16352} the last line data ^X of the data frame DF1 blocks RB1 15503 ^{to X 15488} is input to the AND circuit A1~A16 When calculated, the process proceeds to step S55.

Thereafter, steps S51 to S55 are repeatedly executed, and the data frame DF5 → DF9 → DF13 → DF17 → DF21 → DF25 → DF29 → DF3 → DF7 → DF11 → DF15 → DF19 → DF23 → DF27 → DF31 (FIG. 24C) ( The EDC calculation process in the first column in d) is performed in order. When the intermediate value EDCs is calculated when the data ^X 16 367 ^{to X 16352} of 16 bits including data ^{X 16352} in the last line of the first column of the data frame DF31 of the first read block RB1 is input (step S55 In step S56, the process proceeds to step S58. Next, beta ^-1 vector beta (here, beta ¹⁵⁴⁸⁷⁾ from the computing unit B16 with is stored in the odd register R21, beta ^-1664 vector beta (here, beta ¹³⁸²³⁾ from the computing unit B17 is odd lines Stored in the register R23 (step S58).

Subsequently, the selection circuit 97 outputs a selection signal for causing the vector register 92 to output the vector β from the even line register R22 or the odd line register R23 to the selector 96 (step S59). Specifically, the selection circuit 97 uses the vector β from the even line register R22 when performing the EDC operation processing of the even data frame, and selects the vector β from the odd line register R23 when performing the EDC operation processing of the odd data frame. Is generated by the vector register 92. Here, since the second column of the data frame DF0 in the first read block RB1 is subjected to EDC calculation processing, a selection signal for causing the vector register 92 to output the vector β ¹⁴⁶⁸⁷ from the even line register R22 is output to the selector 96. . Thereby, in step S52, the vectors β ^{14687 to} β ¹⁴⁶⁷² are input from the vector generation circuit 91 to the AND circuits A1 to ^A16 . Also, 16 bits of data X ^{14687 to} X ¹⁴⁶⁷² are input to the AND circuits A1 to ^A16 from the bit data X ^{14687 in} the first row and the second column of the data frame DF0 in the first read block RB1. Then, the EOR circuit 94 calculates a new intermediate value EDCs by taking an exclusive OR of the calculation results from the AND circuits A1 to A16 and the intermediate value EDCs of the same data frame stored in the EDC memory unit 40. To do.

As described above, the bit data X ^k in the first row of each column after the second column in the even data frame of each read block is multiplied by the vector β ^{k + 1665} multiplied by the bit data X ^{k + 1665} of the last row in the previous column in the even data frame. ^Are multiplied by a vector β ^k obtained by performing a shift operation by the β ⁻¹ computing unit B16 and the β ⁻¹⁶⁶⁴ computing unit B17. Similarly, the bit data X ^k of the first row of each column after the second column in the odd-numbered data frame of each read block is multiplied by the vector β ^{k + 1665} multiplied by the bit data X ^{k + 1665} of the last row of the previous column in the odd-numbered data frame. ^Are multiplied by a vector β ^k obtained by performing a shift operation by the β ⁻¹ computing unit B16 and the β ⁻¹⁶⁶⁴ computing unit B17. Here, for example, the bit data X ¹⁴⁶⁸⁷ in the first row and the second column in the data frame ^DF0 of the first read block RB1 is more than the bit data X ^{16352 in} the last row in the first column when the EDC is added {(data block DB The total number of bits in the column direction: 1728 bits) − (the number of bits in the column direction of one read block: 64 bits) +1} = 1655 bits. That is, the bit data X ¹⁴⁶⁸⁷ has the number of times of weighting by the vector β ^{1 smaller} than the bit data X ¹⁶³⁵² by 1665 times. Therefore, as described above, the front row of the bit data X ^{k + 1665} vector beta ^k where the vector beta ^{k + 1665} reverse shift operation by 1665 bits multiplied the last row, the first row of bit data in each column of the second row and subsequent ^Xk is multiplied.

Subsequently, step S52~S54 are repeated, when the first read second column of the 16 bits including bit data ^{X 14624} the last line data ^X of the data frame DF0 blocks RB1 14639 ^{to X 14624} is input When the intermediate value EDCs is calculated, the process proceeds to step S55. Here, since the second column of the final even data frame DF30 and the data frame DF0 which is not the final odd data frame is processed, the process proceeds to step S59. Subsequently, since the second column of the data frame ^DF4 is processed next, the vector β ¹⁴⁶⁸⁷ from the even line register R22 is output to the vector register 92.

Thereafter, steps S51 to S55 and step S59 are repeatedly executed, and the second column in the data frame DF4 → DF8 → DF12 → DF16 → DF20 → DF24 → DF28 → DF2 → DF6 → DF10 → DF14 → DF18 → DF22 → DF26 → DF30 EDC calculation processing is performed in order. When the intermediate value EDCs is calculated when the first read second column of the 16 bits including bit data ^{X 16352} the last line data ^X of the data frame DF30 blocks ^RB1 16367 ~X ¹⁶³⁵² is input (step In S55, YES and the condition C3 of step S56), the process proceeds to step S61. Next, the vector β (β ^{12959 in} this case) from the β- ¹⁶⁶⁴ arithmetic unit B17 at this time is stored in the even line register R22 (step S61).

Subsequently, since the second column of the data frame DF 1 is processed next, the vector β ¹³⁸²³ from the odd line register R 21 is output to the vector register 92. Thereafter, the EDC operation processing of the odd data frame is performed in the same manner as the even data frame. When the EDC operation processing of the second column of the data frame DF31 is completed, the process proceeds to step S61. Next, the vector β (β ^{12095 in} this case) from the β- ¹⁶⁶⁴ arithmetic unit B17 at this time is stored in the odd line register R23 (step S61).

  Similarly, the EDC calculation processing for the third and subsequent columns of each data frame in the first read block RB1 is performed, and the last column (304th column) in the first read block RB1, that is, the last column of the last odd data frame DF31. When the EDC calculation process ends (YES in step S60), the process proceeds to step S62. Next, in step S62, the selection circuit 97 determines whether or not the read block for which the EDC calculation process is completed is the thirteenth read block RB13. If the selection circuit 97 determines that the read block for which the EDC calculation processing has been completed is not the thirteenth read block RB13 (NO in step S62), the selection circuit 97 proceeds to step S63. Next, the selection circuit 97 determines whether or not the EDC operation processing of the final read block RB27 is completed (Step S63). Then, steps S50 to S63 are repeatedly executed until the EDC calculation processing of the read block RB2 to the thirteenth read block RB13 is completed. When the EDC calculation processing of the thirteenth read block RB13 is completed (YES in step S62), step S64 is performed. Move on.

Here, the bit data in the first row and the first column of each odd-numbered data frame in the fourteenth read block RB14 is the first input in each odd-numbered data frame when the EDC operation is performed in the same data input order as when EDC is added. Bit data X ¹⁶⁴¹⁵ . Therefore, before the EDC operation processing of the fourteenth read block RB14 is started, in step S64, the vector β ¹⁶⁴¹⁵ having the same order as the order of the first bit data X ¹⁶⁴¹⁵ of each odd data frame is set as an initial value in the odd register R22. Stored.

Thereafter, steps S50 to S63 are repeatedly executed, and the EDC arithmetic processing of the 14th to 27th read blocks RB14 to RB27 is executed.
According to the embodiment described above, the following effects can be obtained in addition to the effects (1) to (13) of the first and second embodiments.

(14) The EDC syndrome generation circuit 90 includes 16 AND circuits A1 to A16 and a vector generation circuit 91. As a result, the bit data X ^k input to the AND circuits A1 to A16 can be multiplied by the vector β ^k of the desired order k. Therefore, the EDC syndrome EDCS can be generated without performing special arithmetic processing such as the first correction calculation and the second correction calculation in the EDC calculation processing of the first and second embodiments.

(Fourth embodiment)
Hereinafter, a fourth embodiment of the present invention will be described with reference to FIG. The configuration of the built-in memory unit of this embodiment and the connection configuration of the BIS memory unit are different from those of the first embodiment. Hereinafter, the difference from the first embodiment will be mainly described.

  As shown in FIG. 31, the disk data read from the BD 4 is sequentially input to the demodulation circuit 11a of the optical disk controller 10 through the input / output drive circuit 5 (see FIG. 1). The demodulation circuit 11a generates demodulated data, that is, an ECC cluster ECCc (see FIG. 40B) based on the disk data. The demodulation circuit 11a classifies the LDC cluster LDCc and the BIS cluster BISc in the ECC cluster ECCc, outputs the LDC cluster LDCc to the built-in memory unit 12e (LDC data memory unit), and outputs the BIS cluster BISc in the ECC cluster ECCc. The data is output to the BIS memory unit 15a.

  Each of the two buffer memories M11 and M12 constituting the built-in memory unit 12e has a memory capacity capable of storing data of a predetermined number of bytes (2432 bytes in this embodiment). In this embodiment, the LDC cluster LDCc (75392 bytes) in the ECC cluster ECCc is divided into 31 (= 75392/2432) pieces and stored in the buffer memories M11 and M12 of the built-in memory unit 12 every 2432 bytes. That is, the LDC cluster LDCc is divided every 16 rows, and each buffer memory M1 is divided into blocks of 16 rows × 152 columns (= 2432 bytes) (block units excluding the BIS cluster BISc portion in the divided ECC cluster DECC of FIG. 3). , M2.

  The BIS memory unit 15a stores the BIS cluster BISc from the demodulation circuit 11a. The BIS memory unit 15a has the same configuration as that of the first embodiment. When data of the entire BIS cluster BISc is stored, the BIS cluster BISc is output to the BIS syndrome generation circuit 16.

According to the embodiment described above, the following effects are obtained in addition to the effects (1) to (10) of the first embodiment.
(15) The BIS cluster BISc is stored directly in the BIS memory unit 15a from the demodulation circuit 11a, and only the LDC cluster LDCc in the divided ECC cluster DECC is stored in the built-in memory unit 12e. As a result, the access to the built-in memory unit 12e by the BIS memory unit 15a can be omitted, and the access amount to the built-in memory unit 12e can be reduced. Furthermore, since the BIS memory unit 15a receives only the BIS cluster BISc directly from the demodulation circuit 11a, the process of receiving the LDC cluster LDCc can be omitted. From this, the circuit configuration of the controller 10 can be simplified.

(Fifth embodiment)
Hereinafter, a fifth embodiment of the present invention will be described with reference to FIG. The descrambling circuit, BIS syndrome generation circuit, and EDC correction circuit of this embodiment are different from those of the fourth embodiment. Hereinafter, the difference from the fourth embodiment will be mainly described.

  As shown in FIG. 32, the first to twenty-seventh read blocks RB1 to RB27 (scrambled data sDu) sequentially stored in the buffer memory of the built-in memory unit 12e are stored in the external buffer memory 6 and the EDC syndrome generation circuit 30a.

  The EDC syndrome generation circuit 30a reads the first to twenty-seventh read blocks RB1 to RB27 from the built-in memory unit 12e for each read block RB, and performs EDC calculation based on the scrambled data sDu in each read block RB. Generate EDC syndrome. That is, the EDC syndrome generation circuit 30 a generates an EDC syndrome for the scrambled data sDu for each data frame DF in the read block RB, and stores the EDC syndrome in the EDC memory unit 40. The EDC syndrome generation circuit 30a has substantially the same configuration as the EDC syndrome generation circuit 30 shown in FIG. 8, and the scramble data sDu from the built-in memory unit 12e is replaced with EOR as byte data instead of the descramble data dDu. The signal is input to the circuit 31. The EDC syndrome generation circuit 30a may be changed to the EDC syndrome generation circuit 80 shown in FIG. 23 or the EDC syndrome generation circuit 90 shown in FIG.

  As in the first embodiment, the EDC memory unit 40 stores the final EDC syndrome for the scrambled data sDu of each data frame DF in the external buffer memory 6.

  The BIS syndrome generation circuit 16a reads the BIS cluster BISc from the BIS memory unit 15a, and generates a BIS syndrome based on the BIS cluster BISc. The BIS syndrome generation circuit 16 a stores the generated BIS syndrome and the BIS cluster BISc read from the BIS memory unit 15 a in the external buffer memory 6.

  The correction circuit 51a in the error correction circuit 50 performs error correction processing on the data frame DF or the like including the scrambled data sDu stored in the external buffer memory 6 based on the LDC syndrome and BIS syndrome read from the external buffer memory 6 I do. The correction circuit 51a outputs the error position and error value calculated based on various syndromes to the EDC correction circuit 52a. The correction circuit 51a reads the initial value S0 of the scramble value in the BIS cluster BISc after error correction stored in the external buffer memory 6, and outputs the initial value S0 to the EDC correction circuit 52a.

  The EDC correction circuit 52a calculates a scramble value SC based on the initial value S0 input from the correction circuit 51a. The EDC correction circuit 52a performs EDC syndrome correction based on the calculated scramble value SC in addition to EDC syndrome correction based on the error position and error value input from the correction circuit 51.

  When it is confirmed in the EDC check circuit 54 that the error correction has been normally completed, the descrambling circuit 115 similar to the prior art performs the scramble processing on the scrambled data sDu after error correction stored in the external buffer memory 6. Read in the same order. Also, the descrambling circuit 115 reads the value in the BIS cluster BISc after error correction stored in the external buffer memory 6 as the initial value S0 of the scramble value, and sequentially generates the scramble value SC based on the initial value S0. . The descrambling circuit 115 sequentially performs descrambling processing on the read scrambled data sDu, and outputs the descrambling data dDu to the interface circuit 60.

According to the embodiment described above, the following effect is obtained in addition to the effect (15) of the fourth embodiment.
(16) The scramble value SC used for the descrambling process by the descrambling circuit 20 of the fourth embodiment is the address information of the BD4 used for reading the data by the demodulation circuit 11a or the value in the BIS cluster BISc before error correction. Will be used. If there is a possibility that data is incorrect before error correction, the value in the BIS cluster BISc before error correction is not used, but the value in the optical disk controller 10 does not include an error and is demodulated. The address information used by the circuit 11 is used as the scramble value SC. In this case, if the demodulation circuit 11 does not read the correct address, an incorrect scramble value SC is used. This may cause a problem that the descrambling process does not end correctly although error correction and EDC check end correctly.

  On the other hand, in the descrambling circuit 115 of this embodiment, the value in the BIS cluster BISc after error correction can be used as the scramble value SC. Therefore, the descrambling circuit 115 can perform descrambling processing on the data that has been subjected to error correction and EDC check by the error correction circuit 50, using the values in the BIS cluster BISc after error correction. As a result, more correct data can be sent to the host computer.

In addition, the said embodiment can also be implemented in the following aspects which changed this suitably.
In each of the above embodiments, the EDC syndrome generation circuit 30, 80, 90 generates the EDC syndrome EDCS based on the descrambling data dDu input from the descrambling circuits 20, 70. For example, as shown in FIG. 33, an EDC check (generation of EDC syndrome) may be performed based on descrambled data after error correction by an EDC check circuit 116 similar to the conventional one.

  The descrambling circuit 70, the EDC syndrome generation circuit 80 of the second embodiment, or the EDC syndrome generation circuit 90 of the third embodiment may be applied to the controller 10 in the fourth and fifth embodiments.

  In the first embodiment, the descrambling circuit 20 and the EDC syndrome generation circuit 30 may access the built-in memory unit 12 independently. That is, the scrambled data sDu in the divided ECC cluster DECC is input from the built-in memory unit 12 substantially simultaneously to the EOR circuit 21 of the descrambling circuit 20 and the EOR circuit 31 of the EDC syndrome generation circuit 30 in this case.

In each of the above embodiments, the size of the divided ECC cluster DECC stored in each buffer memory M1, M2 of the built-in memory unit 12 is not particularly limited. For example, the data size of the divided ECC cluster DECC may be 32 rows × 155 columns. In this case, the buffer memories M1 and M2 of the built-in memory unit 12 are changed to a memory capacity that can store one divided ECC cluster DECC. In this case, the number of rows in one read block RB is changed. When the number of rows of the read block RB is changed in this way, the 209 shift calculator 25 of the descrambling circuits 20 and 70, the 209 shift calculator 35 of the EDC syndrome generation circuits 30 and 80, and the EDC syndrome generation circuit 90 The number of shift bytes (number of shift bits) in the β- ¹⁶⁶⁴ arithmetic unit B17 is changed.

  The data processing order in the descrambling circuit 70 of the third embodiment may be changed to the same data processing order as in the first embodiment. That is, the descrambling circuit 70 of the third embodiment reads each scrambled data in each read block RB in the same order as the data arrangement in the data block DB, and sequentially descrambles the read scrambled data. Processing may be executed. In this case, the even line register R22 and the odd line register R23 in the EDC syndrome generation circuit 90 may be omitted.

  The data processing order in the descrambling circuit 70 of the second and third embodiments is not particularly limited. For example, first, as shown in FIG. 24A, the descrambling process of the first column of the data frame DF0 → DF4 → DF8 → DF12 → DF16 → DF20 → DF24 → DF28 (even data frame group) is performed. Next, as shown in FIG. 24 (c), the descrambling process in the first column of the data frame DF1-> DF5-> DF9-> DF13-> DF17-> DF21-> DF25-> DF29 (odd data frame group) may be performed. Good.

  Alternatively, after the descrambling process of the first column of the even data frame group as shown in FIGS. 24A and 24B is completed, the descrambling process of the second column of the even data frame group is performed. Also good. In this way, after all the descrambling processing of the even data frame is completed, the processing is shifted to the descrambling processing of the odd data frame, for example, the even work register R4 and the odd work register in the descrambling circuit 70 of the second embodiment. The work register R5 can be changed to a common work register. Similarly, the even line register R22 and the odd line register R23 in the EDC syndrome generation circuit 90 of the third embodiment can be changed to a common line register.

  In each of the above embodiments, each data frame DF is stored in the external buffer memory 6 as a block of 32 rows × 64 columns, but this block size is not particularly limited. For example, each data frame DF may be stored in the external buffer memory 6 in a block of 64 rows × 32 columns. By storing in this way, all even data frames can be stored side by side in the row direction of the external buffer memory 6 and all odd data frames can be stored side by side in the row direction of the external buffer memory 6. Therefore, the descrambling circuit 70 continuously descrambles the scrambled data in the same column of the data frames DF0, DF2, DF4, DF6,..., DF28, DF30, for example, 128 (= 8 × 16). Byte descrambling data can be stored in the same row of the external buffer memory 6.

The 108 shift calculator 26 of the descrambling circuits 20 and 70 in the first and second embodiments may be omitted. In this case, for example, for each initial value, a scramble value SC (S0 · γ ¹⁰⁸ ) obtained by shifting 108 bytes from the initial value according to the generator polynomial Φ (x) is calculated in advance and stored in a memory or the like. May be.

The even register R2 and the odd register R3 of the descrambling circuits 20 and 70 in the first and second embodiments may be omitted. Alternatively, the even register R20 and the odd register R21 of the EDC syndrome generation circuit 90 in the third embodiment may be omitted. In this case, for example, a scramble value SC (vector β ^k ) for the first scrambled data (first bit data) in the first row and first column in each even data frame and each odd data frame of each read block RB is calculated in advance and stored in a memory or the like. It may be stored.

The -211 shift calculator 37 of the EDC syndrome generation circuits 30 and 80 in the first and second embodiments may be omitted. In this case, for example, for the even data frame, it is calculated when EDC calculated value for the data of the first to fourteenth read block RB1~RB14 E14 (76), i.e. the byte data ^{X 0} is input to the EOR circuit 31 EDC The calculated value E14 (76) is stored separately from the intermediate value EDCs. Then, the EDC calculation processing for the data of the fifteenth to twenty-seventh read blocks RB15 to RB27 is executed without reading the EDC calculation value E14 (76). Finally, the calculation result obtained by performing first correction operation in EDC calculated value E14 (76) at which the byte data X ¹⁰⁸ of the last row last column is entered in each even data frame of the 27 read block RB27, the By taking an exclusive OR with the EDC operation value E14 (76), an EDC syndrome EDCS similar to the above equation (1) is generated.

  The correction memory unit 53 in each of the above embodiments may be omitted. In this case, the EDC correction circuit 52 corrects the EDC syndrome EDCS while reading the EDC syndrome EDCS from the external buffer memory 6.

  In the above embodiment, the internal configuration of the built-in memory unit 12, the LDC memory unit 14, the BIS memory unit 15 and the EDC memory unit 40 is embodied in the configuration of FIG. 5, but one buffer memory is stored in the same time. There is no particular limitation as long as it is used for use (for saving syndrome) and the other buffer memory is used for access (for data output and for syndrome output).

  In each of the above embodiments, the built-in memory unit 12, the LDC memory unit 14, the BIS memory unit 15, and the EDC memory unit 40 are configured by two buffer memories M1 and M2, but one buffer memory or three or more The buffer memory may be configured. Further, it may be configured such that data can be simultaneously stored and read from one buffer memory.

  In the above embodiments, the LDC memory unit 14 may be omitted. In this case, it is preferable to provide LDC syndrome calculators for all columns (304 columns) of the parity addition block PB in the LDC syndrome generation circuit 13. Further, an intermediate value of the LDC syndrome generated by the LDC syndrome generation circuit 13 may be stored in the external buffer memory 6.

  The BIS memory unit 15 in each of the above embodiments may be omitted. In this case, the BIS syndrome generation circuit 16 reads the BIS cluster BISc directly from the built-in memory unit 12. At this time, for example, an intermediate value of the BIS syndrome generated by the BIS syndrome generation circuit 16 may be stored in the external buffer memory 6. A syndrome memory for storing an intermediate value of the BIS syndrome may be provided between the BIS syndrome generation circuit 16 and the external buffer memory 6.

  In the above embodiments, the EDC memory unit 40 may be omitted. In this case, EDC syndrome calculators (for example, EOR circuit 31 and various calculators 34 to 37) corresponding to the number of all data frames (30) are provided in the EDC syndrome generation circuits 30, 80, 90. Is preferred.

The various embodiments described above can be summarized as follows.
(Appendix 1)
A predetermined number of data frames are scrambled, the scrambled data frame is converted into a data block of a predetermined size, and a parity addition block in which parity is added to each column of the data block is generated. An error correction device that corrects an error in data in which a cluster is generated by performing interleaving processing on the cluster and the cluster is modulated,
A demodulation circuit that demodulates the modulated data and generates clusters as demodulated data;
A built-in memory unit in which a divided cluster obtained by dividing the cluster from the demodulation circuit into a predetermined number of rows is stored;
Each data in the parity addition block generated by deinterleaving the divided cluster is input in the column direction through the built-in memory unit, and a parity calculation result is generated for each column of the parity addition block based on the data. A parity operation result generation circuit;
Through the built-in memory unit, de-interleave processing is performed on the divided clusters, each scrambled data in the data block is input as a read block in the column direction, and descramble data is generated by performing descrambling processing on each scrambled data A descrambling circuit,
An error correction circuit that performs error correction processing on the descrambling data based on the parity calculation result, and
The descrambling circuit includes:
A first exclusive OR circuit to which the scrambled data from the built-in memory unit is input;
A first shift computing unit that generates a new scramble value by shifting the input scramble value by one byte according to a predetermined first generator polynomial;
The input scramble value is
{(Total number of bytes in the column direction of the data block) + 1− (total number of bytes in the column direction of each read block)}
A second shift calculator for generating a new scrambled value by shifting according to the first generator polynomial for bytes;
One of the scramble values input from the first shift calculator and the second shift calculator according to the position of the scrambled data input to the first exclusive OR circuit in the data block. And a first selection circuit that outputs the scramble value to the first exclusive OR circuit, the first shift calculator, and the second shift calculator.
(Appendix 2)
The parity calculation result generation circuit performs a deinterleaving process on the divided clusters stored in the built-in memory unit, reads each data in the parity addition block in a column direction,
The error correction apparatus according to claim 1, wherein the descrambling circuit performs deinterleaving processing on the divided clusters stored in the built-in memory unit, and reads each scrambled data in the data block in a column direction. .
(Appendix 3)
The built-in memory unit is subjected to deinterleave processing from the demodulation circuit and stores the divided clusters.
The parity calculation result generation circuit reads the parity addition block stored in the built-in memory unit in a column direction,
The error correction apparatus according to claim 1, wherein the descrambling circuit reads the data block stored in the built-in memory unit in a column direction.
(Appendix 4)
The first data frame processed in the data block and every other data frame processed from the data frame are even data frames, and the data frame processed next to the even data frames is odd data frames. sometimes,
The descrambling circuit includes:
A first even register in which a scramble value for the scrambled data in the last row of the first column in the even data frame of each read block is stored through the first shift calculator;
A first odd register in which a scramble value for the scrambled data in the last row of the first column in the odd data frame of each read block is stored through the first shift computing unit;
The first selection circuit includes the first shift computing unit, the second shift computing unit, and the first even number according to a position in the data block of the scrambled data input to the first exclusive OR circuit. One of the scramble values input from the register and the first odd register is output to the first exclusive OR circuit, the first shift calculator, and the second shift calculator. The error correction device according to any one of appendices 1 to 3, characterized by:
(Appendix 5)
The descrambling circuit includes:
A third shift calculator that generates a new scramble value by shifting the input scramble value by 108 bytes according to the first generator polynomial;
In the initial value setting at the start of descrambling processing in each cluster,
The first even register stores an initial value of a scramble value set according to each cluster,
The error correction apparatus according to appendix 4, wherein the initial value is stored in the first odd register through the third shift computing unit.
(Appendix 6)
The descrambling circuit includes:
An even working register in which a scramble value for the scrambled data of the last row of each column in the even data frame of each read block is stored through the second shift computing unit;
An odd work register in which a scramble value for the scrambled data of the last row of each column in the odd data frame of each read block is stored through the second shift computing unit;
The first selection circuit includes the first shift computing unit, the second shift computing unit, and the first even number according to a position in the data block of the scrambled data input to the first exclusive OR circuit. One of the scramble values input from the register, the first odd register, the even work register, and the odd work register is converted into the first exclusive OR circuit, the first shift arithmetic unit, 6. The error correction apparatus according to appendix 4 or 5, wherein the error correction apparatus outputs to the second shift computing unit.
(Appendix 7)
The first selection circuit includes:
When the first scramble data in the first row and the first column in each even data frame of each read block is input to the first exclusive OR circuit, the scramble value input from the first even register is output. ,
When the first scrambled data in the first row and the first column in each odd-numbered data frame of each read block is input to the first exclusive OR circuit, the scramble value input from the first odd-numbered register is output. ,
When scrambled data from the second row of each column in each data frame of each read block is input to the first exclusive OR circuit, a scramble value input from the first shift calculator is output. And
When the scrambled data of the first row of each column after the second column in each data frame of each read block is input to the first exclusive OR circuit, it is input from the second shift arithmetic unit. The error correction apparatus according to appendix 5, wherein a scramble value is output.
(Appendix 8)
The first selection circuit includes:
When the first scramble data in the first row and the first column in each even data frame of each read block is input to the first exclusive OR circuit, the scramble value input from the first even register is output. ,
When the first scrambled data in the first row and the first column in each odd-numbered data frame of each read block is input to the first exclusive OR circuit, the scramble value input from the first odd-numbered register is output. ,
When the scramble data of the first row of each column after the second column in each even data frame of each read block is input to the first exclusive OR circuit, the scramble is input from the even work register. Output the value
When the scrambled data of the first row of each column after the second column in each odd data frame of each read block is input to the first exclusive OR circuit, the scramble input from the odd work register Output the value
When scrambled data from the second row of each column in each data frame of each read block is input to the first exclusive OR circuit, a scramble value input from the first shift calculator is output. The error correction apparatus according to appendix 6, wherein:
(Appendix 9)
The error correction according to any one of appendices 1 to 8, wherein the descrambling circuit reads each scrambled data in the same order as the data arrangement in the data block in each read block. apparatus.
(Appendix 10)
The supplementary note 6 or 8, wherein the descrambling circuit sequentially reads out the scrambled data in the same column of the data frame stored side by side in the row direction of the external memory in each of the read blocks. The error correction device described.
(Appendix 11)
The first selection circuit includes:
Appendix 7 wherein the initial value is output to the first exclusive OR circuit when the top data in the scramble process in each odd data frame is input to the first exclusive OR circuit. 10. The error correction device according to any one of 10.
(Appendix 12)
An EDC syndrome generation circuit that receives the descrambling data from the descrambling circuit and generates an EDC syndrome based on the descrambling data;
An error correction apparatus according to any one of appendices 1 to 11, further comprising: an EDC memory unit that stores an intermediate result of the EDC syndrome generated by the EDC syndrome generation circuit for each data frame. .
(Appendix 13)
The EDC syndrome generation circuit includes:
A second exclusive OR that generates an EDC operation value by taking an exclusive OR of the descrambling data input as byte data from the descrambling circuit and a shift operation value input from the second selection circuit Circuit,
The EDC operation value input from the second exclusive OR is weighted to shift by 1 byte according to a predetermined second generator polynomial to generate a shift operation value, and the shift operation value is stored in the EDC memory unit. A first weighting circuit stored in
For the EDC operation value input from the second exclusive OR,
{(Total number of bytes in the column direction of the data block) + 1− (number of bytes in the column direction of each read block)}
A second weighting circuit that generates a shift operation value by weighting according to the second generator polynomial for bytes,
A shift operation value is generated by weighting an EDC operation value input from the second exclusive OR circuit by 108 bytes according to the second generator polynomial, and the shift operation value is generated in the EDC memory unit. A third weighting circuit stored in
A third exclusive OR circuit that generates a shift operation value by taking an exclusive OR of the shift operation value from the second weighting circuit and the intermediate result of the EDC syndrome from the EDC memory unit. ,
The second selection circuit includes the first weighting circuit, the second weighting circuit, and the third exclusive logic according to a position in the data block of the byte data input to the second exclusive OR circuit. 13. The error correction apparatus according to appendix 12, wherein any one of the shift operation values input from the sum circuit is output to the second exclusive OR circuit.
(Appendix 14)
A predetermined number of data frames are scrambled, the scrambled data frame is converted into a data block of a predetermined size, and a parity addition block in which parity is added to each column of the data block is generated. An error correction device that corrects an error in data in which a cluster is generated by performing interleaving processing on the cluster and the cluster is modulated,
A demodulation circuit that demodulates the modulated data and generates clusters as demodulated data;
A built-in memory unit in which a divided cluster obtained by dividing the cluster from the demodulation circuit into a predetermined number of rows is stored;
Each data in the parity addition block generated by deinterleaving the divided cluster is input in the column direction through the built-in memory unit, and a parity calculation result is generated for each column of the parity addition block based on the data. A parity operation result generation circuit;
An EDC syndrome generation circuit that performs deinterleaving processing on the divided clusters through the built-in memory unit, inputs each scrambled data in the data block as a read block in a column direction, and generates an EDC syndrome based on the scrambled data; ,
An EDC memory unit that stores an intermediate result of the EDC syndrome generated by the EDC syndrome generation circuit for each data frame;
An error correction circuit that performs error correction processing on the scrambled data based on the parity calculation result, and
The EDC syndrome generation circuit includes:
A second exclusive OR circuit that generates an EDC operation value by taking an exclusive OR of the scrambled data input as byte data from the built-in memory unit and the shift operation value input from the second selection circuit When,
The EDC operation value input from the second exclusive OR is weighted to shift by 1 byte according to a predetermined second generator polynomial to generate a shift operation value, and the shift operation value is stored in the EDC memory unit. A first weighting circuit stored in
For the EDC operation value input from the second exclusive OR,
{(Total number of bytes in the column direction of the data block) + 1− (number of bytes in the column direction of each read block)}
A second weighting circuit that generates a shift operation value by weighting according to the second generator polynomial for bytes,
A shift operation value is generated by weighting an EDC operation value input from the second exclusive OR circuit by 108 bytes according to the second generator polynomial, and the shift operation value is generated in the EDC memory unit. A third weighting circuit stored in
A third exclusive OR circuit that generates a shift operation value by taking an exclusive OR of the shift operation value from the second weighting circuit and the intermediate result of the EDC syndrome from the EDC memory unit. ,
The second selection circuit includes the first weighting circuit, the second weighting circuit, and the third exclusive logic according to a position in the data block of the byte data input to the second exclusive OR circuit. An error correction apparatus, wherein any one of shift operation values input from a sum circuit is output to the second exclusive OR circuit.
(Appendix 15)
A correction memory unit storing an EDC syndrome stored in an external memory from the EDC syndrome generation circuit;
Based on the error position and error value calculated by the error correction circuit, the EDC syndrome stored in the correction memory unit is corrected, and the scramble value calculated from the initial value of the scramble value in the BIS cluster of the cluster 15. The error correction apparatus according to appendix 14, further comprising: an EDC correction circuit that corrects the EDC syndrome based on the EDC.
(Appendix 16)
16. The error correction apparatus according to appendix 14 or 15, wherein the EDC syndrome generation circuit reads the byte data in the read block in the same order as the data arrangement in the data block.
(Appendix 17)
Note that the EDC syndrome generation circuit sequentially reads the byte data in the same column of the data frame stored side by side in the row direction of the external memory in each reading block in order. 15. The error correction device according to 15.
(Appendix 18)
The EDC syndrome generation circuit includes:
A shift operation value is generated by weighting the EDC operation value input from the second exclusive OR circuit by a reverse shift of 211 bytes according to the second generator polynomial, and the shift operation value is generated by the EDC. 18. The error correction apparatus according to any one of appendices 13 to 17, further comprising a fourth weighting circuit stored in the memory unit.
(Appendix 19)
The first data frame processed in the data block and every other data frame processed from the data frame are even data frames, and the data frame processed next to the even data frames is odd data frames. sometimes,
In the EDC memory unit,
The EDC operation value generated when the byte data of the last row and last column in each data frame of each read block is input to the second exclusive OR circuit is converted into the EDC syndrome value through the first weighting circuit. Stored as an intermediate result,
An EDC operation value generated when byte data of the last row and last column in each even data frame of the last read block input last to the EDC syndrome generation circuit is input to the second exclusive OR circuit. , Stored as the EDC syndrome through the third weighting circuit,
The EDC operation value generated when the byte data of the last row and last column in each odd data frame of the last read block is input to the second exclusive OR circuit is stored as the EDC syndrome. The error correction device according to any one of Supplementary notes 13 to 18, which is characterized by the following.
(Appendix 20)
The first data frame processed in the data block and every other data frame processed from the data frame are even data frames, and the data frame processed next to the even data frames is odd data frames. sometimes,
In the EDC memory unit,
An EDC operation value generated when the last byte data at the time of EDC addition in each even data frame is input to the second exclusive OR circuit is stored as an intermediate result of the EDC syndrome through the fourth weighting circuit. The error correction device according to appendix 18, characterized in that:
(Appendix 21)
The EDC syndrome generation circuit includes:
An EDC operation value generated when byte data of the last row of each column in each data frame of each read block is input to the second exclusive OR circuit is converted to a shift operation value through the second weighting circuit. With working memory stored as
The working memory outputs the stored shift operation value to the third exclusive OR circuit and the second selection circuit,
The second selection circuit includes the first weighting circuit, the working memory, and the third exclusive OR circuit according to a position in the data block of the byte data input to the second exclusive OR circuit. 21. The error correction apparatus according to any one of appendices 13 to 20, wherein any one of the shift operation values input from is output to the second exclusive OR circuit.
(Appendix 22)
The second selection circuit includes:
When the byte data from the second row in each data frame of each read block is input to the second exclusive OR circuit, the shift operation value input from the first weighting circuit is output,
Input from the second weighting circuit when byte data of the first row of each column excluding the first column and the last column in each data frame of each read block is input to the second exclusive OR circuit Output the shift operation value
Shift operation value input from the third exclusive OR circuit when byte data of the first row of the last column in each data frame of each read block is input to the second exclusive OR circuit The error correction apparatus according to any one of appendices 13 to 20, characterized in that:
(Appendix 23)
The second selection circuit includes:
When the byte data from the second row in each data frame of each read block is input to the second exclusive OR circuit, the shift operation value input from the first weighting circuit is output,
When the byte data of the first row of each column excluding the first column and the last column in each data frame of each read block is input to the second exclusive OR circuit, it is input from the working memory. Output the shift operation value,
Shift operation value input from the third exclusive OR circuit when byte data of the first row of the last column in each data frame of each read block is input to the second exclusive OR circuit The error correction apparatus according to appendix 21, wherein:
(Appendix 24)
The EDC syndrome generation circuit includes:
A first inverse shift computing unit group in which N first inverse shift computing units that reversely shift an input vector by one bit in accordance with a predetermined second generator polynomial to generate a new vector; A vector input from the last stage of the N inverse shift calculators is
{(Total number of bits in the column direction in the data block) − (Number of bits in the column direction of each read block)}
A vector generation circuit configured to include a second inverse shift computing unit that performs a reverse shift according to the second generation polynomial to generate a new vector by a bit amount;
N AND circuits to which the descrambling data is input as bit data from the descrambling circuit and a vector to be input to the N first inverse shift computing units is input;
A fourth exclusive OR circuit that takes an exclusive OR of the operation results from the N AND circuits and generates an intermediate result of the EDC syndrome;
According to the position of the bit data input to the AND circuit in the data block, any one of vectors input from the first inverse shift calculator and the second inverse shift calculator in the final stage 13. The error correction apparatus according to appendix 12, further comprising a third selection circuit that outputs a vector to the first inverse shift computing unit in the first stage.
(Appendix 25)
A predetermined number of data frames are scrambled, the scrambled data frame is converted into a data block of a predetermined size, and a parity addition block in which parity is added to each column of the data block is generated. An error correction device that corrects an error in data in which a cluster is generated by performing interleaving processing on the cluster and the cluster is modulated,
A demodulation circuit that demodulates the modulated data and generates clusters as demodulated data;
A built-in memory unit in which a divided cluster obtained by dividing the cluster from the demodulation circuit into a predetermined number of rows is stored;
Each data in the parity addition block generated by deinterleaving the divided cluster is input in the column direction through the built-in memory unit, and a parity calculation result is generated for each column of the parity addition block based on the data. A parity operation result generation circuit;
An EDC syndrome generation circuit that performs deinterleaving processing on the divided clusters through the built-in memory unit, inputs each scrambled data in the data block as a read block in a column direction, and generates an EDC syndrome based on the scrambled data; ,
An EDC memory unit that stores an intermediate result of the EDC syndrome generated by the EDC syndrome generation circuit for each data frame;
An error correction circuit that performs error correction processing on the scrambled data based on the parity calculation result, and
The EDC syndrome generation circuit includes:
A first inverse shift computing unit group in which N first inverse shift computing units that reversely shift an input vector by one bit in accordance with a predetermined second generator polynomial to generate a new vector; A vector input from the last stage of the N inverse shift calculators is
{(Total number of bits in the column direction in the data block) − (Number of bits in the column direction of each read block)}
A vector generation circuit configured to include a second inverse shift computing unit that performs a reverse shift according to the second generation polynomial to generate a new vector by a bit amount;
N AND circuits to which the scrambled data is input as bit data from the demodulation circuit and a vector to be input to the N first inverse shift calculators are input;
A fourth exclusive OR circuit that takes an exclusive OR of the operation results from the N AND circuits and generates an intermediate result of the EDC syndrome;
According to the position of the bit data input to the AND circuit in the data block, any one of vectors input from the first inverse shift calculator and the second inverse shift calculator in the final stage And a third selection circuit for outputting a vector to the first inverse shift computing unit in the first stage.
(Appendix 26)
The first data frame processed in the data block and every other data frame processed from the data frame are even data frames, and the data frame processed next to the even data frames is odd data frames. sometimes,
The EDC syndrome generation circuit includes:
A second even register in which a vector multiplied by the bit data of the last row of the first column in the even data frame of each read block is stored through the first inverse shift computing unit of the last stage;
A second odd register in which a vector multiplied by the bit data of the first row of the first column in the odd data frame of each read block is stored through the first inverse shift computing unit of the final stage;
The third selection circuit includes a first inverse shift computing unit, a second inverse shift computing unit, and a second even number in the final stage according to the position of the bit data input to the AND circuit in the data block. 26. The error correction apparatus according to appendix 24 or 25, wherein any one of a register and a vector input from the second odd register is output to the first inverse shift computing unit in the first stage.
(Appendix 27)
The EDC syndrome generation circuit includes:
In the initial value setting at the start of EDC calculation processing in each cluster,
The second even register stores a 16415th-order vector generated according to the second generator polynomial,
27. The error correction apparatus according to appendix 26, wherein the second odd number register stores a 15551th-order vector generated according to the second generator polynomial.
(Appendix 28)
The EDC syndrome generation circuit includes:
An even line register in which a vector multiplied by the bit data of the last row of each column in the even data frame of each read block is stored through the second inverse shift calculator;
An odd line register in which a vector multiplied by the bit data of the last row of each column in the odd data frame of each read block is stored through the second inverse shift calculator,
The third selection circuit includes a first reverse shift computing unit at the final stage, the second even register, the second odd register, according to the position of the bit data input to the AND circuit in the data block. 28. The error correction apparatus according to appendix 26 or 27, wherein any one of vectors input from the even line register and the odd line register is output to the first inverse shift computing unit in the first stage.
(Appendix 29)
The third selection circuit includes:
When bit data for N bits including the first bit data in the first row and the first column in each even data frame of each read block is input to the N AND circuits, it is input from the second even register. Output vector,
When bit data for N bits including the first bit data in the first row and the first column in each odd data frame of each read block is input to the N AND circuits, it is input from the second odd register. Output vector,
When the bit data of N bits including the bit data of the first row of each column after the second column in each even data frame of each read block is input to the N AND circuits, the even line The vector input from the register is output,
When the bit data of N bits including the bit data of the first row of each column after the second column in each odd data frame of each read block is input to the N AND circuits, the odd line The vector input from the register is output,
When the bit data for N bits including the bit data of the (N + 1) th and subsequent rows in each data frame of each read block is input to the N AND circuits, the first reverse shift operation in the final stage 29. The error correction apparatus according to appendix 28, wherein the vector input from the device is output.
(Appendix 30)
12. The error correction apparatus according to any one of appendices 1 to 11, further comprising an EDC syndrome generation circuit that generates an EDC syndrome based on the error-corrected descrambling data.
(Appendix 31)
The built-in memory unit includes an LDC data memory unit in which LDC clusters in the divided clusters are stored from the demodulation circuit, and a BIS memory unit in which BIS clusters in the clusters are stored from the demodulation circuit,
The error correction apparatus according to any one of appendices 1 to 30, further comprising a BIS syndrome generation circuit that reads the BIS cluster from the BIS memory unit and generates a BIS syndrome based on the BIS cluster.
(Appendix 32)
The LDC data memory unit includes at least two memories,
At least one memory of the LDC data memory unit is used for data storage for storing LDC clusters in the divided clusters from the demodulation circuit, and at least one memory of the LDC data memory unit generates the parity calculation result The error correction apparatus according to attachment 31, wherein the error correction apparatus is used for access accessed from a circuit.
(Appendix 33)
The built-in memory unit is composed of at least two memories,
At least one memory of the built-in memory unit is used for data storage for storing the divided clusters from the demodulation circuit, and at least one memory of the built-in memory unit is accessed from the parity calculation result generation circuit The error correction apparatus according to any one of supplementary notes 1 to 30, wherein the error correction apparatus is used as an application.
(Appendix 34)
A small-capacity BIS memory unit for storing BIS clusters in the divided clusters from the built-in memory unit;
The error correction apparatus according to any one of appendices 1 to 30, 33, wherein a BIS syndrome is generated based on the BIS cluster read from the BIS memory unit. (Appendix 35)
The BIS memory unit includes at least two memories.
At least one memory of the BIS memory unit is used for data storage for storing the BIS cluster, and at least one memory of the BIS memory unit is used for data output accessed from the BIS syndrome generation circuit. The error correction device according to any one of supplementary notes 31, 32, and 34.
(Appendix 36)
The error correction apparatus according to any one of appendices 1 to 35, further comprising a small-capacity LDC memory unit that stores an intermediate result of the parity calculation result generated by the parity calculation result generation circuit.
(Appendix 37)
The LDC memory unit is composed of at least two memories,
At least one memory of the LDC memory unit is used for saving a syndrome for storing an intermediate result of the parity calculation result from the parity calculation result generation circuit, and at least one memory of the LDC memory unit is used as an external memory. 37. The error correction apparatus according to appendix 36, wherein the error correction apparatus is used for data output for outputting a parity operation result. (Appendix 38)
A correction memory unit storing an EDC syndrome stored in an external memory from the EDC syndrome generation circuit;
Any one of Supplementary notes 12 to 37, further comprising: an EDC correction circuit that corrects an EDC syndrome stored in the correction memory unit based on an error position and an error value calculated by the error correction circuit. The error correction device described in 1.
(Appendix 39)
A data reading device configured to include an error correction device and a memory and read data written on an optical disc,
The error correction device includes:
A predetermined number of data frames are scrambled, the scrambled data frame is converted into a data block of a predetermined size, and a parity addition block in which parity is added to each column of the data block is generated. As a demodulated data, a cluster is generated by performing an interleaving process, and the modulated data is read from a high-density optical disk on which data on which the cluster is modulated is written, and the modulated data is demodulated. A demodulation circuit for generating a cluster;
A built-in memory unit in which a divided cluster obtained by dividing the cluster from the demodulation circuit into a predetermined number of rows is stored;
Each data in the parity addition block generated by deinterleaving the divided cluster is input in the column direction through the built-in memory unit, and a parity calculation result is generated for each column of the parity addition block based on the data. A parity operation result generation circuit;
Through the built-in memory unit, de-interleave processing is performed on the divided clusters, each scrambled data in the data block is input as a read block in the column direction, and descramble data is generated by performing descrambling processing on each scrambled data A descrambling circuit,
An error correction circuit that performs an error correction process on the descrambling data stored in the memory based on the parity calculation result stored in the memory;
The descrambling circuit includes:
A first exclusive OR circuit to which the scrambled data from the built-in memory unit is input;
A first shift computing unit that generates a new scramble value by shifting the input scramble value by one byte according to a predetermined first generator polynomial;
The input scramble value is
{(Total number of bytes in the column direction of the data block) + 1− (total number of bytes in the column direction of each read block)}
A second shift calculator for generating a new scrambled value by shifting according to the first generator polynomial for bytes;
One of the scramble values input from the first shift calculator and the second shift calculator according to the position of the scrambled data input to the first exclusive OR circuit in the data block. And a first selection circuit that outputs the scrambled value to the first exclusive OR circuit, the first shift computing unit, and the second shift computing unit.
(Appendix 40)
A predetermined number of data frames are scrambled, the scrambled data frame is converted into a data block of a predetermined size, and a parity addition block in which parity is added to each column of the data block is generated. Is a descrambling circuit that is used to read data that is interleaved to generate a cluster and the cluster is modulated and written to the optical disc.
A first exclusive OR circuit to which the read data is input;
A first shift computing unit that generates a new scramble value by shifting the input scramble value by one byte according to a predetermined first generator polynomial;
The input scramble value is
{(Total number of bytes in the column direction of the data block) + 1− (total number of bytes in the column direction of each read block)}
A second shift calculator for generating a new scrambled value by shifting according to the first generator polynomial for bytes;
One of the scramble values input from the first shift calculator and the second shift calculator depending on the position of the read data input to the first exclusive OR circuit in the data block. A descrambling circuit comprising: a first selection circuit that outputs the scramble value to the first exclusive OR circuit, the first shift calculator, and the second shift calculator.

1 Optical disk control device (data reading device)
4 Blu-ray Disc (High Density Optical Disc)
6 External buffer memory (external memory)
10 Optical disk controller (error correction device)
11, 11a Demodulation circuit 12, 12e Built-in memory unit 13 LDC syndrome generation circuit (parity calculation result generation circuit)
14 LDC memory unit 15, 15a BIS memory unit 16, 16a BIS syndrome generation circuit 20, 70, 115 Descramble circuit 21 EOR circuit (first exclusive OR circuit)
24 1 shift computing unit (first shift computing unit)
25 209 shift arithmetic unit (second shift arithmetic unit)
26 108 shift operator (third shift operator)
28, 72 selection circuit (first selection circuit)
30, 30a, 80, 90 EDC syndrome generation circuit 31 EOR circuit (second exclusive OR circuit)
34 1 shift computing unit (first weighting circuit)
35 209 shift computing unit (second weighting circuit)
36 108 shift calculator (third weighting circuit)
37-211 shift arithmetic unit (fourth weighting circuit)
38 EOR circuit (third exclusive OR circuit)
39, 82 selection circuit (second selection circuit)
40 EDC memory unit 50 Error correction circuit 51, 51a Correction circuit 52, 52a EDC correction circuit 53 Correction memory unit 81 Working memory 91 Vector generation circuit 94 EOR circuit (fourth exclusive OR circuit)
97 Selection circuit (third selection circuit)
A1 to A16 AND circuit B1 to B16 β-1 shift computing unit (first inverse shift computing unit, first inverse shift computing unit group)
B17 β-1664 shift computing unit (second inverse shift computing unit)
M1 to M8, M11, M12 Buffer memory R2 Even register (first even register)
R3 odd register (second odd register)
R4 Even work register R5 Odd work register R20 Even register (second even register)
R21 Odd register (second odd register)
R22 Even line register R23 Odd line register

Claims (5)

Scrambling process is performed on a predetermined number of data frames, scrambled data frame is converted into data blocks of a predetermined size, parity adding block parity is added to each column of the data block is generated, the parity adding block respect are generated interleave processing is performed cluster, the cluster is a error correction device for correcting errors in data which has been modulated,
A demodulation circuit that demodulates the modulated data and generates clusters as demodulated data;
A built-in memory unit in which a divided cluster obtained by dividing the cluster from the demodulation circuit into a predetermined number of rows is stored;
Through the built-in memory unit, the data in the parity adding block created by deinterleaving the divided cluster is input in the column direction, the parity calculation result generated for each column of the parity addition block based to the each data A parity operation result generation circuit;
Through the built-in memory unit, deinterleaving is performed on the divided clusters, each scrambling data in the data block is inputted as a read block in the column direction, and EDC syndrome generation circuit for generating EDC syndrome on the basis of the scramble data ,
An EDC memory unit that stores an intermediate result of the EDC syndrome generated by the EDC syndrome generation circuit for each data frame;
An error correction circuit that performs error correction processing on the scrambled data based on the parity calculation result, and
The EDC syndrome generation circuit includes:
And the scrambling data input as byte data from said internal memory unit, taking an exclusive OR of the shift operation value input from the channel selection択回passage, a first exclusive OR circuit which generates an EDC calculated value ,
Against EDC operation value inputted from said first exclusive OR, to generate a shift operation value by weighting for one byte shift in accordance with a predetermined generate polynomials, the shift operation value to the EDC memory unit A first weighting circuit for storing;
For the EDC operation value input from the first exclusive OR,
{(Total number of bytes in the column direction of the data block) + 1− (number of bytes in the column direction of each read block)}
A second weighting circuit for generating a shift operation value by the bytes, the weighting is shifted according to the previous Kisei formed polynomial,
Against EDC operation value inputted from said first exclusive OR circuit, and the weighting to be 108 bytes shifted according to the previous Kisei formed polynomial to generate a shift operation value, the said shift operation value EDC memory unit A third weighting circuit stored in
A second exclusive OR circuit that generates a shift operation value by taking an exclusive OR of the shift operation value from the second weighting circuit and the intermediate result of the EDC syndrome from the EDC memory unit. ,
Before cyclohexene択回path, depending on the position in the data block of the byte data input to the first exclusive OR circuit, wherein the first weighting circuit, said second weighting circuit and said second exclusive logical Output any one of the shift operation values input from the sum circuit to the first exclusive OR circuit ;
The data frame processed first in the data block and the data frame processed every other data frame are even data frames, and the data frame processed next to the even data frames is odd data frames. sometimes,
In the EDC memory unit,
An EDC operation value generated when byte data of the last row and last column in each data frame of each read block is input to the first exclusive OR circuit is converted into the EDC syndrome value through the first weighting circuit. Stored as an intermediate result,
An EDC operation value generated when byte data of the last row and last column in each even data frame of the last read block inputted last to the EDC syndrome generation circuit is inputted to the first exclusive OR circuit. , Stored as the EDC syndrome through the third weighting circuit,
An EDC operation value generated when byte data of the last row and last column in each odd data frame of the last read block is input to the first exclusive OR circuit is stored as the EDC syndrome. Characteristic error correction device. A correction memory unit storing an EDC syndrome stored in an external memory from the EDC syndrome generation circuit;
Based on the error position and error value calculated by the error correction circuit, the EDC syndrome stored in the correction memory unit is corrected, and the scramble value calculated from the initial value of the scramble value in the BIS cluster of the cluster The error correction apparatus according to claim 1, further comprising an EDC correction circuit that corrects the EDC syndrome based on the EDC.   3. The error correction apparatus according to claim 1, wherein the EDC syndrome generation circuit reads the byte data in the read block in the same order as the data arrangement in the data block. The EDC syndrome generation circuit includes:
Against EDC operation value inputted from said first exclusive OR circuit, before 211 bytes according Kisei formed polynomial, to generate the shift operation value by weighting to reverse shift, the said shift operation value EDC The error correction apparatus according to any one of claims 1 to 3, further comprising a fourth weighting circuit stored in the memory unit. The EDC syndrome generation circuit includes:
An EDC operation value generated when byte data of the last row of each column in each data frame of each read block is input to the first exclusive OR circuit is converted to a shift operation value through the second weighting circuit. With working memory stored as
The working memory outputs the stored shift operation value to the second XOR circuit and before hexene択回path,
Before cyclohexene択回path, the first according to the position in the data block of the byte data input to the exclusive OR circuit, wherein the first weighting circuit, the working memory and said second exclusive OR circuit errors according to any one of claims 1 to 4, wherein a single shift operation value one of the shift operation value input to output to the first XOR circuit from Correction device.