KR20030073532A - Apparatus and method for operating double product outer syndrome - Google Patents

Abstract

PURPOSE: A double PO syndrome operating device and a method thereof are provided to reduce an access request of ECC block data when operating the PO syndrome so as to improve bandwidth of a buffer memory. CONSTITUTION: A double PO syndrome operating method includes the steps of reading ECC block data by 2 bytes from a buffer memory by a control part, dividing the ECC block data into upper and lower bytes, and operating the syndrome from the upper and lower bytes and roots of a set generating polynomial by odd and even row of PO syndrome operating parts(221,223).

Description

Apparatus and method for operating double product outer syndrome}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a PO syndrome calculating device and method for calculating a PO (Product Outer) syndrome in order to determine an error of a data stream reproduced on an optical disc. The present invention relates to a PO syndrome calculating device and method.

Optical discs such as DVDs (Digital Video Discs) can record and play back certain numbers, letters, figures, and voices, as well as moving images, and use digital recording. There is no need for a very good recording medium.

Optical disc recording and reproducing apparatus such as a DVD player using such an optical disc records, reproduces and outputs a predetermined data stream on the optical disc via the optical pickup unit.

In order to accurately reproduce the data stream recorded on the optical disc, the optical disc recording / reproducing apparatus first inserts an error correction code into the data stream and then inserts the error correction code into the data stream when recording the data stream onto the optical disc. One data stream is modulated for an optical disk and then recorded on the optical disk.

When reproducing the data stream recorded on the optical disc, the data stream is read from the optical disc, demodulated, and the error of the reproduced data stream is corrected by using the error correction code inserted in the data stream when the data stream is recorded. Done.

Algorithms for decoding RS (Reed-Solomon) codes to correct errors in data streams read from optical discs are known as Berlekamp-Massey algorithm, Euclid algorithm, and Peterson algorithm, all of which are error-position polynomials (error) The location of the error can be found using the locator polynomial.

That is, the error location polynomial has information for finding the location of error occurrence from the syndrome, and the above algorithms are a theory of the solution of solving the system of equations with the coefficients of the error location polynomial. The occurrence position of can be found.

In an optical disc recorder, an ECC (Error Correction Code) is composed of an inner code and an outer code as an interleaved product code. That is, as illustrated in FIG. 1, one ECC block has 182 columns of 172 bytes of data symbols and 10 parity bytes, and 208 rows of 192 data symbols and 16 parity symbols. In the PI (Product Inner) correction, up to 5 errors can be corrected, and in PO correction, up to 8 errors or 16 erasers can be corrected.

The basic role of the decoder in the optical disc recording / reproducing apparatus is to restore the original data by correcting the error if the reproduced data stream contains an error. However, it is not possible to know where the error occurred in the data stream, how much the error value is, and how many errors occurred, and the only information is information related to the occurrence of the error from the parity bits inserted while encoding the data stream. .

When the decoder performs PI correction and PO correction, as shown in FIG. 2, the decoder inputs a data stream in step 100, and in step 102, computes a syndrome with ECC block data, and in step 104, the data. Determines whether an error has occurred in the stream.

If it is determined in step 104 that there are no errors, the process ends immediately. If it is determined that an error has occurred in step 104, the error location polynomial is calculated in step 106, and the error location polynomial is found using the calculated error location polynomial in step 108, and the step ( In step 110, the error value corresponding to the occurrence of the found error is determined, and in step 112, the determined error value is added to the data stream to correct the error, and the original data stream without error is restored. .

In the decoder that performs such an error correction operation, in order to perform ECC correction, the ECC block data shown in FIG. 1 must be stored in advance. A 16 Mbit synchronous dynamic random access memory (SDRAM) can be used for high-speed data processing and high capacity. ) Or more high-end memory is used as buffer memory to store ECC block data.

The ECC block data of FIG. 1 is stored in the buffer memory, and ECC block data must be read from the buffer memory when PI correction and PO correction are performed. In this case, the ECC block data is considered in consideration of the access efficiency of the memory. It constitutes a memory map of data.

When the PI correction is performed, the ECC block data is read by accessing the buffer memory in a row direction, so that the ECC block data can be efficiently read using the page mode of the buffer memory. However, when performing the PO correction, it is necessary to access the buffer memory in the column direction to read the ECC block data, which requires more bandwidth of the buffer memory than the PI correction.

Therefore, in order to secure a bandwith of an appropriate buffer memory, accessing the buffer memory to read ECC block data in PO correction is considered as an important factor.

Accordingly, an object of the present invention is to improve the bandwidth of the buffer memory by reducing the access request of the ECC block data when calculating the PO syndrome, dual PO syndrome calculation device that can reduce the calculation time of the PO syndrome by 1/2 times and To provide a method.

Another object of the present invention is to provide a dual PO syndrome calculation device and method that can be shared with the calculation device for PI syndrome calculation.

Another object of the present invention is to provide a dual PO syndrome calculation apparatus and method which is advantageous for PI correction → PO correction → PI correction or PI correction → PO correction → PI correction → PO correction in addition to the PI correction → PO correction, which is an existing ECC correction. have.

The dual PO syndrome calculating device and method of the present invention having such a purpose reduces the access request of the ECC block data, which occupies the most in the bandweed of the buffer memory, by doubling the operation of the PO syndrome in ECC correction.

In other words, when performing ECC correction, ECC block data is required, and a DRAM or the like is used as a buffer memory to store the ECC block data. The buffer memory can write and read ECC block data in word units (2 bytes).

Conventionally, when calculating a single PO syndrome, a buffer memory is read to read ECC block data in word units, and block data of a column corresponding to an odd-numbered PO ECC is selected from the read ECC block data. Calculating

In addition, when the block data of the column corresponding to the even-numbered PO ECC is selected to calculate the PO syndrome, the ECC block data is read again by accessing the same address of the buffer memory, thereby reducing the efficiency of bandweed of the buffer memory.

In the present invention, when reading ECC block data by accessing the buffer memory, two rows of PO syndromes are calculated together to improve the bandweed of the buffer memory as a whole.

Therefore, the dual PO syndrome calculating device of the present invention reads the ECC block data stored in advance by the buffer memory by two bytes according to the control of the control unit controlling the operation of the PO syndrome, and the two-byte ECC read out of the buffer memory. The dual PO syndrome calculating unit receives the block data and the root of the predetermined generation polynomial, and double-operates the PO syndrome.

The dual PO syndrome calculating unit includes an even column PO syndrome calculating unit and an odd column PO syndrome calculating unit. The even column PO syndrome calculating unit includes a lower byte of data and a generated polynomial among 2 bytes of ECC block data received from the buffer memory. The PO column is computed by the operation, and the even-numbered PO PO calculation unit is configured to calculate the PO syndrome from the root of the high-order byte data and the generated polynomial among two-byte ECC block data received from the buffer memory. Both arithmetic unit and odd column PO syndrome arithmetic unit, a plurality of syndrome output unit calculates and outputs a syndrome, and a plurality of logical multiplying means performs a plurality of syndromes outputted from the plurality of syndrome output units and a plurality of roots of a predetermined generation polynomial. Logical multiplication, and the output signal of the logical multiplication means and the ECC block And the logic sum means inputs the data to the syndrome output unit after the logic sum means.

In the dual PO syndrome calculation method of the present invention, the control unit reads ECC block data by 2 bytes from the buffer memory, divides the read 2 byte ECC block data into lower bytes and higher bytes, and divides the lower bytes and The even-order column PO syndrome calculation unit and the odd-numbered column PO syndrome calculation unit respectively receive the upper byte and the root of the set generation polynomial, and calculate a syndrome.

1 is a view showing the configuration of ECC block data for error correction,

2 is a signal flow diagram showing a general PI and PO error correction process,

3 is a view showing the configuration of the syndrome calculation device of the present invention,

4 is a view showing a basic configuration of a syndrome calculation device,

5 is a view for explaining the operation of the dual PO syndrome calculation device of the present invention,

6 is a view showing the configuration of a dual syndrome calculation unit in the dual PO syndrome calculation device of the present invention,

7 is a signal flow diagram illustrating a process of performing error correction after calculating a double PO syndrome according to the dual PO syndrome calculation method of the present invention.

* Description of the symbols for the main parts of the drawings *

200: control unit 210: buffer memory

220: dual PO syndrome calculation unit 221: even column PO syndrome calculation unit

223: odd column PO syndrome calculation unit 300: syndrome output unit

302: logical sum means 304: logical product means

Hereinafter, a dual PO syndrome calculating device and a method of the present invention will be described in detail with reference to the accompanying drawings.

3 is a view showing the configuration of a dual PO syndrome calculation device of the present invention. As shown therein, the control unit 200 controls the operation of the PO syndrome, a buffer memory 210 which stores ECC block data in advance and reads the data by 2 bytes according to the control of the control unit 200, and the buffer memory. It consists of a double PO syndrome calculation unit 220 that calculates the syndrome using the two-byte ECC block data read out at 210 and the roots α ^{O to} α ¹⁵ of the generation polynomial set in advance.

The dual PO syndrome calculating device of the present invention having such a configuration outputs the ECC block data, which is pre-stored by the buffer memory 210, by two bytes when the dual PO syndrome is calculated. At this time, the controller 200 accesses the address of the buffer memory 210 in the column direction, and outputs the ECC block data stored in the buffer memory 210 by 2 bytes in the column direction.

The dual PO syndrome calculation unit 220 receives the ECC block data output from the buffer memory 210, and the double PO syndrome calculation unit 220 receives the root (α ^O ˜ α ¹⁵ ) of the set generation polynomial. And after calculating the syndrome of the odd-numbered column, respectively.

Here, the configuration of the dual PO syndrome calculation unit 220 will be described in more detail.

First, the data stream W (x) in which no error occurs is represented by Equation 1 below.

W (x) = r _n-1 x ^n-1 + r _n-2 x ^n-2 + ... + r ₁ x ¹ + r ₀ x ⁰

Where r is a symbol in bytes of each data stream and x is the position of the symbol.

When the data stream W (x) is recorded on the optical disk and then read, and assuming that an error is included in the read data stream, the total error e (x) of the data stream may be expressed as in Equation 2 below. Can be.

e (x) = e _n-1 x ^n-1 + e _n-2 x ^n-2 + ... + e ₁ x ¹ + e ₀ x ⁰

Where e is an error.

Therefore, the data stream read from the optical disc, that is, the data stream R (x) containing the error, may be expressed as in Equation 3 below.

R (x) = c (x) + e (x)

r _i = c _i + e _i (0≤i≤n-1)

Here, c (x) is a data stream in which no error is inserted.

Therefore, the root of the set polynomial (α ^{0 to} α ^{n -k-1} ), where n is the number of parities added to the data stream and k is the number of data streams, is assigned to the data stream R (x). Then, it is determined whether the value is 0, and whether or not there is an error can be checked. If it is not 0, it is determined that one or more errors exist and error correction is performed.

In this way, the value obtained by substituting all roots α ^{0 to} α ^{n -k-1} of the generated polynomial is a syndrome, which is the basis for ECC correction.

The operation of the syndrome may be expressed as in Equation 5 below.

Equation 5 may be modified as shown in Equation 6 to facilitate the design of hardware.

When Equation 6 is configured as hardware, the syndrome output from the syndrome output unit 300 is logically multiplied by the root α ^j of the generated polynomial by the logical multiplication means 302 as shown in FIG. 4, and the logical multiplication means 302. Logic summation means 306 for logical summation of the output signal and the data stream R (x) to the logic summation unit 304 and output to the syndrome output unit 300.

When the PO syndrome is calculated by the PO syndrome calculating device, the ECC block data must be read in the column direction from the buffer memory 210. The ECC block data is composed of 208 bytes of block data in the column direction, and there are 16 bytes of parity in the 208 bytes of block data.

In the present invention, when calculating the double PO syndrome, the ECC block data stored in the buffer memory 210 is first read in the column direction as described above. At this time, the buffer memory 210 is capable of writing and reading data by two bytes, and thus, the lower byte data B _0,0 , B _1,0 ,... B _207,0 ) is used to compute the even syndrome of PO syndromes. In addition, the PO syndrome of the odd-numbered column is performed using the data of the upper byte B _0,1 , B _{1,1 ,.} do.

That is, according to the present invention, as illustrated in FIG. 5, the ECC block data read from the buffer memory 210 is different from the lower byte data B _0,0 , B _1,0 ,..., B _207,0 . Byte data (B _0,1 , B _1,1 ,..., B _207,1 ) are divided, and the lower byte data (B _0,0 , B _1,0 ,..., B _{207) , 0} ) is input to the even column PO syndrome calculation unit 221 for calculating the PO syndrome of the even column, and the data (B _0,1 , B _1,1 ,..., _207,1 ) of the upper byte is odd. the second column PO PO odd columns as input to a syndrome calculation unit 223, even columns PO syndrome calculation unit 221 and the odd column PO syndrome calculation unit 223, a lower byte, respectively for calculating the syndrome data (B _0,0, B _{1, 0} ,..., B _207,0 ) and the data of the high byte (B _0,1 , B _1,1 ,..., B _207,1 ) and the root α ^{0 to} α ¹⁵ of the set generation polynomial Calculate column PO syndromes and odd column PO syndromes.

When the dual PO syndrome calculation unit 220 of the present invention is displayed in hardware using the syndrome output unit of FIG. 4, the dual PO syndrome calculation unit 220 is configured as shown in FIG. 6. That is, the even-numbered column PO syndrome calculation unit 221 is configured to logically sum the data of the lower byte and the output signals of the plurality of logical multiplication means 302 by inputting them to the plurality of logical sum means 304, respectively, 304 output signal is input to each of the plurality of the syndrome output unit 300 of the plurality of the syndrome output unit 300 are each syndrome as signals received from a plurality of means 304 the logical sum (S ₀ ~S ₁₅ ) And output.

The syndromes S _{0 to} S ₁₅ outputted by the plurality of syndrome output units 300 are respectively input to the plurality of logical multiplication means 302 and the roots of the generated polynomials α ^{0 to} α ¹⁵ and the logical products respectively. The output signal of the logical multiplication means 302 is inputted to the logical sum means 304 to be logically summed with the data of the lower byte, and then input to the syndrome output part 300 to calculate syndromes S _{0 to} S ₁₅ . Will print.

In addition, the odd-numbered column PO syndrome calculation unit 223 outputs the computed syndromes to the muscle α ⁰ ~α ¹⁵ of the generator polynomial of the upper byte data and the set (S ₀ ~S ₁₅₎ as in the above even-numbered columns PO syndrome arithmetic unit 223 do.

7 is a signal flow diagram illustrating an operation of calculating a double PO syndrome and performing error correction according to the dual PO syndrome calculation method of the present invention. As shown in FIG. 4, the ECC block data is read from the buffer memory 210 in the column direction at step 400.

In the next step 402, the read ECC block data is divided into lower byte data and upper byte data, and in step 404, the even-numbered and odd-numbered data are divided into the lower byte data and the upper byte data, respectively. Compute the PO syndrome of the first column.

In the next step 406, it is determined whether an error has occurred in the data stream by using the calculated value of the PO syndrome, and if the error does not occur as a result of the determination, the error is terminated in step 408. A correction process is performed to correct an error in the data stream.

On the other hand, while the present invention has been shown and described with respect to specific preferred embodiments, various modifications and changes of the present invention without departing from the spirit or field of the invention provided by the claims below It can be easily understood by those skilled in the art. For example, in the above description, the PO syndrome is doublely calculated when performing PO correction, and the present invention is not used only for the calculation of PO syndrome, and the computing device of the present invention is simply used when PI correction is performed. You can use it as it is to calculate the PI syndrome.

As described in detail above, in the present invention, when performing ECC correction of a reproduced data stream, the PO syndrome is calculated in duplicate using all ECC block data read out from the buffer memory two bytes at a time. Access to PO data, which consumes the most bandwidth, can be doubled. In addition, since errors are corrected every two bytes when PI correction → PO correction → PI correction as well as PI correction → PO correction → PI correction → PO correction, error correction is easy and there is an advantage in terms of hardware cost and speed. . In other words, since the PO syndrome is calculated in two columns for the data stream, the buffer memory can be efficiently used without additional fast memory or buffer memory addressing method in the system, thereby improving the access efficiency of the buffer memory, and error correction. The processing can be speeded up.

Claims (5)

A control unit controlling an operation of the PO syndrome; A buffer memory for storing ECC block data in advance and reading the stored data under control of the controller; And And a dual PO syndrome calculation unit configured to receive the ECC block data read from the buffer memory and the root of the generated polynomial, and perform syndromes on each of the even and odd column PO syndromes. The apparatus of claim 1, wherein the dual PO syndrome calculating unit comprises; An even-numbered PO syndrome calculation unit configured to calculate a PO syndrome from the low-byte data and the root of the generated polynomial from the 2-byte ECC block data received from the buffer memory; And And an odd-numbered PO syndrome calculation unit configured to calculate a PO syndrome from the two-byte ECC block data input from the buffer memory and the data of the upper byte and the root of the generated polynomial. 3. The apparatus of claim 2, wherein the even and odd column PO syndrome calculation unit; A dual PO syndrome calculation device, characterized in that for calculating the PO syndrome by the following equation (6). Equation 6 Where S is a PO syndrome, r is a symbol in bytes for each data stream, α is the root of the generation polynomial, and n and j are constants. 3. The apparatus of claim 2, wherein the even and odd column PO syndrome calculation unit; A plurality of syndrome output units for calculating and outputting syndromes; A plurality of logical multiplication means for respectively logically multiplying a plurality of syndromes output from the plurality of syndrome output units and a plurality of roots of a set generation polynomial; And And a logical sum means for logically adding the output signal of the logical multiplication means and the ECC block data to the syndrome output unit. A first step of the control unit reading ECC block data by 2 bytes from the buffer memory; A second step of dividing the two bytes of ECC block data read in the first step into a lower byte and an upper byte; And A double PO comprising a third process of calculating the syndrome by receiving the even-numbered PO PO calculation unit and the odd-numbered PO PO syndrome calculation unit respectively from the lower byte and the upper byte and the root of the generated polynomial set in the second process. Syndrome calculation method.