Data correction device and data correction method thereof
Patent Information
- Application Number
- CN202311757035.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-05-26
- Filing Date
- 2023-12-20
- Publication Date
- 2024-11-26
AI Technical Summary
The prior art In the error correction operation, erasing errors in the storage medium cannot be effectively identified and corrected, especially in the absence of externally provided erasing position information, resulting in correction failure.
By generating defect bitmaps and defect symbol maps, the defect bitmaps and erasing positions in the storage medium are identified by themselves, and the errata correction operation is used to perform errata correction operations, and correction can be performed even without externally providing erasing position information.
Improves the success rate and reliability of error correction, enables identification and correction of errors including erasing in the storage medium, and enhances data integrity and reliability.
Smart Images

Figure CN119025326A_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application claims priority to Korean Patent Application No. 10-2023-0068349, filed on May 26, 2023, the disclosure of which is incorporated herein by reference in its entirety. Technical Field
[0003] Various embodiments of the present disclosure relate to data correction techniques, and more particularly, to erasure or errata correction schemes. Background Art
[0004] Data stored in a storage medium may contain errors due to temporary or permanent reasons and may therefore be damaged. Such operations of detecting and correcting errors are generally referred to as error correction operations. Error correction operations include encoding operations and decoding operations.
[0005] During the encoding operation, the error correction code is added to the original data. A set of original data and error correction code is stored in the storage medium as a codeword.
[0006] The codeword stored in the storage medium is later read from the storage medium. In this case, since the original data included in the codeword has been stored in the storage medium, it may contain errors and may be damaged. During the decoding operation, errors included in the damaged data can be detected and corrected based on the error correction code included in the codeword read from the storage medium, and the original data can be finally restored.
[0007] The error correction code may be defined in various ways according to the error correction scheme, and the error correction capability of the error correction code may also vary. When the number of errors included in the damaged data exceeds the error correction capability of the error correction code, the error correction operation is regarded as a failure. Summary of the invention
[0008] Various embodiments of the present disclosure relate to improved errata correction schemes.
[0009] According to an embodiment of the present disclosure, a method for correcting an errata sheet may include: generating first information about defective bits respectively regarded as corresponding to one or more defects included in a storage cell; generating second information about one or more defect symbols based on the first information; generating third information about an erase position based on the second information; and performing an errata correction operation on a read codeword read from the storage cell based on the third information.
[0010] Generating the first information may include: a first operation of storing a test pattern selected from one or more original test patterns in a storage unit; a second operation of reading a test pattern as a read test pattern from the storage unit; a third operation of detecting bits inconsistent between the selected pattern and the read test pattern as defective bits; and a fourth operation of generating the first information based on one or more defective bits as a result of the third operation.
[0011] The original test patterns may have different values from each other.
[0012] Generating the first information may further include a fifth operation of repeating the first operation to the fourth operation for each original test pattern.
[0013] The first information generated through the fifth operation may include pieces corresponding to respective original test patterns.
[0014] The original test pattern may be random in nature.
[0015] These fragments may be defect bitmaps.
[0016] Each of the defect bitmaps may include information on inconsistent bits between a corresponding test pattern among the original test patterns and a corresponding read test pattern among the read test patterns.
[0017] Each of the original test patterns may have the same size as a read codeword.
[0018] The second information may be generated based on information about defective bits included in each of the defective bitmaps.
[0019] Each of the defect symbols may indicate that one or more symbols among corresponding symbols included in the defect bitmap include a defective bit.
[0020] The second information may be a defect symbol map having the same size as the read codeword.
[0021] The third information may be erasure locator information for considering each of the defective symbols as an erasure.
[0022] The errata correction operation may be an operation of correcting at least one error or at least one erasure included in at least one data symbol included in a read codeword.
[0023] The method may further include storing the corrected codeword in a storage unit when the errata correction operation is successful.
[0024] The method may further include storing the read codeword in a storage unit when the errata correction operation fails.
[0025] The read codeword may be a codeword encoded according to a Reed-Solomon scheme.
[0026] Generating the first information may be performed in response to a failure of an error correction operation performed on the read codeword.
[0027] Generating the first information may be performed in response to a failure of an errata correction operation performed on the read codeword.
[0028] According to an embodiment of the present disclosure, an apparatus for correcting erasure may include: a defect symbol position generating unit configured to generate third information about a position of erasure; and an errata decoding unit configured to perform an errata correction operation on a read codeword read from a storage unit based on the third information. The defect symbol position generating unit may be further configured to generate first information about defective bits respectively regarded as corresponding to one or more defects included in the storage unit, and to generate second information about one or more defective symbols based on the first information. The defect symbol position generating unit may generate third information based on the second information.
[0029] Generating the first information may include: a first operation of storing a test pattern selected from one or more original test patterns in a storage unit; a second operation of reading a test pattern as a read test pattern from the storage unit; a third operation of detecting bits inconsistent between the selected pattern and the read test pattern as defective bits; and a fourth operation of generating the first information based on one or more defective bits as a result of the third operation.
[0030] The original test patterns may have different values from each other.
[0031] Generating the first information may further include a fifth operation of repeating the first operation to the fourth operation for each original test pattern.
[0032] The first information generated through the fifth operation may include segments corresponding to respective original test patterns.
[0033] The original test pattern may be random in nature.
[0034] These fragments may be defect bitmaps.
[0035] Each of the defect bitmaps may include information on inconsistent bits between a corresponding test pattern among the original test patterns and a corresponding read test pattern among the read test patterns.
[0036] Each of the original test patterns may have the same size as a read codeword.
[0037] The defect symbol position generating unit may generate the second information based on information about defective bits included in each of the defect bitmaps.
[0038] Each of the defect symbols may indicate that one or more symbols among corresponding symbols included in the defect bitmap include a defective bit.
[0039] The second information may be a defect symbol map having the same size as the read codeword.
[0040] The third information may be erasure locator information for considering each of the defective symbols as an erasure.
[0041] The errata correction operation may be an operation of correcting at least one error or at least one erasure included in at least one data symbol included in a read codeword.
[0042] The errata decoding unit may be further configured to perform an operation of storing the corrected codeword in the storage unit when the errata correction operation is successful.
[0043] The errata decoding unit may be further configured to perform an operation of storing the read codeword in the storage unit when the errata correction operation fails.
[0044] The read codeword may be a codeword encoded according to a Reed-Solomon scheme.
[0045] The defect symbol position generating unit may generate the first information when an error correction operation performed by the errata decoding unit on the read codeword fails.
[0046] The defect symbol position generating unit may generate the first information when an errata correction operation performed by the errata decoding unit on the read codeword fails.
[0047] According to an embodiment of the present disclosure, a method for correcting erasure may include: generating first information about defective bits that are respectively considered to correspond to one or more defects included in a storage cell; generating second information about one or more defect symbols based on the first information; generating third information about an erase position based on the second information; and performing an erasure correction operation on a read codeword read from the storage cell based on the third information.
[0048] Generating the first information may include: a first operation of storing a pattern selected from one or more original test patterns in a storage unit; a second operation of reading a test pattern as a read test pattern from the storage unit; a third operation of detecting bits inconsistent between the selected pattern and the read test pattern as defective bits; and a fourth operation of generating the first information based on one or more defective bits as a result of the third operation.
[0049] The original test patterns may have different values from each other.
[0050] Generating the first information may further include a fifth operation of repeating the first operation to the fourth operation for each original test pattern.
[0051] The first information generated through the fifth operation may include segments corresponding to respective original test patterns.
[0052] The original test pattern may be random in nature.
[0053] These fragments may be defect bitmaps.
[0054] Each of the defect bitmaps may include information on inconsistent bits between a corresponding test pattern among the original test patterns and a corresponding read test pattern among the read test patterns.
[0055] Each of the original test patterns may have the same size as a read codeword.
[0056] The second information may be generated based on information about defective bits included in each of the defective bitmaps.
[0057] Each of the defect symbols may indicate that one or more symbols among corresponding symbols included in the defect bitmap include a defective bit.
[0058] The second information may be a defect symbol map having the same size as the read codeword.
[0059] The third information may be erasure locator information for considering each of the defective symbols as an erasure.
[0060] The erasure correction operation may be an operation of correcting at least one erasure included in at least one data symbol included in the read codeword.
[0061] The method may further include storing the corrected codeword in the storage unit when the erasure correction operation is successful.
[0062] The method may further include storing the read codeword in the memory cell when the erase correction operation fails.
[0063] The read codeword may be a codeword encoded according to a Reed-Solomon scheme.
[0064] Generating the first information may be performed in response to a failure of an error correction operation performed on the read codeword.
[0065] Generating the first information may be performed in response to a failure of an erasure correction operation performed on the read codeword.
[0066] According to an embodiment of the present disclosure, an apparatus for correcting erasure may include: a defect symbol position generating unit configured to generate third information about the position of erasure; and an erasure decoding unit configured to perform an erasure correction operation on a read codeword read from a storage unit based on the third information.
[0067] The defect symbol position generating unit may be further configured to generate first information about defective bits respectively regarded as corresponding to one or more defects included in the storage unit, and generate second information about one or more defective symbols based on the first information, and the defect symbol position generating unit may generate third information based on the second information.
[0068] Generating the first information may include: a first operation of storing a pattern selected from one or more original test patterns in a storage unit; a second operation of reading a test pattern as a read test pattern from the storage unit; a third operation of detecting bits inconsistent between the selected pattern and the read test pattern as defective bits; and a fourth operation of generating the first information based on one or more defective bits as a result of the third operation.
[0069] The original test patterns may have different values from each other.
[0070] Generating the first information may further include a fifth operation of repeating the first operation to the fourth operation for each original test pattern.
[0071] The first information generated through the fifth operation may include segments corresponding to respective original test patterns.
[0072] The original test pattern may be random in nature.
[0073] These fragments may be defect bitmaps.
[0074] Each of the defect bitmaps may include information on inconsistent bits between a corresponding test pattern among the original test patterns and a corresponding read test pattern among the read test patterns.
[0075] Each of the original test patterns may have the same size as a read codeword.
[0076] The defect symbol position generating unit may generate the second information based on information about defective bits included in each of the defect bitmaps.
[0077] Each of the defect symbols may indicate that one or more symbols among corresponding symbols included in the defect bitmap include a defective bit.
[0078] The second information may be a defect symbol map having the same size as the read codeword.
[0079] The third information may be erasure locator information for considering each of the defective symbols as an erasure.
[0080] The erasure correction operation may be an operation of correcting at least one erasure included in at least one data symbol included in the read codeword.
[0081] The erasure decoding unit may be further configured to perform an operation of storing the corrected codeword in the storage unit when the erasure correction operation is successful.
[0082] The erasure decoding unit may be further configured to perform an operation of storing the read codeword in the storage unit when the erasure correction operation fails.
[0083] The read codeword may be a codeword encoded according to a Reed-Solomon scheme.
[0084] The defect symbol position generating unit may generate the first information when an error correction operation performed by the erasure decoding unit on the read codeword fails.
[0085] The defect symbol position generating unit may generate the first information when an erasure correction operation performed by the erasure decoding unit on the read codeword fails. BRIEF DESCRIPTION OF THE DRAWINGS
[0086] Figure 1 is a diagram illustrating a Reed-Solomon (RS) decoding scheme.
[0087] Figure 2 is a diagram illustrating a scheme for identifying an erasure position according to an embodiment of the present disclosure.
[0088] Figure 3 is a block diagram illustrating an errata correction apparatus according to an embodiment of the present disclosure.
[0089] Figure 4 is a flow chart illustrating generation of defect symbol location information according to an embodiment of the present disclosure.
[0090] Figure 5 is a diagram illustrating a memory system according to an embodiment of the present disclosure. DETAILED DESCRIPTION
[0091] In the specification, known parts irrelevant to the essence of the present disclosure can be omitted.It should be understood that when an element is referred to as "connected to" or "coupled to" another element, the element can be directly connected to or coupled to another element, or electrically connected to or coupled to another element by inserting one or more elements therebetween.In the description of the entire specification, it should be understood that when information or data is provided from one element to another element, information or data can be provided directly or indirectly between the elements inserted therebetween.In addition, it should also be understood that, unless otherwise stated, the terms "comprising", "comprising", "including" and "including" do not exclude the existence of one or more other elements when used in this specification, but may further include or have one or more other elements.In the specification, some components are described in the singular, but the present disclosure is not limited thereto, and it should be understood that these components can be formed in the plural form.
[0092] Figure 1 is a diagram illustrating a Reed-Solomon (RS) decoding scheme.
[0093] Figure 1 A codeword consisting of "n" symbols including "k" data symbols and "2t" parity symbols is shown.
[0094] According to the conventional RS decoding scheme, the errors of the maximum "t" data symbols included in the codeword can be corrected. That is, according to the conventional RS decoding scheme, the position of a single data symbol (i.e., an error symbol) containing one or more errors can be detected from the codeword, and these errors can be statistically corrected by two parity symbols. The conventional RS decoding scheme includes a process of calculating the syndrome of the codeword, a process of calculating an error location polynomial and an error evaluation polynomial corresponding to the position and size of the error in the codeword, and a process of calculating the position and size of the error by using the error location polynomial and the error evaluation polynomial.
[0095] In this specification, an RS decoding operation according to a conventional RS decoding scheme is defined as an error decoding operation.
[0096] An RS decoding scheme that is more advanced than the conventional RS decoding scheme is an errata-corrected RS decoding scheme. Here, errata includes erasures and errors in the conventional sense. In addition to errors, data stored in a storage medium may also include erasures having a logical state that cannot be determined as a value of zero (0) or one (1) due to various reasons. Since erasures do not have a fixed value, erasures are targets to be corrected to have a fixed value. According to the errata-corrected RS decoding scheme, errata within a data symbol including an erasure can be statistically corrected by a single parity symbol. In this article, information about the location of the erasure (hereinafter referred to as erasure location information) is provided from an external device. As Figure 1 As shown, according to the errata-corrected RS decoding scheme, errors and erasures of a maximum of "2e+p (≤2t)" errata symbols included in a codeword can be corrected by "2t" parity symbols. In this article, "e" is the number of erroneous symbols whose positions are unknown, and "p" is the number of erasure symbols.
[0097] In this specification, an RS decoding operation according to an errata-corrected RS decoding scheme is defined as an errata decoding operation.
[0098] The errata-corrected RS decoding scheme is disclosed in the paper “NEW SCALABLE DECODER ARCHITECTURES FOR REED-SOLOMON CODES” published in Volume 63 of IEEE TRANSACTIONS ON COMMUNICATIONS in August 2015 and in U.S. Patent No. 8,347,193 “PARALLEL INVERSIONLESS ERROR AND ERASURE PROCESSING”. According to the aforementioned documents, the RS decoder can correct errors of a maximum of “t” data symbols included in a codeword according to a conventional RS decoding scheme when an external device does not provide erasure position information, and can correct errors and erasures of a maximum of “2e+p (≤2t)” errata symbols included in a codeword according to the errata-corrected RS decoding scheme when an external device provides erasure position information. According to the document, an erasure locator corresponding to erasure position information may be generated according to the erasure position information provided by an external device, and erasure may be corrected based on the erasure locator.
[0099] According to the document, when the erasure position information is not provided from an external device, the RS decoder may not perform an errata decoding operation, but may only perform an error decoding operation. That is, unless the erasure position information is provided from an external device, the RS decoder will not perform erasure correction. According to the document, the RS decoder may perform erasure correction only when the erasure position information is provided from an external device. That is, according to the document, the erasure correction of the RS decoder depends only on the erasure position information provided from an external device and predetermined. Therefore, even if the erasure position information has been predetermined for any reason, when a new physical defect appears while the storage medium is operated later, the predetermined erasure position information becomes meaningless, and new erasure position information according to the new physical defect is required.
[0100] According to an embodiment of the present disclosure, in an errata decoding operation, an erasure correction device and an erasure correction method can be provided, in which an erasure position can be identified even if erasure position information is not provided from an external device.
[0101] Figure 2 is a diagram illustrating a scheme for identifying an erasure position according to an embodiment of the present disclosure.
[0102] The errata correction device according to an embodiment of the present disclosure may write an original test pattern into a specific storage unit included in a storage medium, and may store the original test pattern in the storage unit. In an embodiment, the errata correction device according to an embodiment of the present disclosure may perform an error decoding operation on a codeword read from a storage unit according to a conventional RS decoding scheme or an errata-corrected RS decoding scheme, and may store the original test pattern in the storage unit when error correction fails as a result of the error decoding operation. In this case, the codewords already stored in the storage unit may be buffered separately in a buffer, and if necessary, the buffered codewords may be stored in the storage unit again at a later time.
[0103] The original test pattern can have any value as long as it has randomness. For example, all bits included in the original test pattern can each have a value of zero (0) or one (1). For example, among the bits included in the original test pattern, the bits corresponding to the adjacent storage cells can have a flipped value. For example, the original test pattern can have a value that can cause a data transition between bit lines connected to the storage medium. For example, the original test pattern can have an inversion value of another original test pattern. For example, the read codewords RD-CDWD and their inversion codewords to be described later can each be used as the original test pattern. The original test pattern can have a size corresponding to the codeword. Due to the following reasons, it is not necessary to perform error correction code (ECC) encoding and decoding operations on the original test pattern, and therefore, the original test pattern does not need to include a parity symbol.
[0104] The test pattern stored in the storage unit can be read as a read test pattern. The errata correction device according to an embodiment of the present disclosure does not perform an ECC decoding operation on the read test pattern. The errata correction device according to an embodiment of the present disclosure can compare the read test pattern with the original test pattern bit by bit.
[0105] When the comparison result indicates a defective bit, i.e., an inconsistent bit between the read test pattern and the original test pattern, it can be considered that a physical defect has occurred in the position corresponding to the defective bit in the memory cell. The process of detecting the defective bit by comparing the original test pattern with the read test pattern can be repeated multiple times. The above process can be repeatedly performed while randomly changing the value of the original test pattern. As the number of repetitions increases, the reliability of the process of detecting physical defects increases.
[0106] The errata correction device according to an embodiment of the present disclosure can generate information about one or more defective bits identified by repetition. For example, the errata correction device can generate information about defective bits that are identified a predetermined threshold number of times or more during repetition. The threshold number of times is at least once. In addition, the conditions for the defective bits generated as information can be set in various ways.
[0107] In an embodiment, information about defective bits may be generated in the form of a bitmap. The defective bitmap may have a size corresponding to a codeword. Figure 2 As shown, the errata correction apparatus according to the embodiment of the present disclosure may generate one or more defect bit maps as information about defective bits. Figure 2 Defect bitmaps corresponding to the repetitions are shown. Each of the defect bitmaps may have a codeword size, and Figure 2 Defective bits identified by each of the repetitions respectively corresponding to the defective bitmaps are shown.
[0108] The errata correction device according to an embodiment of the present disclosure may generate information about one or more defective symbols based on information about one or more defective bits identified by repetition, for example, based on a defective bitmap. For example, the information about the defective symbol is a single defective symbol map. The defective symbol indicates that one or more symbols among the corresponding symbols included in the defective bitmap include a defective bit. The errata correction device according to an embodiment of the present disclosure may generate a defective symbol map including all defective symbols included in the defective bitmap by performing an OR operation on the defective bitmap symbol by symbol. The defective symbol map has the same size as the defective bitmap, and Figure 2 Shows the defect symbols in the defect symbol diagram.
[0109] According to an embodiment of the present disclosure, a defect symbol generated as information can be regarded as an erasure, and the information about the defect symbol can be used to generate erasure position information. That is, in an errata decoding operation, an errata correction device according to an embodiment of the present disclosure can identify an erasure position even if the erasure position information is not received from an external device. Figure 2 An erase locator corresponding to the erase position information is shown.
[0110] The errata correction device according to an embodiment of the present disclosure can perform an errata decoding operation based on the erasure position information obtained by itself, that is, when no erasure position information is provided by an external device.
[0111] In an embodiment, when error correction fails as a result of performing an error decoding operation on a codeword read from a storage unit according to a conventional RS decoding scheme, as described above, the errata correction device according to an embodiment of the present disclosure can obtain erasure position information by itself, and can perform an errata decoding operation on the codeword for which error correction failed based on the erasure position information obtained by itself.
[0112] In an embodiment, when the errata correction fails as a result of performing an errata decoding operation on a codeword read from a storage unit according to an errata correction RS decoding scheme, as described above, the errata correction device according to an embodiment of the present disclosure can obtain the erasure position information by itself, and can perform the errata decoding operation again on the codeword for which the error correction failed based on the self-obtained erasure position information. In the present embodiment, the previous errata decoding operation can be performed based on the previous erasure position information as information previously provided by an external device or previously obtained by itself. In the present embodiment, the current errata decoding operation can be performed based on the current erasure position information as information currently obtained by itself. That is, according to the present disclosure, the erasure position information reflecting the current state of the storage medium can be used for the errata decoding operation.
[0113] Figure 3 is a block diagram showing an errata correction apparatus 301 according to an embodiment of the present disclosure. Figure 3 The errata correction device 301 may include an errata decoding unit 311, a buffer 313, and a defect symbol location information generating unit 315. The errata decoding unit 311, the defect symbol location information generating unit 315, and the storage unit 303 include all circuits, systems, software, firmware, and devices required for their respective operations and functions.
[0114] The errata decoding unit 311 may read a codeword from the storage unit 303 included in the storage medium as a read codeword RD-CDWD. The read codeword RD-CDWD may have the same Figure 1 That is, the read codeword RD-CDWD may be composed of "n" symbols including "k" data symbols and "2t" parity symbols.
[0115] In an embodiment, the errata decoding unit 311 may buffer the read codeword RD-CDWD in the buffer 313 and may store the buffered read codeword RD-CDWD in the storage unit 303 again later if necessary.
[0116] The errata decoding unit 311 may perform an ECC decoding operation on the read codeword RD-CDWD. In an embodiment, when there is no externally provided or previously obtained erasure position information, the errata decoding unit 311 may perform an error decoding operation on the read codeword RD-CDWD according to a conventional RS decoding scheme. In an embodiment, when there is externally provided or previously obtained erasure position information, the errata decoding unit 311 may perform an errata decoding operation on the read codeword RD-CDWD according to an errata-corrected RS decoding scheme.
[0117] When the error decoding operation or the errata decoding operation fails, the data decoding unit 311 may provide information INFO about the failure to the defect symbol location information generating unit 315. The information INFO may include address information of the storage unit 303 from which the read codeword RD-CDWD has been read.
[0118] The defect symbol location information generating unit 315 may generate defect symbol location information in response to the information INFO.
[0119] Figure 4 is a flow chart illustrating generation of defect symbol location information according to an embodiment of the present disclosure.
[0120] Reference Figure 3 and Figure 4 In operation S401, the defect symbol location information generating unit 315 may generate information about defective bits in response to the information INFO. In an embodiment, the defect symbol location information generating unit 315 may generate "N" defect bitmaps as information about defective bits. The "N" defect bitmaps correspond to the "N" original test patterns PTRN, respectively.
[0121] Operation S401 includes operations S411 to S417. The defect symbol location information generating unit 315 may generate defect bitmaps corresponding to "N" original test patterns PTRN, respectively, by repeating operations S411 to S417. The defect symbol location information generating unit 315 may generate "N" defect bitmaps by sequentially repeating operation S401 "N" times for separate "N" original test patterns PTRN. The "N" original test patterns PTRN may have randomness and may have values different from each other. For example, the read codeword RD-CDWD and its inverse codeword may be used as the original test pattern.
[0122] In operation S411, the defect symbol location information generating unit 315 may store an original test pattern PTRNx (1≤'x'≤'N') selected from 'N' original test patterns PTRN in the storage unit 303. The defect symbol location information generating unit 315 may identify the storage unit 303 according to the information INFO provided from the errata decoding unit 311.
[0123] In operation S413, the defect symbol location information generating unit 315 may read the original test pattern PTRNx as the read test pattern from the storage unit 303. The original test pattern PTRNx does not need to include a parity symbol, and an ECC decoding operation does not need to be performed on the read test pattern read by the defect symbol location information generating unit 315.
[0124] In operation S415, the defect symbol location information generating unit 315 may compare the original test pattern PTRNx with the read test pattern bit by bit. When the result of the bit by bit comparison indicates a defective bit or an inconsistent bit between the read test pattern and the original test pattern PTRNx, it may be considered that a physical defect has occurred in the position corresponding to the defective bit in the memory cell 303.
[0125] In operation S417, the defect symbol location information generating unit 315 may generate a defect symbol location information as shown in the reference image based on the one or more defect bits detected in operation S415. Figure 2 Described defect bitmap BITMAPx.
[0126] When operation S401 is sequentially repeated for the separate “N” original test patterns PTRN, “N” defect bitmaps may be generated eventually.
[0127] In operation S403, the defect symbol location information generating unit 315 may generate information about one or more defect symbols based on the "N" defect bitmaps generated in operation S401. For example, the information about the defect symbol is a single defect symbol map. Figure 2 As described, the defective symbol indicates that one or more symbols among corresponding symbols included in the "N" defective bitmaps include a defective bit.
[0128] In operation S405, the defect symbol location information generating unit 315 may generate erasure location information ERASURE based on the defect symbol map generated in operation S403. For example, the erasure location information ERASURE is an erasure locator. Figure 2 As described, defect symbols included in the defect symbol map may be regarded as erasures, and erasure position information may be generated using information on the defect symbols.
[0129] Return to reference Figure 3, the defect symbol position information generating unit 315 may provide the erasure position information ERASURE generated in operation S405 to the errata decoding unit 311 .
[0130] The errata decoding unit 311 may perform an errata decoding operation on the read codeword RD-CDWD based on the erasure location information ERASURE provided by the defect symbol location information generating unit 315 .
[0131] When the errata decoding operation based on the erasure position information ERASURE provided from the defect symbol position information generating unit 315 fails, the errata decoding unit 311 may store the read codeword RD-CDWD that has been buffered in the buffer 313 in the storage unit 303 again.
[0132] When the errata decoding operation based on the erasure position information ERASURE provided from the defect symbol position information generating unit 315 is successful, the errata decoding unit 311 may store the corrected codeword EC-CDWD in the storage unit 303 .
[0133] According to an embodiment of the present disclosure, the errata correction device 301 can obtain the erasure position information ERASURE by itself, and can perform the errata decoding operation on the read codeword RD-CDWD that fails error correction based on the erasure position information ERASURE according to the errata correction RS decoding scheme.
[0134] Figure 5 is a diagram illustrating a memory system 500 according to an embodiment of the present disclosure.
[0135] Reference Figure 5 The memory system 500 may store therein data provided from an external device, and may provide data previously stored therein to the external device. The memory system 500 may include a memory controller 510 and a memory device 520. The memory controller 510 may control the operation of the memory device 520.
[0136] In response to the control of the memory controller 510, the memory device 520 may store therein data provided from the memory controller 510 or provide the memory controller 510 with data stored therein. The memory device 520 may include a storage medium, and the storage medium may include Figure 3 The storage unit 303 shown. A representative storage medium may be a DRAM, but the present disclosure is not limited thereto. According to an embodiment, as shown in FIG. Figure 3 to Figure 4 As described, the read codeword RD-CDWD can be read from the storage unit 303. According to an embodiment, as shown in FIG. Figure 3 to Figure 4As described above, the original test pattern PTRN may be stored in the storage unit 303 and then read as a read test pattern. Figure 3 to Figure 4 As described, the read codeword RD-CDWD may be stored in the storage unit 303 again, or the corrected codeword EC-CDWD may be stored in the storage unit 303 .
[0137] The memory controller 510 may control the memory device 520 and may write / read data into / from the memory device 520 . The memory controller 510 may include an ECC engine. The ECC engine may include an ECC encoder 501 and an ECC decoder 301 .
[0138] The ECC decoder 301 may perform an ECC decoding operation on the data read from the storage unit 303 as the read codeword RD-CDWD. Figure 3 and Figure 4 According to an embodiment, the ECC decoding operation may be an operation of correcting an error or errata included in the read codeword RD-CDWD according to at least one of the above-mentioned error correction scheme and errata correction scheme.
[0139] The ECC encoder 501 may perform an ECC encoding operation on the data to be stored in the storage unit 303. The ECC encoding operation may be generated according to any ECC encoding scheme. Figure 1 The operation of decoding the codeword shown and generating a parity symbol for the data to be stored in the storage unit 303 is irrelevant in type as long as it can perform an ECC decoding operation. According to an embodiment, the ECC encoding scheme may be a RS encoding scheme.
[0140] According to embodiments of the present disclosure, an improved erasure or errata correction scheme may be provided.
[0141] Although the present disclosure has been shown and described with respect to specific embodiments, the disclosed embodiments are provided for purposes of description and are not intended to be limiting. Further, it is noted that the present disclosure may be implemented in various ways through substitutions, changes, and modifications that fall within the scope of the appended claims, as will be appreciated by those skilled in the art in light of the present disclosure. Furthermore, the embodiments may be combined to form additional embodiments.
Claims
1. A method for correcting erasure, the method comprising: generating first information about defective bits respectively regarded as corresponding to one or more defects included in a memory cell; generating second information about one or more defect symbols based on the first information; generating third information about an erasure position based on the second information; as well as An erasure correction operation is performed on a read codeword read from the memory cell based on the third information.
2. The method according to claim 1, wherein generating the first information comprises: a first operation of storing a test pattern selected from one or more original test patterns in the storage unit; reading the test pattern from the storage unit as a second operation of reading the test pattern; a third operation of detecting inconsistent bits between the selected pattern and the read test pattern as defective bits; as well as A fourth operation of generating the first information based on one or more defective bits as a result of the third operation.
3. The method according to claim 2, wherein: The original test patterns have different values from each other, Generating the first information further includes a fifth operation of repeating the first operation to the fourth operation for each original test pattern, and The first information generated by the fifth operation includes segments corresponding to the respective original test patterns.
4. The method according to claim 2, wherein: The original test pattern is random.
5. The method according to claim 3, wherein: The fragment is a defect bitmap, and Each of the defect bitmaps includes information on inconsistent bits between a corresponding one of the original test patterns and a corresponding one of the read test patterns.
6. The method according to claim 5, wherein: Each of the original test patterns has the same size as the read codeword.
7. The method according to claim 6, wherein: The second information is generated based on information about defective bits included in each of the defective bitmaps.
8. The method according to claim 7, wherein: Each of the defect symbols indicates that one or more symbols among corresponding symbols included in the defect bitmap include the defect bit.
9. The method according to claim 1, wherein: The second information is a defect symbol map having the same size as the read codeword.
10. The method according to claim 1, wherein: The third information is erasure locator information for regarding each of the defective symbols as an erasure.
11. The method according to claim 10, wherein: The erasure correction operation is an operation of correcting at least one erasure included in at least one data symbol included in the read codeword.
12. The method according to claim 11, further comprising: When the erasure correction operation is successful, the corrected codeword is stored in the storage unit.
13. The method according to claim 11, further comprising: When the erasure correction operation fails, the read codeword is stored in the storage unit.
14. The method according to claim 1, wherein: The read codeword is a codeword encoded according to the Reed-Solomon scheme.
15. The method according to claim 1, wherein: Generating the first information is performed in response to a failure of an error correction operation performed on the read codeword.
16. The method according to claim 1, wherein: Generating the first information is performed in response to a failure of an erasure correction operation performed on the read codeword.
17. A device for correcting erasure, the device comprising: a defect symbol position generating unit, generating third information about an erasure position; as well as an erasure decoding unit, performing an erasure operation on a read codeword read from a storage unit based on the third information, Wherein, the defect symbol position generating unit further: generating first information about defective bits respectively regarded as corresponding to one or more defects included in the memory cell, and generating second information about one or more defect symbols based on the first information, and The defect symbol position generating unit generates the third information based on the second information.
18. The device according to claim 17, wherein: Generating the first information includes: a first operation of storing a test pattern selected from one or more original test patterns in the storage unit, reading the test pattern from the storage unit as a second operation of reading the test pattern, a third operation of detecting inconsistent bits between the selected pattern and the read test pattern as defective bits, and A fourth operation of generating the first information based on one or more defective bits as a result of the third operation.
19. The device according to claim 18, wherein: The original test patterns have different values from each other, and Generating the first information further includes a fifth operation of repeating the first operation to the fourth operation for each original test pattern, and The first information generated by the fifth operation includes segments corresponding to the respective original test patterns.
20. The device according to claim 18, wherein The original test pattern is random.
21. The device according to claim 19, wherein The fragment is a defect bitmap, and Each of the defect bitmaps includes information on inconsistent bits between a corresponding one of the original test patterns and a corresponding one of the read test patterns.
22. The device according to claim 21, wherein Each of the original test patterns has the same size as the read codeword.
23. The device according to claim 22, wherein: The defect symbol position generating unit generates the second information based on information about defective bits included in each of the defect bitmaps.
24. The device according to claim 23, wherein: Each of the defect symbols indicates that one or more symbols among corresponding symbols included in the defect bitmap include the defect bit.
25. The device according to claim 17, wherein: The second information is a defect symbol map having the same size as the read codeword.
26. The device according to claim 17, wherein: The third information is erasure locator information for regarding each of the defective symbols as an erasure.
27. The device according to claim 26, wherein: The erasure correction operation is an operation of correcting at least one erasure included in at least one data symbol included in the read codeword.
28. The device according to claim 27, wherein The erasure decoding unit further performs an operation of storing the corrected codeword in the storage unit when the erasure correction operation is successful.
29. The device according to claim 27, wherein: The erasure decoding unit further performs an operation of storing the read codeword in the storage unit when the erasure correction operation fails.
30. The apparatus of claim 17, wherein: The read codeword is a codeword encoded according to the Reed-Solomon scheme.
31. The apparatus of claim 17, wherein: The defect symbol position generating unit generates the first information when an error correction operation performed by the erasure decoding unit on the read codeword fails.
32. The apparatus of claim 17, wherein: The defect symbol position generating unit generates the first information when an erasure correction operation performed by the erasure decoding unit on the read codeword fails.
