U.S. patent application number 17/206632 was filed with the patent office on 2022-05-05 for reed-solomon code soft-decision decoding method and device.
The applicant listed for this patent is Korea University Research and Business Foundation, SK hynix Inc.. Invention is credited to Jin Ho JEONG, Jong Sun PARK, Dong Yeob SHIN, Won Gyu SHIN.
Application Number | 20220140845 17/206632 |
Document ID | / |
Family ID | |
Filed Date | 2022-05-05 |
United States Patent
Application |
20220140845 |
Kind Code |
A1 |
SHIN; Won Gyu ; et
al. |
May 5, 2022 |
REED-SOLOMON CODE SOFT-DECISION DECODING METHOD AND DEVICE
Abstract
Disclosed is an erasure-based Reed-Solomon code soft-decision
decoding method and device, capable of reducing a decoding time
while minimizing the effect on error correction performance. The
Reed-Solomon code soft-decision decoding device includes an erasure
control circuit configured to determine whether a number of errors
in a codeword is odd or even, and to provide a key equation solver
circuit with a first erasure pattern or a second erasure pattern
according to a result of the determining when a decoding failure is
detected by a decoding error detection circuit, the first erasure
pattern being provided when the number of errors is odd, the second
erasure pattern being provided when the number of errors is
even.
Inventors: |
SHIN; Won Gyu; (Seoul,
KR) ; PARK; Jong Sun; (Seoul, KR) ; SHIN; Dong
Yeob; (Seoul, KR) ; JEONG; Jin Ho; (Seoul,
KR) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
SK hynix Inc.
Korea University Research and Business Foundation |
Icheon
Seoul |
|
KR
KR |
|
|
Appl. No.: |
17/206632 |
Filed: |
March 19, 2021 |
International
Class: |
H03M 13/45 20060101
H03M013/45; H03M 13/15 20060101 H03M013/15; H03M 13/37 20060101
H03M013/37 |
Foreign Application Data
Date |
Code |
Application Number |
Oct 29, 2020 |
KR |
10-2020-0141950 |
Claims
1. A Reed-Solomon code soft-decision decoding device, comprising: a
syndrome generation circuit configured to generate a syndrome from
a codeword; a key equation solver circuit configured to generate an
error location polynomial and an error evaluation polynomial based
on the syndrome; a Chien-search & error-evaluator circuit
configured to determine a location and magnitude of an error in the
codeword based on the error location polynomial and the error
evaluation polynomial; a decoding error detection circuit
configured to determine whether decoding on the codeword is
successful or failed based on the location and magnitude of the
error; and an erasure control circuit configured to determine
whether a number of errors in the codeword is odd or even, and to
provide the key equation solver circuit with a first erasure
pattern or a second erasure pattern according to a result of the
determining when a decoding failure is detected by the decoding
error detection circuit, the first erasure pattern being provided
when the number of errors is odd, the second erasure pattern being
provided when the number of errors is even.
2. The Reed-Solomon code soft-decision decoding device of claim 1,
wherein the erasure control circuit receives the syndrome from the
syndrome generation circuit and determines whether the number of
errors in the codeword is odd or even by using characteristics of
the syndrome.
3. The Reed-Solomon code soft-decision decoding device of claim 2,
wherein, when a number of bits each having a value of 1 in the
syndrome is odd, the erasure control circuit determines the number
of errors to be odd, and when the number of bits each having the
value of 1 in the syndrome is even, the erasure control circuit
determines the number of errors to be even.
4. The Reed-Solomon code soft-decision decoding device of claim 1,
wherein, when the number of errors is determined to be odd and the
decoding failure is detected, the erasure control circuit provides,
as the first erasure pattern, 2+4j erasure patterns to the key
equation solver circuit, j being an integer equal to or more than
0.
5. The Reed-Solomon code soft-decision decoding device of claim 4,
wherein, when the number of errors is determined to be even and the
decoding failure is detected, the erasure control circuit provides,
as the second erasure pattern, 4k erasure patterns to the key
equation solver circuit, k being a positive integer.
6. The Reed-Solomon code soft-decision decoding device of claim 5,
wherein the erasure control circuit provides the key equation
solver circuit with the 2+4j erasure patterns or the 4k erasure
patterns in an order in which a symbol with low reliability among
symbols in the codeword is preferentially erased.
7. The Reed-Solomon code soft-decision decoding device of claim 1,
wherein the key equation solver circuit calculates the error
location polynomial and the error evaluation polynomial based on
the syndrome and 2+4j erasure patterns that are the first erasure
pattern or based on the syndrome and 4k erasure patterns that are
the second erasure pattern.
8. A Reed-Solomon code soft-decision decoding method, comprising:
generating a syndrome from a codeword; generating an error location
polynomial and an error evaluation polynomial based on the
syndrome; determining a location and magnitude of an error in the
codeword based on the error location polynomial and the error
evaluation polynomial; determining whether decoding on the codeword
is successful or failed based on the location and magnitude of the
error; and determining whether a number of errors in the codeword
is odd or even, and performing decoding on the codeword based on
the syndrome and a first erasure pattern or based on the syndrome
and a second erasure pattern according to a result of the
determining when a decoding failure is detected, the first erasure
pattern being provided when the number of errors is odd, the second
erasure pattern being provided when the number of errors is
even.
9. The Reed-Solomon code soft-decision decoding method of claim 8,
wherein, in determining whether the number of errors in the
codeword is odd or even, a number of symbol errors in the syndrome
is determined to be odd or even by using characteristics of the
syndrome.
10. The Reed-Solomon code soft-decision decoding method of claim 9,
wherein, in determining whether the number of errors in the
codeword is odd or even, when a number of bits each having a value
of 1 in the syndrome is odd, the number of symbol errors is
determined to be odd, and when the number of bits each having the
value of 1 in the syndrome is even, the number of symbol errors is
determined to be even.
11. The Reed-Solomon code soft-decision decoding method of claim 8,
wherein performing the decoding based on the syndrome and the first
erasure pattern comprises: applying, as the first erasure pattern,
2+4j erasure patterns to the codeword.
12. The Reed-Solomon code soft-decision decoding method of claim
11, wherein performing the decoding based on the syndrome and the
second erasure pattern comprises: applying, as the second erasure
pattern, 4k erasure patterns to the codeword.
13. The Reed-Solomon code soft-decision decoding method of claim
12, wherein the 2+4j erasure patterns or the 4k erasure patterns
are applied to the codeword in an order in which a symbol with low
reliability among symbols in the codeword is preferentially
erased.
14. A semiconductor device, comprising: a memory cell array
configured to store a codeword; and a Reed-Solomon code
soft-decision decoding device configured to read the codeword from
the memory cell array, decode the codeword, and output
error-corrected data corresponding to the codeword, wherein the
Reed-Solomon code soft-decision decoding device comprises: a
syndrome generation circuit configured to generate a syndrome from
the codeword; a key equation solver circuit configured to generate
an error location polynomial and an error evaluation polynomial
based on the syndrome; a Chien-search & error-evaluator circuit
configured to determine a location and magnitude of an error in the
codeword based on the error location polynomial and the error
evaluation polynomial; a decoding error detection circuit
configured to determine whether decoding on the codeword is
successful or failed based on the location and magnitude of the
error; and an erasure control circuit configured to determine
whether a number of errors in the codeword is odd or even, and to
provide the key equation solver circuit with a first erasure
pattern or a second erasure pattern according to a result of the
determining when a decoding failure is detected by the decoding
error detection circuit, the first erasure pattern being provided
when the number of errors is odd, the second erasure pattern being
provided when the number of errors is even.
15. The semiconductor device of claim 14, wherein the erasure
control circuit receives the syndrome from the syndrome generation
circuit and determines whether the number of errors in the codeword
is odd or even by using characteristics of the syndrome.
16. The semiconductor device of claim 15, wherein, when a number of
bits each having a value of 1 in the syndrome is odd, the erasure
control circuit determines the number of errors to be odd, and when
the number of bits each having the value of 1 in the syndrome is
even, the erasure control circuit determines the number of errors
to be even.
17. The semiconductor device of claim 14, wherein, when the number
of errors is determined to be odd and the decoding failure is
detected, the erasure control circuit provides, as the first
erasure pattern, 2+4j erasure patterns to the key equation solver
circuit.
18. The semiconductor device of claim 17, wherein, when the number
of errors is determined to be even and the decoding failure is
detected, the erasure control circuit provides, as the second
erasure pattern, 4k erasure patterns to the key equation solver
circuit.
19. The semiconductor device of claim 18, wherein the erasure
control circuit provides the key equation solver circuit with the
2+4j (erasure patterns or the 4k erasure patterns in an order in
which a symbol with low reliability among symbols in the codeword
is preferentially erased.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This application claims priority under 35 U.S.C. .sctn. 119
to Korean Patent Application No. 10-2020-0141950 filed on Oct. 29,
2020, which is incorporated herein by reference in its
entirety.
BACKGROUND
1. Field
[0002] Exemplary embodiments relate to a Reed-Solomon code
soft-decision decoding method and device, and more particularly, to
an erasure-based Reed-Solomon code soft-decision decoding method
and device, capable of reducing a decoding time.
2. Discussion of the Related Art
[0003] Reed-Solomon code soft-decision decoding includes successive
error-and-erasure decoding (SED). The SED iterates decoding while
changing erasure patterns according to an order decided by an
algorithm. Such a decoding method may expect the improvement in
error correction performance as the number of iterations increases,
but has a disadvantage in that a decoding time is greatly
increased.
[0004] Particularly, the decoding time of the SED may be limited
depending on applications. For example, a phase change memory (PCM)
requires a short decoding time. However, when the number of
iterations of decoding is reduced in order to reduce the decoding
time of the SED, the error correction performance may be
degraded.
[0005] Accordingly, there is a need for a technology capable of
reducing the decoding time while minimizing the degradation of the
error correction performance of the SED.
SUMMARY
[0006] Various embodiments are directed to providing an
erasure-based Reed-Solomon code soft-decision decoding method and
device, capable of reducing a decoding time while minimizing the
degradation of error correction performance.
[0007] In an embodiment, a Reed-Solomon code soft-decision decoding
device may include: a syndrome generation circuit configured to
generate a syndrome from a codeword; a key equation solver circuit
configured to generate an error location polynomial and an error
evaluation polynomial based on the syndrome; a Chien-search &
error-evaluator circuit configured to determine a location and
magnitude of an error in the codeword based on the error location
polynomial and the error evaluation polynomial; a decoding error
detection circuit configured to determine whether decoding on the
codeword is successful or failed based on the location and
magnitude of the error; and an erasure control circuit configured
to determine whether a number of errors in the codeword is odd or
even, and to provide the key equation solver circuit with a first
erasure pattern or a second erasure pattern according to a result
of the determining when a decoding failure is detected by the
decoding error detection circuit, the first erasure pattern being
provided when the number of errors is odd, the second erasure
pattern being provided when the number of errors is even.
[0008] In an embodiment, a Reed-Solomon code soft-decision decoding
method may include steps of: generating a syndrome from a codeword;
generating an error location polynomial and an error evaluation
polynomial based on the syndrome; determining a location and
magnitude of an error in the codeword based on the error location
polynomial and the error evaluation polynomial; determining whether
decoding on the codeword is successful or failed based on the
location and magnitude of the error; and determining whether a
number of errors in the codeword is odd or even, and performing
decoding on the codeword based on the syndrome and a first erasure
pattern or based on the syndrome and a second erasure pattern
according to a result of the determining when a decoding failure is
detected, the first erasure pattern being provided when the number
of errors is odd, the second erasure pattern being provided when
the number of errors is even.
[0009] In an embodiment, a semiconductor device may include: a
memory cell array configured to store a codeword; and a
Reed-Solomon code soft-decision decoding device configured to read
the codeword from the memory cell array, decode the codeword, and
output error-corrected data corresponding to the codeword, wherein
the Reed-Solomon code soft-decision decoding device comprises: a
syndrome generation circuit configured to generate a syndrome from
the codeword; a key equation solver circuit configured to generate
an error location polynomial and an error evaluation polynomial
based on the syndrome; a Chien-search & error-evaluator circuit
configured to determine a location and magnitude of an error in the
codeword based on the error location polynomial and the error
evaluation polynomial; a decoding error detection circuit
configured to determine whether decoding on the codeword is
successful or failed based on the location and magnitude of the
error; and an erasure control circuit configured to determine
whether a number of errors in the codeword is odd or even, and to
provide the key equation solver circuit with a first erasure
pattern or a second erasure pattern according to a result of the
determining when a decoding failure is detected by the decoding
error detection circuit, the first erasure pattern being provided
when the number of errors is odd, the second erasure pattern being
provided when the number of errors is even.
[0010] In accordance with embodiments, it is possible to correct an
error by using only an erasure pattern corresponding to one of an
even number of errors and an odd number of errors. Accordingly, it
is possible to reduce a decoding time while minimizing the
degradation of error correction performance.
BRIEF DESCRIPTION OF THE DRAWINGS
[0011] FIG. 1 is a block diagram illustrating a Reed-Solomon code
soft-decision decoding device in accordance with an embodiment.
[0012] FIG. 2 illustrates a method of determining whether the
number of errors is odd or even by using a syndrome.
[0013] FIG. 3 illustrates a first erasure pattern corresponding to
a case where the number of errors is odd.
[0014] FIG. 4 illustrates a second erasure pattern corresponding to
a case where the number of errors is even.
DETAILED DESCRIPTION
[0015] Various embodiments will be described below in detail with
reference to the accompanying drawings such that the present
disclosure can be easily carried out by those skilled in the art to
which the present disclosure pertains. The same reference numerals
among the reference numerals illustrated in each drawing denote the
same members.
[0016] In the description of the present disclosure, when it is
determined that detailed descriptions of related publicly-known
technologies may obscure the subject matter of the present
disclosure, detailed descriptions thereof will be omitted.
[0017] The terms such as `first` and `second` may be used to
distinguish one component from another component. However, the
components are not limited by the terms such as the `first` and
`second.`
[0018] Embodiments of the present disclosure provide an
erasure-based Reed-Solomon code soft-decision decoding method and
device, capable of reducing a decoding time while minimizing the
degradation of error correction performance.
[0019] In accordance with embodiments, in a Reed-Solomon code, a
codeword may include a plurality of symbols and one symbol may
include 8-bit data. The data may include real data and parity
data.
[0020] In accordance with embodiments, `error correction` may be
defined as correcting an error whose error location may be unknown
in the codeword, and `erasure correction` may be defined as
correcting an error whose error location may be known in the
codeword.
[0021] In accordance with embodiments, in the relation between
error correction performance and erasure correction performance,
one error correction performance may correspond to two erasure
correction performances.
[0022] In accordance with embodiments, a syndrome may be generated
based on the symbols of the codeword, and may indicate a form of
damage to the codeword due to an error. Each of the syndrome and
the symbols may include a plurality of bits. The plurality bits of
the syndrome may be obtained by logically calculating values of the
bits between the symbols. Such the syndrome can be used to
determine the number of symbol errors in the codeword as odd or
even. Such the syndrome may be used for generating an error
location polynomial and an error evaluation polynomial capable of
computing the location and magnitude of the error.
[0023] In accordance with embodiments, the plurality of bits of the
syndrome may be used for determining whether the number of errors
in symbols is odd or even. For example, an error correction decoder
of the Reed-Solomon code may determine the number of errors in the
symbols to be odd when the number of bits each having a value of
`1` in the syndrome is odd, and may determine the number of errors
in the symbols to be even when the number of bits each having the
value of `1` in the syndrome is even.
[0024] In accordance with embodiments, an odd error number may be
defined as the number of errors in symbols in the codeword that is
odd, and an even error number may be defined as the number of
errors in symbols in the codeword that is even.
[0025] In accordance with embodiments, an erasure pattern may be
defined as a pattern for erasing the symbols of the codeword in
order to correct an error of the codeword in the Reed-Solomon code.
In accordance with embodiments, symbols with low reliability among
the symbols of the codeword may be preferentially erased using the
erasure pattern.
[0026] FIG. 1 is a block diagram illustrating a Reed-Solomon code
soft-decision decoding device 100 in accordance with an
embodiment.
[0027] Referring to FIG. 1, the Reed-Solomon code soft-decision
decoding device 100 may include a syndrome generation circuit 10, a
key equation solver circuit 20, a Chien-search &
error-evaluator (CSEE) circuit 30, a decoding error detection
circuit 40, and an erasure control circuit 50.
[0028] The syndrome generation circuit 10 may receive a codeword
including a plurality of symbols and generate a syndrome from the
codeword.
[0029] The syndrome generation circuit 10 may generate the syndrome
by computing bits of each symbol of the codeword by using a preset
algorithm. Since the preset algorithm for generating the syndrome
is well known, descriptions of generating the syndrome is omitted
herein.
[0030] The key equation solver circuit 20 may generate an error
location polynomial and an error evaluation polynomial from the
syndrome.
[0031] The CSEE circuit 30 may compute a location and magnitude of
an error in the codeword based on the error location polynomial and
the error evaluation polynomial.
[0032] The decoding error detection circuit 40 may determine
whether decoding is successful or failed based on the location and
magnitude of the error.
[0033] The erasure control circuit 50 may determine whether the
number of errors in the codeword is odd or even, and may provide
the key equation solver circuit 20 with a first erasure pattern
corresponding to an odd error number or a second erasure pattern
corresponding to an even error number according to a result of the
determining when a decoding failure is detected by the decoding
error detection circuit 40.
[0034] The erasure control circuit 50 may receive the syndrome from
the syndrome generation circuit 10 and determine the number of
errors in the codeword to be odd or even by using the
characteristics of the syndrome. For example, in order to determine
the number of errors in the symbols of the codeword, the erasure
control circuit 50 may determine the number of symbol errors in the
codeword to be odd or even depending on the number of bits having a
value of `1` in the syndrome.
[0035] For example, when the number of bits each having a value of
`1` in the syndrome is odd, the erasure control circuit 50 may
determine the number of symbol errors to be odd, and when the
number of bits each having the value of `1` in the syndrome is
even, the erasure control circuit 50 may determine the number of
symbol errors to be even. When the number of symbol errors is
determined to be even, the number of errors in the symbols of the
codeword is determined to be even. When the number of symbol errors
is determined to be odd, the number of errors in the symbols of the
codeword is determined to be odd.
[0036] Then, when the number of symbol errors is determined to be
odd and the decoding failure is detected by the decoding error
detection circuit 40, the erasure control circuit 50 may provide,
as the first erase pattern, 2+4j erasure patterns to the key
equation solver circuit 20, j being an integer equal to or more
than 0. For example, the erase control circuit 50 may provide two
erase patterns to the key equation solver circuit 20 when the
number of symbol errors is determined as odd and the decoding
failure is detected. In addition, the erase control circuit 50 may
provide six erase patterns to the key equation solver circuit 20
when the decoding failure is detected even after erasure
correction. Here, the 2+4j erasure patterns may be provided in an
order in which a symbol with low reliability among the symbols of
the codeword is preferentially erased.
[0037] The erasure control circuit 50 may provide the 2+4j erasure
patterns in an order in which a symbol with low reliability among
the symbols of the codeword is preferentially erased. This will be
described in detail in the description associated with FIG. 3.
[0038] On the other hand, when the number of symbol errors is
determined to be even and the decoding failure is detected by the
decoding error detection circuit 40, the erasure control circuit 50
may provide, as the second erasure pattern, 4k erasure patterns to
the key equation solver circuit 20, k being a positive integer. For
example, the erase control circuit 50 may provide four erase
patterns to the key equation solver circuit 20 when the number of
symbol errors is determined as even and the decoding failure is
detected. Here, the 4k erasure patterns may be provided in an order
in which a symbol with low reliability among the symbols of the
codeword is preferentially erased.
[0039] The erasure control circuit 50 may provide the 4k erasure
patterns in an order in which a symbol with low reliability among
the symbols of the codeword is preferentially erased. This will be
described in detail in the description associated with FIG. 4.
[0040] When the number of errors in the symbols in the codeword is
odd, the key equation solver circuit 20 may receive the 2+4j
erasure patterns from the erasure control circuit 50 and compute
the error location polynomial and the error evaluation polynomial
based on the syndrome and the 2+4j erasure patterns.
[0041] On the other hand, when the number of errors in the symbols
in the codeword is even, the key equation solver circuit 20 may
receive the 4k erasure patterns from the erasure control circuit 50
and compute the error location polynomial and the error evaluation
polynomial based on the syndrome and the 4k erasure patterns.
[0042] FIG. 2 illustrates a method of determining whether the
number of errors is odd or even by using the syndrome.
[0043] Referring to FIG. 2, a codeword R[0:n] may include a
plurality of symbols R[0] to R[n], and one symbol may include 8-bit
data, n being a positive integer.
[0044] A syndrome1 S1 may be generated based on each symbol of the
codeword R[0:n]. Such a syndrome may indicate a form of damage to
the codeword R[0:n] due to an error, and may be used for generating
an error location polynomial and an error evaluation polynomial
capable of computing a location and magnitude of the error.
[0045] Furthermore, the syndrome1 S1 may be used for determining
whether the number of errors in the symbols of the codeword R[0:n]
is odd or even. For example, when the number of bits each having a
value of `1` in the syndrome1 S1 is odd, the number of errors in
the symbols of the codeword R[0:n] may be determined to be odd, and
when the number of bits each having the value of `1` in the
syndrome1 S1 is even, the number of errors in the symbols of the
codeword R[0:n] may be determined to be even.
[0046] As illustrated in FIG. 2, when a bit error has occurred in
the first symbol R[0], the second symbol R[1], and the third symbol
R[2], since the number of bits each having the value of `1` in the
syndrome1 S1 is 3. Thus, the number of symbol errors may be
determined to be odd.
[0047] In this way, in an embodiment, whether the number of errors
in the symbols of the codeword R[0:n] is odd or even may be
determined using the bit values of the syndrome1 S1.
[0048] FIG. 3 illustrates a first erasure pattern corresponding to
a case where the number of errors is odd.
[0049] Referring to FIG. 3, the reliability of symbols in a
codeword may be computed during decoding, and symbols with the
lowest reliability among the symbols in the codeword may be
arranged to be preferentially subjected to the erasure correction
using an erasure pattern.
[0050] In an embodiment, when the number of errors in the symbols
of the codeword is determined to be odd, the erasure correction may
be performed using 2+4j erasure patterns, j being an integer equal
to or more than 0.
[0051] Specifically, in an embodiment, the symbols with low
reliability may be decoded based on a syndrome.
[0052] Subsequently, when a decoding failure is detected and the
number of errors is determined to be odd, the erasure correction
may be performed by applying the 2+4j erasure patterns to the
codeword.
[0053] For example, when the number of errors is determined to be
odd, a first symbol and a second symbol with low reliability among
the symbols in the codeword may be subjected to the erasure
correction using two erasure patterns among the 2+4j erasure
patterns, and then whether the decoding is successful or failed may
be determined based on a location and magnitude of an error
computed after the erasure correction is performed on the first
symbol and the second symbol.
[0054] Subsequently, when the decoding is failed, the first symbol
and a third symbol may be subjected to the erasure correction using
two erasure patterns among the 2+4j erasure patterns, and then
whether the decoding is successful or failed may be determined
based on a location and magnitude of an error computed after the
erasure correction is performed on the first symbol and the third
symbol.
[0055] Subsequently, when a decoding failure is detected, the
second symbol and the third symbol may be subjected to the erasure
correction using two erasure patterns among the 2+4j erasure
patterns, and then whether the decoding is successful or failed may
be determined based on a location and magnitude of an error
computed after the erasure correction is performed on the second
symbol and the third symbol.
[0056] In this way, symbols with the lowest reliability among the
symbols in the codeword may be preferentially subjected to the
erasure correction using the 2+4j erasure patterns, and the erasure
correction process may be iterated until a decoding success is
detected.
[0057] As described above, in an embodiment, when the number of
errors occurring in the symbols of the codeword is odd, an error
may be corrected using the 2+4j erasure patterns, and thus the
iterative process of performing the erasure correction by applying
4k erasure patterns corresponding to an even error number can be
skipped, so that it is possible to reduce a total decoding time by
half.
[0058] FIG. 4 illustrates a second erasure pattern corresponding to
a case where the number of errors is even.
[0059] Referring to FIG. 4, the reliability of symbols in a
codeword may be computed during decoding, and symbols with the
lowest reliability among the symbols in the codeword may be
arranged to be preferentially subjected to the erasure correction
using an erasure pattern.
[0060] In an embodiment, when the number of errors in the symbols
of the codeword is determined to be even, the erasure correction
may be performed using 4k erasure patterns, k being a positive
integer.
[0061] Specifically, symbols with low reliability among the symbols
in the codeword may be decoded based on a syndrome.
[0062] Subsequently, when a decoding failure is detected and the
number of errors is determined to be even, the erasure correction
may be performed by applying the 4k erasure patterns to the
codeword.
[0063] For example, when the number of errors is determined to be
even, first to fourth symbols with low reliability among the
symbols in the codeword may be subjected to the erasure correction
using four erasure patterns among the 4k erasure patterns, and then
whether the decoding is successful or failed may be determined
based on a location and magnitude of an error computed after the
erasure correction is performed on the first to fourth symbols.
[0064] Subsequently, when the decoding is failed, the first symbol,
the second symbol, the third symbol, and a fifth symbol may be
subjected to the erasure correction using four erasure patterns
among the 4k erasure patterns, and then whether the decoding is
successful or failed may be determined based on a location and
magnitude of an error computed after the erasure correction is
performed on the first symbol, the second symbol, the third symbol,
and the fifth symbol.
[0065] Subsequently, when a decoding failure is detected, the first
symbol, the second symbol, the third symbol, and the fifth symbol
may be subjected to the erasure correction using four erasure
patterns among the 4k erasure patterns, and then whether decoding
is successful or failed may be determined based on a location and
magnitude of an error computed after the erasure correction is
performed on the first symbol, the second symbol, the third symbol,
and the fifth symbol.
[0066] In this way, symbols with the lowest reliability among the
symbols in the codeword may be preferentially subjected to the
erasure correction using the 4k erasure patterns, and the erasure
correction process may be iterated until a decoding success is
detected.
[0067] As described above, when the number of errors occurring in
the symbols of the codeword is even, an error may be corrected
using the 4k erasure patterns, and thus the iterative process of
performing erasure correction by applying the 2+4j erasure patterns
corresponding to an odd error number can be skipped. Accordingly,
it is possible to reduce the total decoding time by half.
[0068] Meanwhile, the embodiment may be applied to a semiconductor
device including a memory cell array.
[0069] The semiconductor device may include the memory cell array
(not illustrated) and the Reed-Solomon code soft-decision decoding
device 100 shown in FIG. 1.
[0070] The memory cell array may store a codeword encoded with the
Reed-Solomon code.
[0071] The Reed-Solomon code soft-decision decoding device 100 may
read the codeword from the memory cell array, and correct an error
of the codeword using a corresponding erasure pattern.
[0072] Furthermore, the embodiment may be also applied to a
communication device that decodes data of a communication channel
that is encoded with the Reed-Solomon code.
[0073] As described above, in accordance with the embodiments, it
is possible to correct an error of a codeword or data by using only
an erasure pattern corresponding to an even error number or an
erasure pattern corresponding to an odd error number, so that it is
possible to reduce a total decoding time while minimizing the
degradation of the error correction performance.
[0074] Although various embodiments have been illustrated and
described, various changes and modifications may be made to the
described embodiments without departing from the spirit and scope
of the invention as defined by the following claims.
* * * * *