KR101882620B1 - Apparatus and Method for Soft-Decision BCH Decoding based on Syndrome - Google Patents

Abstract

Disclosed are a soft-decision BCH decoding apparatus based on syndrome and a method thereof. A soft-decision BCH decoding apparatus disclosed according to the present invention includes: a test syndrome operation unit for extracting a position of a least reliable bit for each section in which a codeword is divided by the number of least reliable bits and converting the extracted position into a test syndrome; a pattern syndrome generation unit for receiving the syndrome and the test syndrome from the syndrome operation unit to generate a pattern syndrome; a syndrome coefficient calculation unit for receiving the pattern syndrome to calculate a syndrome coefficient of a candidate codeword, and separately calculating a syndrome coefficient for all orders while sharing a common part of syndrome coefficients for each order of a syndrome matrix; a pre-Chien search unit and a metric checking unit for receiving the pattern syndrome to sum the number of errors; and an error syndrome coefficient determining unit and a Chien search unit for determining a syndrome coefficient of the candidate codeword through a control signal, selecting an operation result of the least reliable bit, and calculating a vector to correct an error.

Description

[0001] The present invention relates to a soft-decision BCH decoding apparatus based on a syndrome, and a BCH decoding apparatus and a soft-

The present invention relates to a soft-decision Bose-Chaudhuri-Hocquenghem (BCH) decoding method and apparatus using syndromes and a hard-decision kernel.

Recently, BCH codes have been widely used in forward error correction techniques for detecting and correcting errors occurring during transmission and reception in data transmission in short distance wireless communication, memory systems, and optical communication systems.

The BCH code is represented by BCH (n, k, t), where n is the number of codeword symbols including data and parity, k is the number of symbols in the data, Indicates the number of error symbols. The maximum correctable number of error symbols is defined as the relationship between the exponent m of the Galois field and the equation n-k ≥ mt. Usually, a hard-decision BCH code refers to a binary BCH code in which each symbol is 1 bit.

The Berlekamp-Massey (BM) algorithm, the Modified Euclidean (ME) algorithm and the modified SBS (Modified Step-by-Step) algorithm have been used for decoding the hard decision BCH codes. In particular, the modified SBS algorithm can perform syndrome determinant operation without feedback circuit and latency when the number of correctable errors is less than 4 (t≤4) compared to BM and ME algorithms. Therefore, it is possible to generate the syndromes related to the soft decision code sequentially and apply it to the determinant calculation as it is.

로 정의되며, 때문에 정정 가능한 최대 심볼의 수는 t + p가 된다. 따라서 연판정 BCH 부호는 경판정 BCH 부호보다 비트오류율(Bit Error Rate, BER)이 낮아 경판정 BCH 부호보다 높은 성능을 발휘한다.The soft decision BCH code measures channel information using LLR (Log Likelihood Ratio). The p bits having the smallest size determine that the probability of error occurrence is the highest, and this is called Least Reliability Bit. The least confidence bit is

, So the number of correctable maximum symbols is t + p. Therefore, the soft-decision BCH code has lower bit error rate (BER) than the hard-decision BCH code, so that it performs better than the hard-decision BCH code.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems occurring in the prior art, and it is an object of the present invention to provide a soft decision BCH decoder, which uses hard decision BCH decoding as a kernel and a test pattern generator as a pattern syndrome generator, Generator), thereby providing a high performance soft decision BCH decoding apparatus with low hardware complexity and a method therefor. Further, in order to solve the problem that the complexity increases exponentially when the necessary calculations are shared by increasing the order of the existing soft decision BCH decoder (the number of correction error symbols), it is necessary to use the syndrome of the hard decision codeword as a candidate code A syndrome of a word (Candidate codeword) In other words, a pattern syndrome generation unit for generating a pattern syndrome is configured, and a hard decision kernel is to be applied to decoding of a soft decision code while suppressing an increase in area. Further, in order to provide a soft decision BCH decoding apparatus in which the length of the codeword does not affect the complexity of the decoder, the error syndrome coefficient determination unit including the D-flip flop is replaced by the error syndrome coefficient determination unit.

In one aspect, the soft decision BCH decoding apparatus using the test syndrome proposed in the present invention extracts the positions of the least reliable bits one by one for each section obtained by dividing a code word by the number of the least reliable bits, A pattern syndrome generating unit for receiving the syndrome from the syndrome calculating unit and the converted test syndrome to generate a pattern syndrome, calculating a syndrome coefficient of the candidate codeword by receiving the generated pattern syndrome, A syndrome coefficient calculation unit for separately calculating a syndrome coefficient for all orders while sharing a common part of syndrome coefficients for each order, a line-to-parent search unit for calculating a sum of error numbers by receiving the pattern syndrome, A syndrome coefficient of the candidate codeword is determined through an inspection unit and a control signal , It selects the calculation result of the minimum confidence bits, and for computing a vector to correct errors include errors the syndrome coefficient determining unit, and search hit portion.

The test syndrome operation unit compares the size of the LLR of the received symbol from the channel based on the chase algorithm, extracts one minimum confidence bit for each zone in which the symbol is divided by the number of least confidence bits, Into a test syndrome.

The test syndrome operation unit performs a test syndrome operation by inputting the test syndrome value into a shift register without a separate sorting operation process.

The pattern syndrome generating unit sequentially receives the syndrome from the syndrome calculating unit and the converted test syndrome, and sequentially calculates a syndrome for all candidate codewords using a D-flip-flop and a feedback circuit.

The error syndrome coefficient determination unit and the data search unit determine the syndrome coefficient of the candidate code word through the control signal and transmit the determined syndrome coefficient to the data search unit, select the calculation result of the least confidence bit for soft decision, The result of the operation is summed with a GF (Galois field) adder, and an error vector is generated to correct errors of the received word.

The pre-searcher and the metric checking unit sums the number of errors generated from the syndrome coefficient without outputting an error, and outputs the syndrome coefficient by receiving the syndrome coefficient of the candidate code word.

According to another aspect of the present invention, there is provided a soft decision BCH decoding method using a test syndrome according to the present invention, wherein a minimum confidence bit position is extracted one by one for each interval obtained by dividing a codeword by the number of least reliable bits, Generating a pattern syndrome by receiving the syndrome from the syndrome operation unit and the converted test syndrome, calculating a syndrome coefficient of the candidate codeword by receiving the generated pattern syndrome, Calculating syndrome coefficients of all the orders separately while sharing a common part of the syndrome coefficients, summing the number of errors by receiving the pattern syndrome, and determining a syndrome coefficient of the candidate codeword through a control signal , Selects the operation result of the least reliable bit, and corrects the error And a step of calculating the emitter.

According to embodiments of the invention the soft decision other words, by using the test syndrome and pattern syndrome in the BCH decoder design, again a syndrome for a candidate code word, without having to generate 2 ^p number of candidate code words, light and create a pattern syndrome It is possible to provide an efficient soft-decision BCH decoding apparatus having a low hardware complexity and a method thereof using the decision BCH decoder structure. In addition, the hard-decision error locations of all candidate codewords are not stored by applying the pre-check unit and the error syndrome coefficient determination unit. By determining only the error syndrome coefficient of the correctable candidate codeword, the D-flip-flop is removed by the product of the length of the codeword and the number of candidate codewords so that the length of the codeword and the hardware complexity are independent of each other. And a method therefor.

1 is a diagram illustrating a configuration of a soft decision BCH decoding apparatus using a test syndrome according to an embodiment of the present invention.
2 is a diagram illustrating a test syndrome operation unit for extracting a position of a least reliable bit according to an embodiment of the present invention.
3 is a diagram illustrating a pattern syndrome generating unit according to an embodiment of the present invention.
4 is a diagram illustrating a syndrome coefficient calculation unit according to an embodiment of the present invention.
5 is a diagram illustrating a pre-searcher and a metric checker according to an embodiment of the present invention.
6 is a diagram showing a test syndrome search unit according to an embodiment of the present invention.
7 is a diagram illustrating an error syndrome coefficient determination unit and a parent search unit according to an embodiment of the present invention.
8 is a flowchart illustrating a soft decision BCH decoding method using a test syndrome according to an embodiment of the present invention.
9 is a diagram illustrating a bit error rate according to an embodiment of the present invention.

The present invention relates to a soft-decision Bose-Chaudhuri-Hocquenghem (BCH) decoding method and apparatus using syndromes and a hard-decision kernel. More specifically, a soft decision BCH code is decoded by using a kernel of a hard decision BCH code structure, and a low complexity degree is obtained through a test syndrome operation unit, a pattern syndrome generation unit, a syndrome coefficient calculation unit, a pre- BCH decoding method and apparatus. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram illustrating a configuration of a soft decision BCH decoding apparatus using a test syndrome according to an embodiment of the present invention.

The soft decision BCH decoding apparatus using the proposed test syndrome includes a syndrome operation unit 110, a test for extracting the positions of the least reliable bits one by one for each interval in which a code word is divided into the number of least reliable bits, A syndrome calculating unit 120, a pattern syndrome generating unit 130 for receiving the syndrome from the syndrome calculating unit and the converted test syndrome and generating a pattern syndrome, and calculating a syndrome coefficient of the candidate codeword by receiving the generated pattern syndrome A syndrome coefficient calculation unit 140 separately calculating a syndrome coefficient for all orders while sharing a common part of syndrome coefficients for each order of the syndrome matrix, a syndrome coefficient calculation unit 140 for receiving the pattern syndrome, A search unit and a metric checking unit 150, a syndrome counting unit for calculating a syndrome coefficient of the candidate codeword And includes selects the operation result of the least reliable bits and to correct errors determined error syndrome coefficient for computing a vector portion and a pro-search section 160, a control unit 170, and test syndrome search section 180. The

The test syndrome operation unit 120 compares the size of LLRs of symbols received from a channel based on a chase algorithm and extracts one minimum confidence bit for each zone in which the symbol is divided by the number of least reliable bits, Converts the location of the bit to the test syndrome. At this time, the test syndrome operation unit 120 performs a test syndrome operation by inputting the test syndrome value into a shift register without requiring a separate sort operation operation.

The pattern syndrome generation unit 130 receives the syndrome from the syndrome operation unit and the converted test syndrome, and sequentially calculates the syndrome for all the candidate codewords using the D-flip-flop and the feedback circuit.

The error syndrome coefficient determination unit and the parent search unit 160 determine the syndrome coefficient of the candidate codeword through the control signal and transmit the determined syndrome coefficient to the parent search unit and select the calculation result of the minimum confidence bit for soft decision After performing the sum operation using the GF adder, the error vector is generated to correct the error of the received word.

The pre-searcher and the metric checking unit 150 sums the number of errors generated from the syndrome coefficient without outputting an error, outputs the syndrome coefficient by receiving the syndrome coefficient of the candidate code word, and outputs the syndrome coefficient. The soft decision BCH decoding apparatus using the test syndrome will be described in more detail below.

A soft decision BCH decoding apparatus using a test syndrome according to an embodiment of the present invention is a forward error correction (FEC) system for correcting errors occurring in data during a data transmission process in a transmitter of a digital communication system by a receiver A syndrome operation unit 110, a test syndrome operation unit 120 for extracting a location of the smallest LLR, a pattern syndrome generation unit 130 for receiving a syndrome and a test syndrome as input, A syndrome coefficient calculation unit 140 for calculating a syndrome coefficient of the candidate code, a line-to-parent search unit and a metric checking unit 150 for measuring information of the syndrome coefficient of the candidate code, a control unit 170, A test syndrome good search unit 180 for restoring the position of the LLR, Error syndrome relates to a soft decision BCH decoding apparatus and method including a coefficient determiner, and pro-search section 160.

Embodiments of the present invention are described with reference to (1020, 990, 3) codes, and the present invention can be equally applied to the implementation of a soft decision BCH decoder structure for various communication systems. According to an embodiment of the present invention, a modified SBS algorithm having a low complexity can be used when the number of correctable errors is four or less as an algorithm for decoding a soft decision BCH code.

The soft decision BCH decoding apparatus and method according to the present invention calculates an operation on a candidate codeword based on a syndrome and generates a syndrome of all candidate codes, that is, a pattern syndrome using a test syndrome and a hard decision syndrome. Then, the proposed soft decision BCH decoding apparatus does not store the error position of the candidate codeword but selects the syndrome coefficient of the candidate codeword and generates the position of the error by using this to correct the error of the received codeword. By using this method, the D-flip-flop for storing the error location of the candidate codeword can be eliminated.

The syndrome coefficient calculation unit 140 applies the existing modified SBS algorithm and reduces the hardware complexity by sharing the coefficients required for the calculation. The syndrome coefficient calculation unit 140 receives the input from the pattern syndrome generation unit 130 and calculates a syndrome coefficient .

The test syndrome bad searcher 180 receives an input from the test syndrome calculating unit 120 and restores the position of the LLR to generate an error location for soft decision.

The error syndrome coefficient determination unit and the parent search unit 160 do not store the search result of all the candidate codewords in the D-flip-flop, and thus have complexity independent of the length of the codeword. The syndrome coefficient of the candidate codeword may be replaced by selecting the error syndrome coefficient and checking the error syndrome coefficient. The syndrome coefficient of the candidate codeword and the control signal from the control unit are inputted from the pre- And uses only the syndrome coefficients of the selected one candidate code word to generate a hard decision error position of the candidate code word.

Table 1 below shows the pseudo code of the decoding process of the soft decision BCH code decoding apparatus.

<Table 1>

The symbol of the codeword received from the channel is an LLR value of a two's complement system and is input to the soft decision BCH decoder and converted into a size value and input to the test syndrome operation unit. The most significant bit is input to the syndrome operation unit as a hard decision code.

The syndrome of the syndrome operation unit and the test syndrome operation unit test syndrome are input to the pattern syndrome generation unit and the pattern syndrome is output. The output of the pattern syndrome generation unit is input to the syndrome coefficient calculation unit, and the syndrome coefficients for all the candidate codewords are sequentially output. When calculating a pattern syndrome pattern syndromes for the first candidate code words are S _{i, tp0} = S _i, two pattern syndromes for the second candidate code words are S _{i, tp1} = S _{i, tp0} + (α ^{t0) i,} two pattern syndromes for the second candidate code words are _{S i, tp2 = S i,} tp1 + (α t1) i, pattern syndromes for the third candidate code words S _{i, tp3} = S _{i, tp2} + (alpha ^t0 ) ⁱ , the pattern syndrome corresponding to the fourth candidate codeword is _{Si, tp4} = _{Si, tp3} + (alpha ^t2 ) ⁱ , and the pattern syndrome corresponding to the fifth candidate codeword is _{Si, tp5 = S i, tp4 + (α} t0) i, pattern syndrome for the sixth candidate code words _{S i, tp6 = S i,} tp5 + (α t1) i, pattern syndrome for the seventh candidate codewords S _{i, tp 7} = S _{i, tp 6} + (α ^{t 0} ) ⁱ .

When calculating the syndrome coefficients corresponding to the candidate ^{_{codewords, S 1, tpj 3 + S}} 3, when _tpj = 0 is satisfied ^{_{R tpj = S 1, tpj 3}} + S 3, tpj, A tpj = S 1, tpj 2, _Tpj = S _{1, tpj} , C _tpj = 0, deg _{2, tpj} = 1 are output. At this time, S _{1, tpj} (S _{1, tpj} ² S _{3, tpj} + S _{5, Deg 3 and tpj} = 0 are output _{, and} deg _{3 and tpj} = 1 are not output. If S _{1, tpj} ³ + S _{3 and tpj} = 0 are not established, R _tpj = (S _{1, tpj} ³ + S _{3, tpj} ) ² + S _{1, tpj} (S _{1, tpj} ² S _{3, 5, tpj), A tpj =} S 1, tpj 2 S 3, tpj + S 5, tpj, B tpj = S 1, tpj 4 + S 1, tpj S 3, tpj, C tpj = S 1, tpj 3 + S _{3, tpj,} deg _{2, tpj} = the value of 1 is output, wherein ^{_{(S 1, tpj 3 + S}} 3, tpj) 2 + S 1, tpj (S 1, tpj 2 S 3, tpj + S 5, _tpj = 0, deg _{3 and tpj} = 0 are established, and deg _{3 and tpj} = 1 are not output. Where j = 2 ^p -1, ... 0. Order information of the syndrome coefficient _{deg 2, tp0, deg 3,} tp0, deg 2tp1, deg 3, tp1, deg 2, tp2, deg 3, tp2, deg 2, tp3, deg 3, tp3, deg 2, tp4, deg 3 _{, tp4, deg2, tp5, deg3, tp5, deg2, tp6} , _deg3 , _tp6 , _{deg2 , tp7 , deg3 and tp7} are input to the controller.

The line-to-lead search unit and the metric check unit include a line-to-lead search unit for receiving the syndrome coefficient of the candidate code word as an input and outputting the error position of each candidate code word, and a metric checking unit for calculating the sum of the number of error locations.

Line - The pro-search section and the metric checker receives the syndrome coefficient for every candidate code words sequentially checks the condition _{R tpj, i = A tpj,} i α P + B tpj, i α 2P + C tpj, i α 3P , Where j = N-1, ..., 0. If R _{tpj, i} = A _{tpj, i} α ^P + B _{tpj, i} α ^2P + C _{tpj, i} α ^3P is satisfied, it means that an error occurs in the position of the jth bit. Add 1 to the value. The outputs M _tp0 , M _tp1 , M _tp2 , M _tp3 , M _tp4 , M _tp5 , M _tp6 , and M _tp7 of the metric checking unit are input to the control unit.

The controller receives the degree information deg ₂ and deg ₃ of the syndrome coefficients of all the candidate codewords as an input and then receives the metric value M of the hatching and metric checking unit as input as described above so that the candidate codeword judgment . First, it is determined whether deg ₂ is 0, and since there is no error or 0 (t = 1) at 0, the syndrome coefficient of the candidate codeword in which M is 0 or 1 is selected. On the other hand, if deg ₂ is 1, the number of errors is 2 or 3, and deg ₃ is always 0 when the number of errors is 2. Therefore, if the number of errors is 3, In this case, the value of M is compared with {deg ₂ , deg ₃ }. If the two values are the same, the syndrome coefficient of the candidate codeword suitable for correcting the error is equal to the number of errors and the degree information of the syndrome coefficient. Determines the syndrome coefficient of the correctable candidate codeword according to the above condition, and outputs a control signal.

,

,

, 에 1을 출력하고 만족하지 않을 경우 0을 출력한다.Test syndrome pro-search unit In the ^{α i + α j0 = 0 (} i = N-1, ..., 0, j = p-1, ..., 0) from the test and determining whether the syndrome is established, the condition If satisfied, the location of the least confidence bit

,

,

, And outputs 1 when it is not satisfied.

The error syndrome coefficient determination and the parent search unit receives the control signal received from the control unit, the syndrome coefficient of the candidate codeword from the line-to-parent search unit and the metric checking unit, and the position of the restored minimum confidence bit of the test syndrome search unit. Position of the appropriate syndrome coefficient and the soft decision error to correct the error in the syndrome of the candidate code words in response to the control signal _{e 0, e 1, e 2} , e 3, e 4, e 5, e 6, the e ₇ Table 1 , And performs an operation as shown in the error syndrome coefficient determination portion.

2 is a diagram illustrating a test syndrome operation unit for extracting a position of a least reliable bit according to an embodiment of the present invention.

The test syndrome operation unit 200 according to an embodiment of the present invention includes a comparator 210. Compares the magnitude of the LLR of the symbol received from the channel based on the chase algorithm with the comparator 210, extracts one minimum confidence bit for each zone in which the symbol is divided by the number of least confidence bits, Converts the location to the test syndrome.

According to an embodiment, when the minimum confidence bit is set to 3 (p = 3) as shown in Table 1, the test syndrome calculating unit 200 divides the interval of the received word by 3 and calculates a minimum confidence bit To a test syndrome value. And the pattern syndrome is calculated without generating the code word using the converted test syndrome.

3 is a diagram illustrating a pattern syndrome generating unit according to an embodiment of the present invention.

The pattern syndrome generating unit 300 according to the embodiment of the present invention includes a test syndrome selector 310. Pattern for every candidate code words in a manner joined to provide a single test syndrome as described by testing syndrome selector 310 from the test syndrome calculation unit shown in Table 1, the α ^t0, α ^t1, α ^t2 to input the syndrome S _{i, tp0} and it generates the _{i S, tp1,} S _{i, tp2,} S _{i, tp3,} S _{i, tp4,} S _{i, tp5,} S _{i, tp6,} S _i, sequentially _tp7 (i = 1,3,5).

4 is a diagram illustrating a syndrome coefficient calculation unit according to an embodiment of the present invention.

The pattern syndromes S _{1, tpj} , S _{3, tpj} , S _{5, tpj} , (j = 2 ^p -1, ..., 0) of the pattern syndrome generation unit described above are supplied to the syndrome coefficient calculation unit S ₁ , S ₃ S ₅ , and the syndrome coefficient calculation unit shares the common operation and performs calculation by optimizing the operation as shown in the result table 1 when the order of the error syndrome coefficients is 2 and 3

5 is a diagram illustrating a pre-searcher and a metric checker according to an embodiment of the present invention.

The metric checking unit 510 of the line-to-line searching unit and the metric checking unit 500 has a parallel structure according to the number of candidate codewords, and a syndrome coefficient of the corresponding candidate codeword is input into one block structure, The syndrome coefficients R, A, B, and C and the metric value M are output.

6 is a diagram showing a test syndrome search unit according to an embodiment of the present invention.

The test syndrome is received from the test syndrome operation unit and the position of the least reliable bit is recovered. The position of the least reliable bit in the first codeword interval is restored to e _t0 , the second position is e _t1 , and the third position is restored to e _t2 .

7 is a diagram illustrating an error syndrome coefficient determination unit and a parent search unit according to an embodiment of the present invention.

The error syndrome coefficient determination unit and the parent search unit 710 of the parent search unit 700 receive a control signal from the control unit, p minimum confidence bit positions from the test syndrome search unit, a candidate search word from the metric checking unit, Lt; / RTI &gt; The syndrome coefficient for hard decision of the candidate codeword is selected by the control signal and transmitted to the inner parent search unit. And selects the operation result of the least confidence bit for soft decision as shown in Table 1 according to the control signal. The Galois field (GF) adder adds the result of the search and the selected least confidence bit to the GF adder, and the result is added to the hard decision code of the receiver word and the GF adder. And performs restoration of the received word.

8 is a flowchart illustrating a soft decision BCH decoding method using a test syndrome according to an embodiment of the present invention.

The soft-decision BCH decoding method using the proposed test syndrome includes a step 810 of extracting the positions of the least confidence bits for each interval obtained by dividing the codeword by the number of the least reliable bits and converting the positions of the least confidence bits into a test syndrome, And generating a pattern syndrome by receiving the converted test syndrome and calculating a syndrome coefficient of the candidate codeword by receiving the generated pattern syndrome and calculating a syndrome coefficient for each degree of the syndrome matrix, A step 830 of calculating a syndrome coefficient for all orders while sharing a common part, a step 840 of receiving the pattern syndrome and summing the number of errors, and determining a syndrome coefficient of the candidate codeword through a control signal Selecting a computation result of the least confidence bit, computing (850) a vector to correct the error, It includes.

In step 810, the position of the least reliable bit is extracted one by one for each interval obtained by dividing the code word by the number of least reliable bits, and converted into a test syndrome.

In step 820, a syndrome from the syndrome operation unit and the converted test syndrome are input to generate a pattern syndrome. In step 820, the size of the LLRs of symbols received from the channel are compared based on the chase algorithm, and one minimum confidence bit is extracted for each zone in which the symbol is divided by the number of least confidence bits, Into a test syndrome. At this time, a test syndrome operation is performed by inputting the test syndrome value into a shift register without a separate sorting operation process.

In step 830, the generated syndrome is computed, and syndrome coefficients for all orders are computed separately while sharing a common part of the syndrome coefficients for each degree of the syndrome matrix. The syndrome from the syndrome operation unit, and the converted test syndrome, and sequentially calculates a syndrome for all candidate codewords using a D-flip-flop and a feedback circuit.

Thereafter, in step 840, the pattern syndrome is received and the number of errors is calculated.

In step 850, a syndrome coefficient of the candidate codeword is determined through a control signal, a calculation result of the least reliable bit is selected, and a vector is calculated to correct the error. The syndrome coefficient of the candidate codeword is determined through the control signal, the determined syndrome coefficient is transmitted to the parent search unit, the result of the least confidence bit for soft decision is selected, After summing with GF adder, error vector is generated to correct error of receiver. At this time, the number of errors generated from the syndrome coefficient is summed up without outputting an error, and the syndrome coefficient of the candidate code word is received and the syndrome coefficient is output.

The soft decision BCH decoding method proposed in the present invention computes an operation on a candidate codeword based on a syndrome, and generates a syndrome of all candidate codes, that is, a pattern syndrome using a test syndrome and a hard decision syndrome. Then, the proposed soft decision BCH decoding apparatus does not store the error position of the candidate codeword but selects the syndrome coefficient of the candidate codeword and generates the position of the error by using this to correct the error of the received codeword. By using this method, the D-flip-flop for storing the error location of the candidate codeword can be eliminated.

9 is a diagram illustrating a bit error rate according to an embodiment of the present invention.

Referring to FIG. 9, the bit error rate performance of the hard decision BCH code and the bit error rate performance of the soft decision BCH code can be compared based on the (1020, 990, 3) code. In this case, when the error performance is based on the bit error rate of 10-6, the soft decision BCH code has a coding gain of about 0.65dB than the hard decision BCH code and the bit error rate performance is improved.

Therefore, by using the method proposed in the present invention, it is possible to implement a soft decision BCH decoding apparatus which has a better bit error rate performance than the hard decision BCH decoding apparatus.

The apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the apparatus and components described in the embodiments may be implemented within a computer system, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA) , A programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing unit may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of the foregoing, and may be configured to configure the processing device to operate as desired or to process it collectively or collectively Device can be commanded. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device As shown in FIG. The software may be distributed over a networked computer system and stored or executed in a distributed manner. The software and data may be stored on one or more computer readable recording media.

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI &gt; or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (7)

A soft decision BCH decoding apparatus using a test syndrome,
A test for comparing the size of the LLR of the codeword based on the Chase algorithm and for converting the position of the least reliable bit to a test syndrome by extracting one minimum confidence bit for each zone in which the codeword is divided by the number of least confidence bits A syndrome operation unit;
A pattern syndrome generation unit for receiving the syndrome from the syndrome operation unit and the converted test syndrome and generating a pattern syndrome;
A syndrome coefficient calculation unit for calculating a syndrome coefficient of a candidate codeword by receiving the generated pattern syndrome, separately calculating a syndrome coefficient for all orders while sharing a common part of syndrome coefficients for each order of the syndrome matrix;
A line-to-edge search unit and a metric check unit for receiving the pattern syndrome and summing the number of errors; And
An error syndrome coefficient determination unit for determining a syndrome coefficient of the candidate codeword through a control signal, selecting a minimum confidence bit for soft decision, computing a vector to correct the error using the selected minimum confidence bit, part
And a soft decision BCH decoding apparatus using the test syndrome. delete The method according to claim 1,
Wherein the test syndrome calculating unit comprises:
A test syndrome operation is performed by inputting the test syndrome into a shift register without a separate sorting operation process
Soft decision BCH decoding apparatus using test syndrome. The method according to claim 1,
Wherein the pattern syndrome generating unit comprises:
The syndrome from the syndrome operation unit and the converted test syndrome are received, and the syndrome for all the candidate codewords is sequentially calculated using the D-flip-flop and the feedback circuit
Soft decision BCH decoding apparatus using test syndrome. The method according to claim 1,
Wherein the error syndrome coefficient determination unit and the parent search unit determine,
The syndrome coefficient of the candidate codeword is determined through the control signal, the determined syndrome coefficient is transmitted to the parent search unit, the result of the least confidence bit for soft decision is selected, After performing a summation operation using a GF adder, an error vector is generated to correct errors in the codeword
Soft decision BCH decoding apparatus using test syndrome. The method according to claim 1,
Wherein the pre-searcher and the metric checker comprise:
A syndrome coefficient of the candidate codeword is summed, and the syndrome coefficient of the candidate codeword is output by receiving the syndrome coefficient of the candidate codeword
Soft decision BCH decoding apparatus using test syndrome. A soft decision BCH decoding method using test syndrome,
Comparing the size of the LLR of the codeword based on the Chase algorithm, extracting one minimum confidence bit for each zone where the codeword is divided by the number of least confidence bits, and converting the location of the least confidence bit to a test syndrome ;
Receiving a syndrome from the syndrome operation unit and the converted test syndrome and generating a pattern syndrome;
Calculating a syndrome coefficient of a candidate codeword by receiving the generated pattern syndrome, separately calculating a syndrome coefficient for all orders while sharing a common portion of syndrome coefficients for each order of the syndrome matrix;
Receiving the pattern syndrome and summing the number of errors; And
Determining a syndrome coefficient of the candidate codeword via a control signal, selecting a minimum confidence bit for soft decision, computing a vector to correct the error using the selected minimum confidence bit
A soft decision BCH decoding method using a test syndrome.

Images