WO2022161147A1 - Decoding method, apparatus and device, and computer readable storage medium - Google Patents

Abstract

A decoding method, apparatus and device, and a computer readable storage medium, relating to the technical field of communications. Data of m rows and n columns, for example, data of m rows or n columns, in a codeword can form data of multiple dimensions, and in the present method, data of m rows and n columns in a codeword is stored in one storage address, thereby facilitating reading of the data of m rows and n columns in the codeword from the storage address to allow for storage and reading of the data of dimensions in the codeword, and facilitating subsequent multi-dimensional decoding, for example, row-direction decoding or column-direction decoding, of the read data of multiple dimensions in the codeword.

Description

Decoding method, apparatus, device, and computer-readable storage medium

This application claims the priority of the Chinese patent application filed on January 27, 2021, with the application number of 202110112883.5 and the application title of "Decoding Method, Apparatus, Equipment and Computer-readable Storage Medium", all of which The contents are incorporated herein by reference.

technical field

The present application relates to the field of communication technologies, and in particular, to a decoding method, apparatus, device, and computer-readable storage medium.

Background technique

Forward error correction coding (FEC) is widely used in communication systems to provide reliable digital transmission with less overhead. In particular, multi-dimensional algebraic codes in FEC can provide higher performance gains at lower complexity and power consumption, such as turbo product codes (TPC), staircase codes (staircase codes), and continuous alternating BCH codes ( continuously interleaved Bose Ray Hocquenghem, CIBCH) etc.

At present, a general decoding device decodes the multi-dimensional algebraic code by means of iterative decoding. During each iterative decoding process, the decoding device performs at least one time on the data of each dimension of the multi-dimensional algebraic code in the cache. Decoding, taking the TPC codeword as an example, each iterative decoding process of the TPC codeword includes a row decoding process and a column decoding process. The column decoding process refers to that the decoding device decodes each column subcode in the TPC codeword.

Since the decoding device decodes the data in the multi-dimensional algebraic code from multiple dimensions, in the process of iteratively decoding the multi-dimensional algebraic code, the decoding device needs to read the data of each dimension in the multi-dimensional algebraic code from the cache. Therefore, how to store and read the data of each dimension in the multi-dimensional algebraic code is compatible with the decoding process, which is an urgent problem to be solved.

SUMMARY OF THE INVENTION

Embodiments of the present application provide a decoding method, apparatus, device, and computer-readable storage medium, which are compatible with data storage and reading of various dimensions in a codeword. The technical solution is as follows:

In a first aspect, a decoding method is provided, the method comprising:

The codeword is stored in a plurality of storage addresses, wherein one storage address is used to store m*n data of m rows and n columns in the codeword, and both the m and the n are integers greater than or equal to 1; For any storage address in the plurality of storage addresses, read m*n data stored in the any storage address from the any storage address; decoding.

Since the data of m rows and n columns in the code word can form data of multiple dimensions, such as m rows of data or n columns of data, this method stores the data of m rows and n columns in the code word in a storage address, which is convenient for data from the The data of m rows and n columns in the codeword is read from the storage address, which is compatible with the data storage and reading of each dimension in the codeword, and facilitates the subsequent multi-dimensional decoding of the data of multiple dimensions in the read codeword. , such as row-wise decoding or column-wise decoding.

In a possible implementation manner, the decoding of the read m*n data includes:

If the current decoding process is a row-wise decoding process, determine m row data groups based on the positions of the read m*n data in the codeword; perform the m row data groups on the m row data groups Decoding, wherein one row data group includes n pieces of data among the read m*n pieces of data, and the n pieces of data are located in the same row in the codeword.

Based on the above possible implementation manners, each row of data in the data group is decoded, so that the row-wise decoding of the data group can be realized, that is, the row-wise decoding of the data group is realized. code.

In a possible implementation manner, the decoding the m row data groups includes:

The m line data groups are respectively sent to m first decoding nodes, and each first decoding node decodes the received line data groups.

Based on the above possible implementations, by sending m line data groups to m first decoding nodes respectively, each first decoding node decodes a received line data group, so as to realize parallel processing of m data groups. Each row data group is decoded, which improves the decoding efficiency of the data group.

In a possible implementation manner, the decoding of the read m*n data includes:

If the current decoding process is a column-oriented decoding process, based on the positions of the read m*n data in the codeword, determine n column data groups; Decoding, wherein one column data group includes m data in the read m*n data, and the m data are located in the same column in the codeword.

Based on the above possible implementation manners, each column of data in the data group is decoded, so that the column-wise decoding of the data group can be realized, that is, the decoding of the data group in the column-wise dimension is realized. code.

In a possible implementation manner, the decoding the n column data groups includes:

The n column data sets are respectively sent to n second decoding nodes, and each second decoding node decodes the received column data sets.

Based on the above possible implementations, by sending n row data groups to n second decoding nodes respectively, and each second decoding node decodes a column data group received, thereby realizing parallel pairing of n data groups. Each column data group is decoded, which improves the decoding efficiency of the data group.

In a possible implementation manner, after the read m*n pieces of data are decoded, the method further includes:

Modify m*n pieces of data stored in any of the storage addresses into m*n pieces of decoded data, and the m*n pieces of decoded data are the decoded pieces of the read m*n pieces of data. data.

In a possible implementation manner, the storing the codeword in multiple storage addresses includes:

According to the granularity of m rows and n columns, the data in the codeword is divided into a plurality of data groups; the plurality of data groups are stored in the plurality of storage addresses, wherein one data group includes the data in the codeword m*n data in m rows and n columns, one data group is stored in one storage address.

In a possible implementation manner, the multiple storage addresses belong to the same storage node.

In a possible implementation manner, the multiple storage addresses belong to multiple storage nodes;

The storing the plurality of data groups at the plurality of storage addresses includes:

combining the multiple data sets into multiple data sets; storing the data sets in the multiple data sets at the multiple storage addresses, wherein one data set includes adjacent multiple data sets in the multiple data sets; One data group is stored in one storage address, and multiple data groups in the same data set are stored in storage addresses in different storage nodes.

In a possible implementation manner, after storing the data groups in the multiple data sets at the multiple storage addresses, the method further includes:

For any data set in the plurality of data sets, a plurality of data groups of the any data set are read from the plurality of storage nodes.

Based on the above possible implementation manners, one data set can be read from multiple storage nodes at a time, so that multiple data sets in the read data set can be decoded in parallel, which improves the degree of parallelism in the decoding process and the decoding process. efficiency.

In a possible implementation manner, the codeword includes multiple subcodes, and after the codeword is stored in multiple storage addresses, the method further includes:

For a plurality of adjacent target subcodes among the plurality of subcodes, at least one target data set related to the plurality of target subcodes is determined from the plurality of data sets, wherein the at least one target data set includes At least one target data group related to the plurality of target subcodes, one target data group includes data of part of the target subcodes in the plurality of target subcodes;

Read the target data set in the at least one target data set from the plurality of storage nodes; and decode the read target data set.

In a possible implementation manner, the read target data group includes at least one odd target data group and at least one even target data group, and one odd target data group is a target located at an odd row layer of the codeword data group, an even target data group is the target data group located at the even-numbered row layer of the codeword;

The decoding of the read multiple target data groups includes:

sending the at least one odd target data group to a plurality of third decoding nodes, and decoding the received odd target data group by the plurality of third decoding nodes; sending the at least one even target data group It is sent to multiple fourth decoding nodes, and the multiple fourth decoding nodes decode the received even data group.

In a second aspect, a decoding apparatus is provided for executing the above decoding method. Specifically, the decoding apparatus includes a functional module for executing the decoding method provided in the above-mentioned first aspect or any optional manner of the above-mentioned first aspect.

In a third aspect, a decoding device is provided, the decoding device includes a processor and a memory, the memory stores at least one piece of program code, and the program code is loaded and executed by the processor to implement the above-mentioned decoding method. operation.

In a fourth aspect, a computer-readable storage medium is provided, where at least one piece of program code is stored in the storage medium, and the program code is loaded and executed by a processor to implement the operations performed by the above decoding method.

In a fifth aspect, a computer program product or computer program is provided, the computer program product or computer program includes program code, the program code is stored in a computer-readable storage medium, and the processor of the decoding device is obtained from the computer-readable storage medium. The program code is read, and the processor executes the program code, so that the computer device executes the method provided in the first aspect or various optional implementation manners of the first aspect.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the drawings that are used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without creative effort.

1 is a schematic structural diagram of a TCP codeword provided by an embodiment of the present application;

2 is a schematic structural diagram of an OFEC codeword provided by an embodiment of the present application;

3 is a schematic diagram of a decoding system provided by an embodiment of the present application;

4 is a schematic structural diagram of a decoding device provided by an embodiment of the present application;

5 is a flowchart of a decoding method provided by an embodiment of the present application;

6 is a schematic diagram of a codeword storage provided by an embodiment of the present application;

7 is a schematic diagram of an iterative decoding provided by an embodiment of the present application;

8 is a flowchart of a decoding method provided by an embodiment of the present application;

9 is a schematic diagram of a codeword storage provided by an embodiment of the present application;

10 is a schematic diagram of the correspondence between a data group and a storage node in a TPC codeword provided by an embodiment of the present application;

11 is a decoding process of a convolutional code provided by an embodiment of the present application;

12 is a schematic diagram of an OFEC codeword decoding process provided by an embodiment of the present application;

FIG. 13 is a schematic structural diagram of a decoding apparatus provided by an embodiment of the present application.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

In order to facilitate the understanding of the decoding process provided by this application, the codeword structures of the following various codewords are first introduced as follows.

1. TPC code word

FIG. 1 is a schematic structural diagram of a TCP codeword provided by an embodiment of the present application. Referring to FIG. 1 , the TCP codeword in FIG. 1 includes R*K data in R rows and K columns, and each data occupies 1 bit. Wherein, R and K are both integers greater than 1. In a possible implementation manner, the TPC codeword includes a plurality of subcodes, and each subcode includes a plurality of data in the TPC codeword. For example, each row of data in the TPC codeword is a subcode, or, each column of data in the TCP codeword is a subcode.

2. Open Forward Error Correction Coding (open FEC, OFEC) codewords

FIG. 2 is a schematic structural diagram of an OFEC codeword provided by an embodiment of the present application. Referring to FIG. 2 , the OFEC codeword is a wireless extended convolutional code. The OFEC codeword includes multiple data in multiple rows and multiple columns, and each data occupies one bit. The data in the OFEC codeword can be divided into historical data and future data based on the order of generation. Taking a row of data in the OFEC codeword as an example, the data in the next row of the row of data is future data relative to the row of data. And the next row of data relative to this row of data is historical data.

In a possible implementation manner, the OFEC codeword includes a plurality of blocks (blocks), each block includes m*n data of m rows and n columns, and each data occupies one bit, wherein m and n are both is an integer greater than 1, for example, m=n=16 in FIG. 2 .

Multiple blocks in the same row in the OFEC codeword are one row layer. For example, as shown in FIG. 2, each row layer includes 8 blocks. If m=n=16, then each row layer includes 8*6*16 blocks. data. Optionally, each row layer is identified with a row layer number. For example, the OFEC codeword in FIG. 2 includes [-∞, +∞] row layers, and each row layer has a row layer number. For convenience of description, row layers with odd row layer numbers are denoted as odd row layers, and row layers with even row layer numbers are denoted as even row layers. Among them, two adjacent row layers, one odd and one even, are an OFEC frame, that is, an OFEC frame includes two adjacent radix row layers and even row layers. An OFEC frame in FIG. 2 includes 8*16 *16*2=128*16*2 data.

Multiple blocks in the same column in the OFEC codeword are a column layer. Optionally, each column level is identified with a column level number, for example, the OFEC codeword in FIG. 2 includes column levels 0 to 7. In a possible implementation manner, the location information of a piece of data in the codeword includes the row layer number of the target row layer where the data is located, the column layer number of the target column layer where the data is located, and the row where the data is located in the target row layer. , and the column number of the column where the data is located in the target column layer. Taking the 2 data in the first block in the -2 row layer in the OFEC codeword shown in FIG. 2 as an example, the position information of the data located in the first row and the first column in the block is expressed as [-2, 0,0,0], where -2,0,0,0 are used to indicate the -2 row layer, the 0 column layer, the 1st row in the -2 row layer and the 1st column in the 0 column layer, respectively; The position information of the data in the 2nd row and 1st column in the block can be expressed as [-2, 0, 1, 0], where -2, 0, 1, 0 are used to indicate -2 row layer, 0 column respectively layer, the 2nd row in the -2 row layer, and the 1st column in the 0 column layer; it should be noted that the mark B at the block in Figure 2 is the side length of the block, which can be understood as a m*n size The block, m and n are all equal to B.

In a possible implementation manner, the OFEC codeword includes a plurality of subcodes, and the first half of the data of each subcode is composed of a plurality of oblique short columns and a plurality of horizontal short columns in the OFEC codeword, wherein the oblique Each short column in the multiple short columns is a column of data in a block in a different row layer, each short column in the horizontal multiple short columns is a row of data in a block, and the horizontal multiple short columns are in a the same row in the row layer. Taking the subcode in the OFEC codeword shown in Figure 2 as a BCH (256, 239) subcode, and each block includes 16*16 data as an example, in Figure 2 multiple short columns of the same line type constitute a BCH (256, 239) subcode, a plurality of short columns of a line type include 8 short columns arranged in a diagonal line and 8 short columns located in the same row, wherein the 8 short columns arranged in a diagonal line are BCH The multiple short columns in the oblique direction in the (256, 239) subcode are the data of the first 128 bits of the BCH (256, 239) subcode, and the 8 short columns in the same row are in the BCH (256, 239) subcode. The multiple short columns in the horizontal direction are the data of the last 128 bits of the BCH (256, 239) subcode.

It can be seen from FIG. 2 that the data of the subcodes in the OFEC codeword are interleaved together, and the interleaving relationship in the OFEC codeword is represented by the following formulas (1)-(2).

If t<W:{(H^1)–2G-2W/B+2[t/B],[t/B],(t%B)^p,p} (1)

If t≥W:{H, [(t–W)/B], p, (t％B)^p} (2)

Among them, t is the position number of a certain data of a subcode in the OFEC codeword in the subcode, for example, the subcode is a BCH(256,239) subcode, the first one in the BCH(256,239) subcode The position sequence number of the data on the bit is 1, and the position sequence number of the data on the second bit is 2. W is half of the code length of the subcode. For example, if the subcode is a BCH (256, 239) subcode, the value of W is 128. (H^1)–2G-2W/B+2[t/B] is the row layer number of the row layer where the data is located when the data is the first half of the subcode. Among them, B is the side length of a block, for example, the value is 16. [t/B] is the column layer number of the column layer where the data is located when the data is the first half of the subcode. For example, t=1, B=16, 0<1/16<1, then 0 is the column layer number of the column layer where the data is located. For another example, t=17, B=16, 1<17/16<2, Then 1 is the column layer number of the column layer where the data is located. G is an integer greater than or equal to 1, which is used to represent the guard interval. There is no coding relationship between the data of each line in the continuous 2G+2 line layers, so the decoding device can record the continuous 2G+2 line layers as a complex frame, and encodes and decodes a multiframe in parallel. (t%B)^p is the row number of the column where the data is located when the data is the first half of the subcode, where "%" is used to indicate the round-down operation, and "^" is used to indicate the bitwise operation Perform an exclusive OR operation. When the data is the data of the first half of the subcode, p is the column number of the column where the data is located, and when the data is the data of the second half of the subcode, p is the row number of the row where the data is located. H is the row layer number of the row layer where the data is located when the data is the data in the second half of the subcode. [(t-W)/B] is the column layer number of the column layer where the data is located when the data is the data in the second half of the subcode. From the above description of formulas (1)-(2), it can be known that the above formulas (1) and (2) are the position information of the first half data and the position information of the second half of the subcode, respectively.

FIG. 3 is a schematic diagram of a decoding system provided by an embodiment of the present application. Referring to FIG. 3 , the system 300 includes a data processing node 301 , a storage node 302 , a data distribution node 303 , and a decoding node 304 . Wherein, the data processing node 301 is used to acquire the codeword, and store the acquired codeword in the storage node 302 . In a possible implementation manner, the data processing node 301 divides the data in the codeword into multiple data groups, each data group includes m*n data in m rows and n columns in the codeword, and divides each data group into a plurality of data groups. Each data group is stored at one storage address within storage node 302 . During the row iterative decoding process of the codeword, the data distribution node 303 reads out one data group from a storage address of the storage node 302 at a time; if the current decoding process is a row decoding process, the The data distribution node 303 determines a plurality of row data groups read into the data group, and distributes each row data group to a decoding node 304, and the plurality of decoding nodes 304 decode the received row data groups. code, so that the data decoding process of one dimension on one data group is implemented by a plurality of decoding nodes 304 in parallel. If the current decoding process is a column decoding process, the data distribution node 303 determines a plurality of column data groups to be read into the data group, and distributes each column data group to a decoding node 304, where the A plurality of decoding nodes 304 decode the received column data group, so that a data decoding process of another dimension on one data group is implemented by the plurality of decoding nodes 304 in parallel.

In another possible implementation manner, the data processing node 301 combines multiple adjacent data sets in the codeword into one data set, and stores multiple data sets in the data set in multiple storage nodes 302 , and the multiple data groups in the data set are located in different storage nodes 302 . In the process of iteratively decoding the codeword, the data distribution node 303 reads out multiple data groups of the data set from multiple storage nodes 302 at a time. If the current decoding process is a row decoding process, the data distribution node 303 determines a plurality of row data groups of each data group in the read data set, and distributes each row data group of each data group to a decoding code node 304, at this time, a decoding node 304 is used to receive a row data group in each data group in the data set, and the plurality of decoding nodes 304 decode each received row data group, so that multiple The decoding nodes 304 implement the data decoding process of one dimension on one data set in parallel. If the current decoding process is a column decoding process, the data distribution node 303 determines a plurality of column data groups in each data group in the read data set, and distributes each column data group of each data group to a Decoding node 304, at this time, one decoding node 304 is used to receive one column data group in each data group in the data set, and the plurality of decoding nodes 304 decode each received column data group, so that the multiple decoding nodes 304 decode each column data group received. The decoding nodes 304 implement in parallel the data decoding process of another dimension on one data set. And compared with the process of decoding one data group at a time, the process of decoding multiple data groups in one data set at a time has a higher degree of parallelism and higher decoding efficiency.

In a possible implementation manner, the decoding node 304 is any decoder with a decoding function, such as a soft in soft out (soft in soft out, SISO) BCH sub-decoder, a hard in hard out ( hard in hard out, HIHO) BCH sub-decoder, etc.

After each decoding node 304 completes the decoding of a row data group or a column data group, the decoding node 304 writes the decoded data obtained by the decoding back to the storage address in the storage node 302 in the original way, so as to Overwrites the corresponding data stored in this memory address. So that the subsequent data distribution node 303 reads data from the storage address again, and performs the next decoding process to realize iterative decoding.

In a possible implementation manner, each node in the system 300 is an independent device. In another possible implementation manner, the system 300 may be deployed on a decoding device, and each node in the system 300 is a module in the decoding device. Of course, the functions of multiple nodes in the system 300 may also be implemented by the same module on one decoding device, or implemented by multiple devices.

For example, as shown in FIG. 4 is a schematic structural diagram of a decoding device provided by an embodiment of the present application, the decoding device 400 may vary greatly due to different configurations or performance, including one or more processors 401 and one or more More than one memory 402, wherein the processor includes a central processing unit (central processing units, CPU). At least one piece of program code is stored in the memory 402, and the at least one piece of program code is loaded and executed by the processor 401 In order to realize the decoding methods provided by the following method embodiments. For example, each memory 401 is a storage node. For example, when the decoding device 400 includes a processor 401, the processor 401 is used to implement the functions of a data processing node, a data distribution node, and a decoding node. When the decoding device 400 includes multiple processors 401, one processor 401 may be used to implement the function of one of the data processing node, the data distribution node, and the decoding node.

Of course, the decoding device 400 may also have components such as a wired or wireless network interface, a keyboard, and an input/output interface for input and output. The decoding device 400 may also include other components for realizing device functions, which are not described here. Repeat.

In an exemplary embodiment, a computer-readable storage medium, such as a memory including program codes, is also provided, and the program codes can be executed by a processor in the terminal to implement the decoding methods in the following embodiments. For example, the computer-readable storage medium is a non-transitory computer-readable storage medium, such as read-only memory (ROM), random access memory (RAM), compact disc read-only memory, CD-ROM), magnetic tapes, floppy disks, and optical data storage devices, etc.

Exemplarily, for further explanation, the decoding device adopts the storage mode and reading mode of the data group to store and read the code word, and decode the read data group to realize the decoding of the code word. For the process, refer to the flowchart of a decoding method provided by an embodiment of the present application shown in FIG. 5 .

501. A decoding device acquires a codeword.

The codeword is a codeword to be decoded, and the codeword is any type of multi-dimensional code, such as a TPC codeword, a convolutional codeword, a ladder codeword, and a CIBCH. The dimension of the codeword is greater than or equal to 2. When the codeword is a two-dimensional code, the codeword includes R*K data in R rows and K columns, and each data occupies 1 bit, for example, as shown in Figure 1 TPC codeword. When the codeword is a three-dimensional code, the codeword includes F*R*K data in R rows and K columns with a height of F, and each data occupies 1 bit, where F is an integer greater than 1. Herein, the embodiment of the present application does not limit the type and dimension of the codeword.

In a possible implementation manner, this step 501 is performed by a data processing node in the decoding device. The data processing node obtains the codeword from a target node, where the target node is any node that stores the codeword, or any node that encodes the codeword. Optionally, the target node is a node other than the decoding device, for example, when the encoding device that encodes the codeword and the decoding device are not the same device, the encoding device that encodes the codeword is the target node, After the encoding device encodes the codeword, it sends the codeword to a data transmission interface of the decoding device, and a data processing node in the decoding device receives the codeword from the data transmission interface. Or, the target node is a module in the decoding device. For example, the decoding device also has an encoding function. After encoding the codeword, the target node in the decoding device sends the codeword to the decoding device. The data processing node in the code device.

502. The decoding device stores the codeword in multiple storage addresses, and one storage address is used to store m*n pieces of data in m rows and n columns in the codeword, and both m and n are integers greater than or equal to 1 .

Optionally, the multiple storage addresses belong to the same storage node, and the storage space provided by the storage node is indicated by the storage address. For example, the storage node is a RAM, and the decoding device can store the codeword in the RAM. in multiple storage addresses.

When the codeword is the TCP codeword shown in FIG. 1 , m is an integer greater than or equal to 1 and less than R, and n is an integer greater than or equal to 1 and less than K. Among them, m and n are equal or unequal, and R and K are equal or unequal. In some embodiments, R divides m and/or K divides n. In some embodiments, R is not divisible by m and/or K is not divisible by n.

In some embodiments, the decoding device first divides the codeword into a plurality of data groups, and then stores the plurality of data groups in the plurality of storage addresses. In a possible implementation manner, this step 502 is implemented by the processes shown in the following steps 5021-5022.

Step 5021, the decoding device divides the data in the codeword into multiple data groups according to the granularity of m rows and n columns, and one data group includes m*n data in m rows and n columns in the codeword.

In a possible implementation manner, the decoding device sets a sliding window of m rows and n columns of bits, and the decoding device divides the codeword into multiple data groups by sliding the sliding window on the codeword.

Optionally, initially, the decoding device slides the sliding window to the upper left corner of the codeword. At this time, the sliding window covers the data in the 1st to mth rows and the 1st to nth columns in the codeword, then The decoding device determines all the data covered by the sliding window at this time as a data group; the decoding device uses n bits as the sliding step, slides the sliding window to the right, and when the sliding is completed, the sliding window Covering the data in rows 1 to m and columns n+1 to 2n in the codeword, the decoding device determines all the data covered by the sliding window at this time as a data group; and so on, the decoding The code device slides the sliding window to the right until the sliding window can cover the data in the 1st to mth row and the Kth column in the codeword, and the decoding device ends sliding on the 1st to mth row in the codeword; And with reference to the above-mentioned sliding mode, the decoding device slides the sliding window in the m+1 to 2mth row to divide the data in the m+1 to 2mth row in the codeword into a plurality of data groups; and so on. , until the data node slides the sliding window until it can cover the data in the Rth row and the Kth column of the codeword, and ends the sliding. For example, a schematic diagram of a codeword storage provided by an embodiment of the present application shown in FIG. 6, each rectangular frame on the codeword in FIG. 6 is the area covered by the sliding window after each sliding, and all the data in each rectangular frame form a data set.

It should be noted that when R can divide m and K can divide n, after the decoding device slides the sliding window every time, the sliding window can cover m*n data in the codeword. When R cannot divide m or K cannot divide n, if the sliding window slides to the Kth column or the sliding window slides in the Rth row, the sliding window cannot cover the m*n data in the codeword. There are no m*n pieces of data in the data group determined by the decoding device.

In some embodiments, after the codeword is divided into a plurality of data groups, the decoding device treats the plurality of data groups in the same row in the codeword as a row layer, and will be in the same row in the codeword as a row layer. Multiple data groups of columns act as a column layer. The decoding device sets a row layer number for each row layer to identify each row layer, and sets a column layer number for each column layer to identify each column layer. A data group is identified by the position information of the data group, and the position information of the data group includes the row layer number of the row layer where the data group is located and the column layer number of the column layer where the data group is located. For example, the position information of the first data group in the codeword in FIG. 6 is 1-1, which indicates the data group in the first row layer and the first column layer.

It should be noted that the process shown in the above step 5021 is the grouping of the data in one dimension of the codeword, and the decoding device can, according to the process shown in the above step 5021, group the data in another dimension of the codeword. The data above is grouped once.

Step 5022: The decoding device stores the plurality of data groups in the plurality of storage addresses, and each storage address stores one data group.

Optionally, the decoding device sends the multiple data groups to a storage node, and the storage node stores each of the multiple data groups in a storage address in the storage node respectively. As shown in FIG. 6 , the storage node stores data groups 1-1, 1-2 and 1-3 in the codeword at storage addresses 1 to 3 in the storage node, respectively.

For any data group in the plurality of data groups, the storage node stores the data in the any data group at a storage address in the storage node row by row. As shown in FIG. 6 , the storage node sequentially stores four data in three rows of data group 1-1 in storage address 1, and 12 data in storage address 1 are arranged in one row.

It should be noted that, in some embodiments, the process shown in this step 502 is performed by a data processing node in the decoding device. The processes shown in the steps 501-502 only need to be performed once, and do not need to be performed multiple times.

After the decoding device stores the codeword in the storage node, the decoding device decodes the codeword stored in the storage node. Since the codeword is a multidimensional code, the decoding of the codeword by the decoding device is multidimensional decoding, or it can also be understood as iterative decoding, and the process of iteratively decoding the codeword by the decoding device includes row decoding. coding process and column decoding process. For any decoding process in the row decoding process or the column decoding process, the decoding device first reads each data group belonging to the codeword from the storage node, and uses the read data in each data group. data is decoded. To further describe any of the decoding processes, please refer to the following steps 503-504.

It should be noted that this step 502 is described by taking a two-dimensional codeword as an example. In some embodiments, when the dimension of the codeword is greater than 2, the decoding device divides the codeword into a plurality of multidimensional data blocks, each multidimensional data block is located on all dimensions of the codeword, and the decoding The code device stores each multidimensional block of data in a memory address. Taking a 3-dimensional codeword as an example, the decoding device divides the data in the 9*9*9 3-dimensional codeword into multiple 3*3*3* 3-dimensional data blocks, and converts each 3*3* 3* 3-dimensional data blocks are stored in one storage address respectively.

503. For any storage address in the plurality of storage addresses, the decoding device reads m*n data stored in the any storage address from the any storage address.

A data group stored in a storage address is the basic unit for the decoding device to read and write data, and the decoding device can read out m*n data in a data group from any storage address at a time. Optionally, the process shown in this step 503 is performed by a data distribution node in the decoding device.

504. The decoding device decodes the read m*n pieces of data.

In a possible implementation manner, if the current decoding process is a row-wise decoding process, the decoding device determines the arrangement of the m*n pieces of data read this time in the row-wise dimension of the codeword, And decode the data arranged in the row dimension. Optionally, the process shown in this step 504 is implemented by the processes shown in the following steps 5041-5042.

Step 5041: If the current decoding process is a row-wise decoding process, the decoding device determines m row data groups, one row data group based on the position of the read m*n data in the codeword. Including n data among the read m*n data, the n data are located in the same row in the codeword.

The m row data groups are the arrangement of the m*n data read out this time in the row dimension of the data group. Since the m*n data stored in any storage address is a row of data at any storage address, the m*n data read by the decoding device from any storage address is also a row of data. In order to facilitate the decoding code, the decoding device also needs to determine the position of the m*n data read this time in the codeword, so as to determine the read data based on the position of the m*n data in the codeword The arrangement of each data in the data group in the row dimension.

The decoding device converts the (i-1)*n+1th data to the i*th data in the read m*n data based on the position of the read m*n data in the codeword n data as a row data group. Among them, i is an integer greater than or equal to 1 and less than or equal to m. For example, when i=1, the decoding apparatus regards the 1st data to the nth data among the m*n pieces of data arranged in a line as a line data group. For another example, when i=2, the decoding device converts the n+1 th data to the 2n th data among the m*n pieces of data arranged in a line into a line data group.

It should be noted that, in some embodiments, the process shown in this step 5041 is performed by the data distribution node in the decoding device.

Step 5042: The decoding device decodes the m row data groups.

In some embodiments, the decoding device decodes the m row data groups in parallel, so as to improve the decoding efficiency of the decoding device for the data groups stored in any storage address. In a possible implementation manner, this step 5042 is completed by the data distribution node and a plurality of first decoding nodes in the decoding device. After the data distribution node determines m line data groups, the m line data groups are respectively sent to m first decoding nodes, and each first decoding node analyzes one of the received line data groups. The row data group is decoded in the direction, and the row data groups received by different first decoding nodes are different.

In a possible implementation manner, the data distribution node is provided with a data distribution queue of m*n bits, and the (i-1)*n+1th bit to the i*nth bit in the data distribution queue bits are connected to a first decoding node. The data distribution node sequentially arranges the read m*n data on each bit in the data distribution queue, so that the (i-1)*n+1th bit to the th The n data on i*n bits is also a row data group. For any row data group in the m row data groups, the data distribution queue, through the connection between the bit position where the any row data group is located and a first decoding node, distributes n data in the any row data group The pieces of data are sent to the first decoding node, and the first decoding node decodes the n pieces of data in any row of data groups.

In order to further illustrate the process shown in steps 5041-5042, please refer to the decoding process of the first dimension data shown in FIG. 7, wherein FIG. That is, the data in which m*n data are read and arranged in the dimension of the codeword row. After the data distribution node reads m*n data arranged in a row from any storage address, the data distribution node divides the m*n data into m row data groups, which are row data group 1 respectively. to m, and distribute m row data groups to m first decoding nodes, which are respectively the first decoding nodes 1 to m, so as to realize the distribution of the first dimension data. The n pieces of data in the row data group are decoded.

In a possible implementation manner, if the current decoding process is a column-wise decoding process, the decoding device determines the arrangement of the m*n pieces of data read this time in the column-wise dimension of the codeword , and decode the data arranged in the column dimension. Optionally, the process shown in this step 504 is implemented by the processes shown in the following steps 504A-504B.

Step 504A, if the current decoding process is a row-wise decoding process, the decoding device determines n column data groups and one column data group based on the position of the read m*n data in the codeword. Including m data among the read m*n data, the m data are located in the same column in the codeword.

The n column data groups are the arrangement of the m*n data read out this time in the column dimension of the data group. Since the m*n data stored in any storage address is a row of data at any storage address, the m*n data read by the decoding device from any storage address is also a row of data. In order to facilitate the decoding code, the decoding device also needs to determine the position of the m*n data read this time in the codeword, so as to determine the read data based on the position of the m*n data in the codeword The arrangement of each data in the data group in the column dimension of the data group.

The decoding device determines m row data groups based on the positions of the read m*n data in the codeword, and determines the jth data in the m row data groups as the n data The jth column data group in the column data group, where j is an integer greater than or equal to 1 and less than or equal to n. For example, the decoding device forms the first data group in the m row data groups into the first column data group in the n column data groups. For another example, the decoding device will form the second data group in the m row data groups into the second column data group in the n column data groups.

It should be noted that, in some embodiments, the process shown in this step 504A is performed by the data distribution node in the decoding device.

Step 504B, the decoding device decodes the n column data groups.

In some embodiments, the decoding device decodes the n column data groups in parallel, so as to improve the decoding efficiency of the decoding device for the data groups stored in any storage address. In a possible implementation manner, this step 504B is completed by a data distribution node and a plurality of second decoding nodes in the decoding device. After the data distribution node determines n column data groups, the n column data groups are respectively sent to n second decoding nodes, and each second decoding node analyzes one of the received column data groups. The column data sets are decoded, and the column data sets received by different second decoding nodes are different.

In a possible implementation manner, for any row of data groups, n bits located in the data distribution queue in the data distribution node of any row of data groups are respectively connected to a second decoding node, so that a The second decoding node can be connected to the bit where one data of the m data groups is located. It can be understood that, m pieces of data connected to m bits by one second decoding node are a column data group. The data distribution node sequentially arranges the read m*n data on each bit in the data distribution queue. For any column data group in the n column data groups, the data distribution queue uses the connection between the bits where the any column data group is located and a second decoding node, and m in the any column data group The pieces of data are sent to the second decoding node, and the second decoding node decodes the m pieces of data in any column of data groups.

In order to further illustrate the process shown in steps 504A-504B, please refer to the decoding process of the second dimension data shown in FIG. 7 , the second dimension data is the read m*n data in the column dimension of the codeword Arranged data. After the data distribution node reads m*n data arranged in a row from any storage address, the data distribution node splits the m*n data into n column data groups, which are column data group 1 respectively. to n, and distribute n column data groups to n second decoding nodes, which are respectively second decoding nodes 1 to n, so as to realize the distribution of second dimension data, and each second decoding node pairs one The m pieces of data in the column data group are decoded.

It should be noted that, in some embodiments, the m first decoding nodes and the n second decoding nodes are different decoding nodes. In other embodiments, the first decoding node and the second decoding node may be multiplexed. If m is greater than or equal to n, in the process of performing column-wise decoding, n first decoding nodes among the m first decoding nodes are multiplexed into n second decoding nodes. If m is less than n, in the process of row-wise decoding, m second decoding nodes in the n second decoding nodes are multiplexed into m first decoding nodes, so as to reduce the number of decoding equipment. The number of decoding nodes in . Since the decoding device performs the iterative decoding process for a data group in the codeword, it must first perform the row decoding process, and then perform the column decoding process, or perform the column decoding process first, and then perform the row decoding process. It can be seen that the column decoding process and the row decoding process performed by the decoding device for a data group do not occur simultaneously, so the first decoding node and the second decoding node can be multiplexed.

It should be noted that this step 504 is described by taking a 2-dimensional codeword as an example. In some embodiments, when the dimension of the codeword is greater than 2, and what the decoding device reads from any storage address is a multi-dimensional data block, not a two-dimensional data group, the decoding The device may first divide the multi-dimensional data block into multiple two-dimensional data groups, and then perform step 504 on each two-dimensional data group.

505. The decoding device modifies m*n pieces of data in any storage address into m*n pieces of decoded data, and the m*n pieces of decoded data are decoded for the read m*n pieces of data data after.

The m*n pieces of data in any storage address are in one-to-one correspondence with the m*n pieces of decoded data. After the decoding device completes the decoding of m*n data in any storage address, m*n decoded data is obtained, and the decoding device according to the m*n data in any storage address and The m*n pieces of decoded data have a one-to-one correspondence, and the m*n pieces of decoded data are written back to the any storage address in the original way to cover the m*n pieces of data stored in the any storage address, thereby realizing Revise.

In some embodiments, the process shown in this step 505 is completed by the data distribution node and the decoding node in the decoding device. In a possible implementation manner, if the current decoding process is a row-wise decoding process, for any first decoding node among the m first decoding nodes, any first decoding node corresponds to one After the n pieces of data in the row data group are decoded, n pieces of decoded data are obtained, and the n pieces of decoded data correspond to the n pieces of data one-to-one. The any first decoding node writes the n pieces of decoded data back to the data distribution node, and when the m first decoding nodes all write the n pieces of decoded data obtained after decoding back to the data distribution node Afterwards, the data distribution node can obtain m*n pieces of decoded data, and the data distribution node writes the m*n pieces of decoded data back to any storage address.

In another possible implementation, if the current decoding process is a column-wise decoding process, for any second decoding node in the n second decoding nodes, any second decoding node pairs After the m pieces of data in a column data group are decoded, m pieces of decoded data are obtained, and the m pieces of decoded data are in one-to-one correspondence with the m pieces of data. The any second decoding node writes the m pieces of decoded data back to the data distribution node, and when the n second decoding nodes all write the m pieces of decoded data obtained after their respective decoding back to the data distribution node Afterwards, the data distribution node can obtain m*n pieces of decoded data, and the data distribution node writes the m*n pieces of decoded data back to any storage address.

When the decoding device performs the process shown in the above steps 503-505 for each data group belonging to the codeword, the decoding process performed by the decoding device in one dimension of the codeword is completed. The decoding device jumps to execute the process shown in the above steps 503-505, and continues to perform the decoding process of another dimension on the codeword, thereby realizing the iterative decoding process of the codeword. For example, after the decoding device performs the row-wise decoding process for each data group in the codeword through the above steps 503-505, the decoding device continues to perform the above steps 503-505 to perform the decoding process on the code word. A column-wise decoding process is performed for each data group in the word.

Since m*n pieces of data stored in any data group satisfy the distribution of row dimension and the distribution of column dimension, in some embodiments, the decoding device uses the data group as the basic unit of iterative decoding. The decoding device performs the above steps 503-505 to complete the row-wise decoding of a data group stored in any storage address, without waiting for the data groups stored in other storage addresses to complete the row-wise decoding. The device directly performs the column decoding process on the data group currently stored in any storage address through the above steps 503-505, thereby realizing the iterative decoding process of one data group in the codeword.

Taking the data group stored in any storage address as [1, 2; 3, 4] as an example, in the row decoding process, the data distribution node reads the data group [1, 2; 3 from any storage address] , 4], and based on the read data groups [1, 2; 3, 4], two row data groups are determined, which are row data groups [1, 2] and row data groups [3, 4] respectively. The data distribution node sends the row data group [1, 2] and the row data group [3, 4] to the decoding nodes 1 and 2, respectively. Decoding node 1 decodes the row data set [1, 2] to obtain decoded data 5 and 6, and decoding node 2 decodes the row data set [3, 4] to obtain decoded data 7 and 8. The decoding node 1 writes the decoded data 5 and 6 back to the data distribution node, and the decoding node 2 writes the decoded data 7 and 8 back to the data distribution node. The data distribution node writes the decoded data 5, 6, 7, and 8 back to any storage address, and overwrites the 1, 2, 3, and 4 stored in any storage address. At this time, the data group stored in any storage address is updated. is [5, 6; 7, 8]. In the process of column decoding, the data distribution node reads the data set [5, 6; 7, 8] from any storage address, and determines the data set [5, 6; 7, 8] based on the read data set [5, 6; 7, 8]. 2 column data groups, namely column data group [5, 7] and column data group [6, 8]. The data distribution node sends the column data set [5, 7] and the column data set [6, 8] to the decoding nodes 3 and 4, respectively. The decoding node 3 decodes the column data set [5, 7] to obtain the decoded data 9 and 10, and the decoding node 4 decodes the column data set [6, 8] to obtain the decoded data 11 and 12. The decoding node 3 writes the decoded data 9 and 10 back to the data distribution node, and the decoding node 4 writes the decoded data 11 and 12 back to the data distribution node. The data distribution node writes the decoded data 9, 11, 10, and 12 back to any storage address, and overwrites the 5, 6, 7, and 8 stored in any storage address. At this time, the data group stored in any storage address is updated to [9, 11; 10, 12], and the data stored in any storage address undergoes one row decoding and one column decoding. When each data group in the word performs a row decoding process and a column decoding process, that is, an iterative decoding process is performed for the codeword. The above-mentioned decoding nodes 1 and 2 are both first decoding nodes, and decoding nodes 3 and 4 are both second decoding nodes.

Since the data of m rows and n columns in the codeword can form data of multiple dimensions, for example, the data of m rows or n columns, the embodiment of the present application stores the data of m rows and n columns in the codeword in a storage address, It is convenient to read the data of m rows and n columns in the code word from the storage address, so as to be compatible with the data storage and reading of each dimension in the code word, and it is convenient for subsequent data of multiple dimensions in the read code word to be performed. Multi-dimensional decoding, such as row-wise decoding or column-wise decoding. And it can support flexible design of any degree of parallelism without being limited by the length of subcodes, and support high-traffic FEC design requirements.

The process shown in FIG. 5 is a process in which the decoding device uses one data group as a decoding processing unit to decode the codeword. When there is a lot of data in the codeword, the decoding device can also use multiple data groups as a decoding processing unit, and decode the multiple data groups at the same time, so as to improve the parallel degree of decoding and the decoding efficiency. To further illustrate the process, please refer to the flowchart of a decoding method provided by an embodiment of the present application shown in FIG. 8 .

801. The decoding device obtains a codeword.

The process shown in this step 801 is the same as the process shown in the above-mentioned step 501, and this step 801 is not repeated in this embodiment of the present application.

802. The decoding device divides the data in the codeword into multiple data groups according to the granularity of m rows and n columns.

The process shown in this step 802 is the same as the process shown in the above-mentioned step 5021, and here, this embodiment of the present application does not repeat the description of this step 802.

803. The decoding device combines the multiple data sets into multiple data sets, and one data set includes multiple adjacent data sets among the multiple data sets.

Wherein, the adjacent multiple data groups have multiple adjacent forms. Optionally, the adjacent multiple data groups are consecutive multiple data groups in the same row layer. Taking the code word in FIG. 9 as an example, the decoding device uses the data groups 1-1, 1-2 and 1-3 in the first row layer of the code word as a data set, wherein FIG. 9 is this A schematic diagram of a codeword storage provided by an embodiment of the application. Optionally, the adjacent multiple data groups are continuous multiple data groups in the same column and row layer. Taking the code word in FIG. 9 as an example, the decoding device uses the code word in the first column layer. The datasets 1-1, 2-1, and 3-1 are used as a dataset. Optionally, the adjacent multiple data groups are multiple data groups in adjacent row layers or adjacent column layers. Taking the code word in FIG. 9 as an example, the data in the code word of the decoding device is Groups 1-1, 1-2, 2-1 and 2-2 are used as a dataset, wherein the row and column layers where the data groups 1-1, 1-2, 2-1 and 2-2 are located are adjacent .

It should be noted that, the embodiment shown in FIG. 9 is taken as an example for the decoding device to combine multiple consecutive data in the same row layer into one data set. In some embodiments, the process shown in this step 803 is performed by a data processing node in the decoding device.

804. The decoding device stores the data groups in the multiple data sets in multiple storage addresses in the multiple storage nodes, wherein one storage address stores one data group, and multiple data groups in the same data set are stored in different storage addresses. in the storage address within the storage node.

For any data set in the plurality of data sets, the decoding device stores the plurality of data groups in the any data set respectively at a storage address in the plurality of storage nodes, wherein each data set in the any data set Each data group corresponds to different storage nodes. Taking FIG. 9 as an example, a data set includes data groups 1-1, 1-2 and 1-3. The decoding device stores the data group 1-1 at the storage address 1 of the storage node 1, and stores the data group 1-2 at the storage address 1 of the storage node 1. The data groups 1-3 are stored at the storage address 1 of the storage node 2, and the data groups 1-3 are stored at the storage address 1 of the storage node 3.

In a possible implementation manner, the decoding device uses a cyclic shift manner to store data groups in the multiple data sets in multiple storage nodes. Optionally, the decoding device adopts a cyclic shift method to establish the correspondence between each data group in the plurality of data sets and the plurality of storage nodes, and stores each data group in one of the corresponding storage nodes. address.

In a possible implementation manner, the decoding device adopts a cyclic shift manner, and establishing the correspondence between each data group in the plurality of data sets and the plurality of storage nodes includes: if the number of data groups in each data set is equal to is s, there are s multiple storage nodes, and in the adjacent s row layers in the codeword, for the zth row layer in the s row layers, the decoding device uses the zth row layer After the s data groups in each data set are shifted to the right by (z-1) bits, they correspond to s storage nodes in turn, where s is an integer greater than 1 and less than n, and z is greater than or equal to 1 and less than or equal to z the integer.

When z=1, in the first row layer in the s row layers, the decoding device sequentially corresponds to the s data groups in each data set in the first row layer with the s storage nodes. For example, the first data group in each dataset in the first row layer corresponds to the first storage node in the s storage nodes, and the second data group in each dataset in the first row layer corresponds to the s storage node. The 2nd storage node in the storage node corresponds, and so on. For example, s=3, z=1, the first data set in the first row layer in the codeword shown in FIG. 9 includes data group 1-1, data group 1-2 and data group 1-3, data group 1-1 corresponds to storage node 1, data group 1-2 corresponds to storage node 2, and data group 1-3 corresponds to storage node.

When z=2, in the second row layer in the s row layers, the decoding device shifts the s data groups in each data set in the second row layer to the right by 1 bit, and sequentially Corresponds to s storage nodes. For example, after the data group in each data set in the second row layer is shifted to the right by 1 bit, the second data group in each data set corresponds to the first storage node in the s storage nodes. The third data group corresponds to the fourth storage node in the s storage nodes, and so on, the s-1 data group in each data set corresponds to the s storage node in the s storage nodes, and each The s-th data group in the data sets corresponds to the first storage node in the s-storage nodes. For example, s=3, z=2, the first data set in the second row layer in the codeword shown in FIG. 9 includes data group 2-1, data group 2-2 and data group 2-3, data group 2-2 corresponds to storage node 1, data group 2-3 corresponds to storage node 2, and data group 2-1 corresponds to storage node 3.

When z=3, in the third row layer in the s row layers, the decoding device shifts the s data groups in each data set in the third row layer to the right by 2 bits, and sequentially Corresponds to s storage nodes. For example, after the data group in each data set in the third row layer is shifted to the right by 2 bits, the third data group in each data set corresponds to the first storage node in the s storage nodes. The 4th data group corresponds to the 4th storage node in the s storage nodes, and so on, the s-2nd data group in each data set corresponds to the sth storage node in the s storage nodes, and each The s-1th data group in each data set corresponds to the first storage node in the s storage nodes, and the s-th data group in each data set corresponds to the second storage node in the s storage nodes. For example, s=3, z=3, the first data set in the third row layer in the codeword shown in FIG. 9 includes data group 3-1, data group 3-2 and data group 3-3, data group 3-3 corresponds to storage node 1, data group 3-1 corresponds to storage node 2, and data group 3-2 corresponds to storage node 3.

In order to further illustrate the process of establishing the correspondence between each data group in the plurality of data sets and the plurality of storage nodes using the cyclic shift method, refer to a TPC code provided by an embodiment of the present application shown in FIG. 10 . Schematic diagram of the correspondence between the data group in the word and the storage node. Figure 10 takes the code word as TPC(256,239)(256,239) as an example, TPC(256,239)(256,239) is a two-dimensional block code, and each row subcode or each column subcode in TPC(256,239)(256,239) is To extend the BCH(256,239) subcode, the total code length of TPC(256,239)(256,239) is 256*256=65536 bits. With m=n=16, the decoding device divides the 65536 data in TPC(256,239)(256,239) into 256 data groups, which are respectively data groups 1 to 256, and each data group includes 16*16 data. If s=4, the decoding device combines four adjacent data in the same row layer into one data set according to the row layer, so that one data set includes 4*16*16=1024 data. For the first four row layers in TPC(256,239)(256,239), the decoding device sequentially corresponds each data set in the first row layer to four storage nodes, for example, a data set in the first row layer includes Data groups 1 to 4 correspond to storage nodes 1 to 4 respectively. After the decoding device shifts each data group in the second row layer to the right by 1 bit, it corresponds to 4 storage nodes in sequence, for example, starting from the second data group in the second row layer, data groups 18 to 21 Corresponding to storage nodes 1 to 4 in sequence, ..., data groups 30 to 32 correspond to storage nodes 1-3 in sequence, after data group 32 is corresponding, loop to the first data group of the second row layer, from the second The first data group of the row layer starts, and the data group 17 corresponds to the storage node 4, so that a cyclic shift process of shifting 1 bit to the right is formed in the second row layer. And from the perspective of data sets, the second row layer includes 4 data sets, which respectively include data sets 17 to 20, data sets 21 to 24, data sets 25 to 28, and data sets 29 to 32. , after shifting 1 bit to the right, the second data group (data groups 18, 22, 26 and 30) in these four data sets all correspond to storage node 1, and the third data group ( Data groups 19, 23, 27, and 31) all correspond to storage node 2, and the fourth data group ( data groups 20, 24, 28, and 32) in these four data sets all correspond to storage node 3. These four data sets correspond to storage node 3. The first data group ( data groups 17 , 21 , 25 and 29 ) in the collection all correspond to the storage node 1 . Then, according to the shifting method of the second row layer, the decoding device shifts each data group in the third row layer to the right by 2 bits, and then sequentially corresponds to the four storage nodes, and converts the data groups in the fourth row layer to the right. After each data group is shifted to the right by 3 bits, it sequentially corresponds to 4 storage nodes.

It should be noted that, in a possible implementation manner, the process shown in this step 804 is performed by a data processing node in the decoding device. The process shown in the above steps 803-804 is a process in which the decoding device stores multiple data groups of the codeword in multiple storage addresses, and the multiple storage addresses belong to multiple storage nodes. In addition, the decoding device only needs to store the code word once, and does not need to store it multiple times.

805. For any data set in the plurality of data sets, the decoding device reads a plurality of data groups of the any data set from the plurality of storage nodes.

The decoding device determines a plurality of target storage addresses for storing the any data set from a plurality of storage addresses of a plurality of storage nodes, the plurality of target storage addresses belong to different storage nodes, and one target storage address stores There is one dataset in either dataset. The decoding device reads a plurality of data groups of the arbitrary data set from the plurality of target storage addresses. In a possible implementation manner, in one data read cycle, the decoding device reads out a data group in any one of the data sets from the plurality of target storage addresses, respectively. Taking any data set as the first data set of the first row layer in the codeword shown in Fig. 10 as an example, in one data read cycle, the decoding device reads from the four storage addresses of the four storage nodes. Data sets 1 to 4 are each read out.

In a possible implementation manner, the process shown in this step 805 is performed by the data distribution node in the decoding device.

806. The decoding device decodes the multiple data groups in the any data set.

In a possible implementation manner, the decoding device simultaneously decodes a plurality of data groups in any data set, wherein, for any data group in the plurality of data groups in the any data set, the decoding The coding device decodes m*n pieces of data in any data group to obtain m*n pieces of decoded data.

In a possible implementation manner, the process shown in this step 806 is completed by the data distribution node and the decoding node in the decoding device. If the current decoding process is a row decoding process, after the data distribution node reads a plurality of data groups in any of the data sets, the decoding device determines m rows of each data group in the plurality of data groups data groups, and respectively send the m line data groups of each data group to the m first decoding nodes, so that each first decoding node can receive one line data group among the plurality of data groups. It can be understood that each first decoding node can receive a plurality of row data groups, and each received row data group is from one of the plurality of data groups respectively. After each first decoding node receives multiple row data groups, it decodes n pieces of data in each received row data group.

If the current decoding process is a column decoding process, after the data distribution node reads a plurality of data groups in any data set, the decoding device determines n columns of each data group in the plurality of data groups data groups, and respectively send the n column data groups of each data group to the n second decoding nodes, so that each second decoding node can receive one column data group among the plurality of data groups. It can be understood that each second decoding node can receive a plurality of column data groups, and each received column data group is respectively from one data group among the plurality of data groups. After each second decoding node receives multiple column data groups, it performs column decoding on m data in each received column data group.

Take Figure 10 as an example. If the current decoding process is a row decoding process, after the data distribution node reads the data groups 1 to 4 in a data set, the data distribution node sends the 16 lines of data in each data group in the data groups 1 to 4. Up to 16 first decoding nodes, each first decoding node decodes one line of data (including 4*16=64 data) of data groups 1 to 4. If the current decoding process is a column decoding process, after the data distribution node reads the data groups 1 to 4, the data distribution node sends the 16-column data of each data group in the data groups 1 to 4 to the 16th Two decoding nodes, each second decoding node decodes a column of data (including 4*16=64 data) of data groups 1 to 4.

It should be noted that when the decoding device performs the process shown in the above steps 805-806 for each data set belonging to the codeword, the decoding device performs decoding on one dimension of the codeword. Process complete. The decoding device jumps to execute the process shown in steps 805-806, and continues the decoding process of the codeword in another dimension, thereby realizing the iterative decoding process of the codeword. For example, after the decoding device performs the row-wise decoding process for each data set in the codeword through the above steps 805-806, the decoding device continues to perform the above steps 805-806 to perform the decoding process on the code word. A column-wise decoding process is performed for each data set in the word.

Since the m*n pieces of data stored in any data set satisfy the distribution of the row dimension and the distribution of the column dimension, in some embodiments, the decoding device uses the data set as the basic unit of iterative decoding. By performing the above steps 805-806, the decoding device does not need to wait for other data sets to complete the column-direction decoding after completing the row-direction decoding for any data set, and the decoding device directly performs the above-mentioned steps 805-806. The encoded data set is subjected to a column decoding process, thereby realizing an iterative decoding process of a data set in the codeword.

807. For any data group in the any data set, the decoding device modifies m*n pieces of data in the storage addresses where the any data group is stored into m*n pieces of decoded data, the m*n pieces of decoded data. The pieces of decoded data are decoded data of the read m*n pieces of data.

The process shown in this step 807 is the same as the process shown in the above-mentioned step 505, and this step 807 is not described in detail here in this embodiment of the present application.

In the method provided by the embodiment of the present application, a decoding device is used to decode a plurality of data groups in a data set in parallel, thereby improving the parallel degree of decoding and the decoding efficiency. And it can support flexible design of any degree of parallelism without being limited by the length of subcodes, and support high-traffic FEC design requirements.

For different types of codewords, the decoding device can store different types of codewords in the storage address according to the storage mode of the data group. Moreover, the decoding device can also decode multiple data groups in a codeword in parallel, but due to the different structures of different types of codewords, when the decoding device decodes multiple data groups in parallel, the decoding device reads multiple data groups. Group methods may vary. For TCP codewords, the decoding device can use the process shown in FIG. 8 to decode multiple data groups in TCP codewords in parallel, while the construction of convolutional codes is different from that of TCP codewords. When decoding multiple data groups of a convolutional code in parallel, the way of reading multiple data groups is different from that of reading multiple data groups in a TCP codeword. To further illustrate the difference, refer to the decoding process of a convolutional code provided by an embodiment of the present application shown in FIG. 11 .

1101. A decoding device acquires a codeword.

Optionally, the codeword is an OFEC codeword, such as the OFEC codeword shown in FIG. 2 . The process shown in this step 1101 is the same as the process shown in the above-mentioned step 501, and this step 1101 is not described repeatedly in this embodiment of the present application.

1102. The decoding device divides the data in the codeword into multiple data groups according to the granularity of m rows and n columns.

If the codeword is an OFEC codeword, and each block in the OFEC codeword includes m*n data in m rows and n columns, the decoding device determines each block in the OFEC codeword as a piece of data group, so that the decoding device divides the data in the codeword into multiple data groups according to the granularity of m rows and n columns.

Taking the OFEC codeword shown in FIG. 12 as an example, FIG. 12 is a schematic diagram of a decoding process of an OFEC codeword provided by an embodiment of the present application. The OFEC codeword in FIG. 12 includes 24 row layers and 8 column layers, that is, the OFEC codeword includes 24*8 blocks, and each block includes 16*16 data, the decoding device converts 24* The 8 blocks are respectively used as a data group, which are data groups 1 to 192 respectively.

1103. The decoding device combines the multiple data sets into multiple data sets.

The decoding apparatus combines a plurality of data in adjacent row layers or adjacent column layers into one data set. Taking FIG. 12 as an example, the decoding device combines the data groups 1, 2, 9 and 10 in the OFEC codeword into a data set, and the data groups 1, 2, 9 and 10 are in two adjacent row layers and 2 adjacent column layers.

1104. The decoding device stores data groups in the plurality of data sets in a plurality of storage addresses in a plurality of storage nodes, wherein the plurality of data groups in the same data set are stored in storage addresses in different storage nodes, and Data groups of different column layers in two adjacent datasets in an OFEC frame pair are stored in different storage nodes.

An OFEC frame pair is two adjacent OFEC frames, that is, four consecutive line layers of OFEC codewords. It can be seen from the OFEC codeword shown in FIG. 2 that the multiple oblique short columns in a subcode of the OFEC codeword are located in different OFEC frames, and all are located in the even-numbered row layer or the odd-numbered row layer. During the decoding process, if you want to decode multiple oblique short columns in the subcode at the same time, you need to ensure that these multiple short columns are stored in different storage nodes, and these multiple short columns may be located in one OFEC frame. Therefore, in the process of storing multiple data groups in each data set into multiple storage nodes, the decoding device also needs to ensure that two adjacent data in the same OFEC frame pair are Centralized data groups of different column layers are stored in different storage nodes.

In a possible implementation manner, for two adjacent data sets in an OFEC frame pair, if the two data sets both include s data sets, the decoding device converts the zth data in the two data sets into The groups are all stored on the zth storage node among the s storage nodes. Taking Figure 10 as an example, the first 4 row layers in the OFEC codeword shown in Figure 10 are an OFEC frame pair. In this OFEC frame pair, data groups 7, 8, 15 and 16 are data set 1, and data group 23. , 24, 31 and 32 are dataset 2, and the datasets 1 and 2 are adjacent in this OFEC frame pair. The decoding device stores both the first data group (data group 7 and data group 23) in the data sets 1 and 2 in the storage node 1, and the decoding device stores the second data group in the data sets 1 and 2 (Data set 8 and data set 24) are both stored in storage node 2, and the decoding device stores the third data set (data set 15 and data set 31) in data sets 1 and 2 in storage node 3. The decoding device stores the fourth data group (data group 16 and data group 32 ) in both data sets 1 and 2 in the storage node 4 . This storage method not only ensures that each data group in dataset 1 or 2 is stored in different storage nodes, but also ensures that data groups in different column layers in datasets 1 and 2 are also stored in different storage nodes.

1105. For a plurality of adjacent target subcodes among the plurality of subcodes, the decoding device determines at least one target data set related to the plurality of target subcodes from the plurality of data sets, and the at least one target data set includes a At least one target data group related to the plurality of target subcodes, one target data group includes partial data of some target subcodes in the plurality of target subcodes.

The codeword includes a plurality of subcodes. At least one target data set related to the plurality of target subcodes is a data set including data of the plurality of target subcodes.

It can be seen from the above formulas (1)-(2) that a subcode in the OFEC codeword is located at multiple row layers and multiple column layers, and the first half of the data of the subcode is located at non-adjacent row layers and column layers, The data of the second half of the subcode is located in the adjacent row layer and column layer. When the decoding device stores the OFEC codeword, it combines multiple data in adjacent row layers or adjacent column layers into a data set, so a data set in a data set may include multiple subcodes. part of the data. In the process of decoding a plurality of target subcodes, the decoding device may first determine at least one target data set related to the plurality of target subcodes from the plurality of data sets, and then perform the decoding on the at least one target data set. At least one target data set associated with the plurality of target subcodes is decoded.

For any target subcode in the plurality of target subcodes, the decoding device determines the position information of each data in any target subcode based on the above formulas (1)-(2), and according to any target subcode The location information of each data in any target subcode determines the data group where each data is located, and uses any data group where the data in any target subcode is located as the target data group, and the data to which the target data group belongs set as the target dataset.

Taking Figure 10 as an example, if there are at least two target data sets, namely Data Set 1 and Data Set 2 in Figure 10, the 16 short columns in the data set 1 in the column upwards of the data set 1 belong to 16 The adjacent subcodes are respectively subcodes 1 to 16, and the 16 short columns in the column direction in the data group 23 in the data set 2 also belong to the subcodes 1 to 16, respectively. In this dataset 1, the 16 short columns in the column-up direction in the data group 16 belong to 16 adjacent subcodes, which are sub-codes 17 to 32, respectively. The 16 short columns in the column-up direction in the data group 23 in this data set 2 Also belong to subcodes 17 to 32, respectively. It can be seen from FIG. 10 that the data in the adjacent subcodes 1 to 32 are all located in the data sets 1 and 2, then the decoding device determines the data sets 1 and 2 as the target data sets related to the subcodes 1 to 32, Determine the data set 6 in the data set 1 and the data set 23 in the data set 2 as the target data sets related to the subcodes 1 to 16, and determine the data set 16 in the data set 1 and the data set 2. The data set 31 is determined as the target data set associated with the subcodes 17 to 32.

If there is at least one target data set, it is the data set 3 in FIG. 10, wherein the data set 3 includes data 167, 168, 175 and 176, and 16 of the data sets 167 and 168 in the horizontal direction of each data set The short columns belong to subcodes 1 to 16, respectively, and the 16 short columns in the lateral direction of each data set in data sets 175 and 176 belong to subcodes 17 to 32, respectively. Then the decoding device determines the data set 3 as a target data set related to the subcodes 1 to 32, and determines the data sets 167 and 168 in the data set 3 as the target data related to the subcodes 1 to 16 group, the data groups 175 and 176 in the data set 3 are determined as the target data groups related to the subcodes 17 to 32.

1106. The decoding device reads the target data group in the at least one target data set from the plurality of storage nodes.

Since multiple target data sets in the at least one target data set are stored in different storage nodes, the decoding device reads out one target data set from multiple storage nodes in one data reading cycle.

1107. The decoding device decodes the read target data group.

The read target data group includes at least one odd target data group and at least one even target data group, that is, the target data group in the at least one target data set, wherein one odd target data group is located in the codeword. The target data group of the odd-numbered line layer, and an even target data group is the target data group located at the even-numbered line layer of the codeword. Taking the at least one target data group as the data groups 8, 16, 23 and 31 in FIG. 10 as an example, the data groups 8 and 23 are respectively located in the first row layer and the third row layer of the codeword, then the data group 8 and 23 are odd target data groups, while data groups 16 and 31 are located at the 2nd row layer and the 4th row layer of the codeword, respectively, so data groups 16 and 31 are both even target data groups.

Since among the multiple target data groups, some target data groups are located in the odd-numbered row layer, some target data groups are located in the even-numbered row layer, and the target data groups located in the odd-even row layer belong to different target subcodes, therefore, the decoding device is in the In the process of decoding the plurality of target subcodes, two groups of parity decoding nodes can respectively decode the target data groups located at the odd-numbered row layer and at the even-numbered row layer. In a possible implementation manner, this step 1107 is implemented by the process shown in the following steps 11071-11072.

Step 11071: The decoding device sends the at least one odd target data group to a plurality of third decoding nodes, and the plurality of third decoding nodes decodes the received odd target data group.

The plurality of third decoding nodes are a group of decoding nodes for decoding odd data groups in the codeword, and may also be referred to as odd decoding nodes. Wherein, the odd data group in the codeword is the data group in the odd row layer in the codeword. For example, in the codeword shown in FIG. 10, all data groups of the first row layer are odd data groups, and odd target data groups are also odd data groups.

If the at least one odd target data group is located in the first half of multiple target subcodes related to the at least one odd target data group, for any odd target data group in the at least one odd target data group, the data distribution node is based on The position of m*n data in the codeword in any odd target data group determines n column data groups, each column data group belongs to a target subcode and is a short column in a target subcode ; The data distribution node sends the n column data groups to the n third decoding nodes respectively. Each third decoding node can receive at least one column data group, and at least one column data group is respectively from at least one odd target data group, and each third decoding node can receive at least one column data group on the received at least one column data group. to decode.

Taking the at least one target data group as the data groups 8, 16, 23 and 31 in FIG. 10 as an example, since the data groups 8 and 23 are in the first half of the target subcodes 1 to 16 and are both odd target data groups, the The data distribution node divides the 16*16 data of each data group in data groups 8 and 16 into 16 column data groups, and sends the 16 column data groups of each data group to 16 third decoding points respectively , so that each third decoding node can receive 2 column data groups and decode 16*2 data in the 2 column data groups.

If the at least one odd target data group is located in the second half of a plurality of target subcodes related to the at least one odd target data group, for any odd target data group in the at least one odd target data group, the data distribution node Based on the positions of m*n data in the codeword in any odd target data group, m row data groups are determined, each row data group belongs to a target subcode, a short column in a target subcode ; The data distribution node sends the m line data groups to the m third decoding nodes respectively. Each third decoding node can receive at least one row data group, and at least one row data group is respectively from at least one odd target data group, and each third decoding node performs processing on the received at least one row data group. decoding.

Taking the at least one target data group as the data groups 167, 168, 175 and 176 in FIG. 10 as an example, since the data groups 167 and 168 are in the first half of the target subcodes 17 to 32 and are both odd target data groups, the The data distribution node divides the 16*16 data of each data group in the data groups 167 and 168 into 16 line data groups, and sends the 16 line data groups of each data group to the 16 third decoding points respectively , so that each third decoding node can receive 2 row data groups and decode 16*2 data in the 2 row data groups.

11072. The decoding device sends the at least one even target data group to a plurality of fourth decoding nodes, and the plurality of fourth decoding nodes decodes the received even target data group.

The plurality of fourth decoding nodes are a group of decoding nodes for decoding even data groups in the codeword, and may also be referred to as even decoding nodes. Wherein, the even data group in the codeword is the data group in the even-numbered row layer in the codeword. For example, in the codeword shown in FIG. 10, all data groups of the second row layer are even data groups, and even target data groups are also even data groups.

If the at least one even target data group is located in the first half of a plurality of target subcodes related to the at least one even target data group, for any even target data group in the at least one even target data group, the data distribution node is based on The position of m*n data in the codeword in any even target data group determines n column data groups, each column data group belongs to a target subcode and is a short column in a target subcode ; The data distribution node sends the n column data groups to the n fourth decoding nodes respectively. Each fourth decoding node can receive at least one column data group, and at least one column data group is respectively from at least one even target data group, and each fourth decoding node performs processing on the received at least one column data group. decoding.

Taking the at least one target data group as the data groups 8, 16, 23 and 31 in FIG. 10 as an example, since the data groups 16 and 31 are in the first half of the target subcodes 1 to 16, and are all even target data groups, The data distribution node divides the 16*16 data of each data group in the data groups 16 and 31 into 16 column data groups, and sends the 16 column data groups of each data group to 16 fourth decodings respectively Therefore, each fourth decoding node can receive 2 column data groups, and decode 16*2 data in the received 2 column data groups.

If the at least one even target data group is located in the second half of a plurality of target subcodes related to the at least one even target data group, for any even target data group in the at least one even target data group, the data distribution node Based on the positions of m*n data in the codeword in any even target data group, m row data groups are determined, and each row data group belongs to a target subcode; the data distribution node uses the m rows The data groups are respectively sent to the m fourth decoding nodes. Each fourth decoding node can receive at least one row data group, and at least one row data group is respectively from at least one even target data group, and each fourth decoding node performs processing on the received at least one row data group. decoding.

Taking the at least one target data group as the data groups 167, 168, 175 and 176 in FIG. 10 as an example, since the data groups 175 and 176 are in the first half of the target subcodes 17 to 32 and are both even target data groups, the The data distribution node divides the 16*16 data of each data group in the data groups 175 and 176 into 16 line data groups, and sends the 16 line data groups of each data group to the 16 fourth decoding points respectively , so that each fourth decoding node can receive 2 row data groups, and decode 16*2 data in the received 2 row data groups.

It should be noted that the above steps 11071 and 11072 are parallel steps, and there is no sequence in the execution order, so that the decoding device can decode the multiple target data groups at the same time, which improves the degree of parallelism in the decoding process. and decoding efficiency.

In a possible implementation manner, the decoding device first determines at least one odd target data group and at least one even target data group among the plurality of target data groups based on the row layer where the plurality of target data groups are located, and then , and then perform the above steps 11071-11072. Optionally, the process of determining the at least one target data group and the at least one even target data group by the decoding device is performed by a data distribution node in the decoding device.

1108. For any data group in the read target data group, the decoding device modifies m*n pieces of data in the storage address of the any data group into m*n pieces of decoded data, The m*n pieces of decoded data are decoded data of the m*n pieces of data in any data group.

The process shown in this step 1108 is the same as the process shown in the above-mentioned step 505, and here, this embodiment of the present application does not repeat the description of this step 1108.

In the method provided by the embodiment of the present application, a decoding device is used to decode multiple target data groups in parallel, thereby improving the parallelism of decoding and the decoding efficiency. And it can support flexible design of any degree of parallelism without being limited by the length of subcodes, and support high-traffic FEC design requirements.

FIG. 13 is a schematic structural diagram of a decoding apparatus provided by an embodiment of the present application. The apparatus 1300 includes:

The storage module 1301 is used to store the codeword in a plurality of storage addresses, one storage address is used to store m*n data of m rows and n columns in the codeword, and both the m and the n are greater than or equal to an integer of 1;

a reading module 1302, configured to read m*n pieces of data stored in the any storage address from the any storage address for any storage address in the plurality of storage addresses;

The decoding module 1303 is used for decoding the read m*n data.

Optionally, the decoding module 1303 includes:

The first determination unit is used to determine m row data groups based on the positions of the read m*n data in the codeword if the current decoding process is a row-wise decoding process, one row The data group includes n data in the read m*n data, and the n data are located in the same row in the codeword;

The first decoding unit is used for decoding the m row data groups.

Optionally, the first decoding unit is used for:

The m line data groups are respectively sent to m first decoding nodes, and each first decoding node decodes the received line data groups.

Optionally, the decoding module 1303 includes:

The second determining unit is configured to, if the current decoding process is a column-oriented decoding process, determine n column data groups, one column The data group includes m data in the read m*n data, and the m data are located in the same column in the codeword;

The second decoding unit is used for decoding the n column data groups.

Optionally, the second decoding unit is used for:

The n column data sets are respectively sent to n second decoding nodes, and each second decoding node decodes the received column data sets.

Optionally, the apparatus 1300 further includes:

A modification module, configured to modify m*n pieces of data stored in any of the storage addresses into m*n pieces of decoded data, where the m*n pieces of decoded data are the read m*n pieces of data Data decoded data.

Optionally, the storage module 1301 is used for:

According to the granularity of m rows and n columns, the data in the codeword is divided into multiple data groups, and one data group includes m*n data of m rows and n columns in the codeword;

The plurality of data groups are stored in the plurality of storage addresses, and one data group is stored in one storage address.

Optionally, the multiple storage addresses belong to the same storage node.

Optionally, the multiple storage addresses belong to multiple storage nodes;

The storage module 1301 is used for:

combining the plurality of data sets into a plurality of data sets, and one data set includes a plurality of adjacent data sets in the plurality of data sets;

The data groups in the multiple data sets are stored in the multiple storage addresses, wherein one storage address stores one data group, and multiple data groups in the same data set are stored in storage addresses in different storage nodes.

The reading module 1302 is also used for:

For any data set in the plurality of data sets, a plurality of data groups of the any data set are read from the plurality of storage nodes.

Optionally, the apparatus 1300 further includes:

a determining module, configured to, for a plurality of adjacent target subcodes among the plurality of subcodes, determine at least one target data set related to the plurality of target subcodes from the plurality of data sets, the at least one target data set The data set includes at least one target data group related to the plurality of target subcodes, and one target data group includes data of part of the target subcodes in the plurality of target subcodes;

The reading module 1302 is further configured to read the target data group in the at least one target data set from the plurality of storage nodes;

The decoding module 1303 is further configured to decode the read target data group.

Optionally, the read target data group includes at least one odd target data group and at least one even target data group, an odd target data group is a target data group located at an odd-numbered row layer of the codeword, an even target data group The target data group is the target data group located at the even-numbered row layer of the codeword;

The decoding module 1303 is used for:

sending the at least one odd target data group to a plurality of third decoding nodes, and decoding the received odd target data group by the plurality of third decoding nodes;

The at least one even target data group is sent to a plurality of fourth decoding nodes, and the received even data group is decoded by the plurality of fourth decoding nodes.

All the above-mentioned optional technical solutions can be combined arbitrarily to form optional embodiments of the present disclosure, which will not be repeated here.

It should be noted that when the decoding apparatus provided in the above embodiment decodes the code word, only the division of the above functional modules is used as an example for illustration. In practical applications, the above functions can be allocated to different functions as required. Module completion means dividing the internal structure of the device into different functional modules to complete all or part of the functions described above. In addition, the decoding method embodiments provided by the above embodiments belong to the same concept, and the specific implementation process thereof is detailed in the method embodiments, which will not be repeated here.

Embodiments of the present application also provide a computer program product or computer program, where the computer program product or computer program includes computer instructions, where the computer instructions are stored in a computer-readable storage medium, and the processor of the decoding device stores the computer-readable storage medium. The medium reads the computer instructions, and the processor executes the computer instructions, so that the processor executes the above decoding method.

Those of ordinary skill in the art can understand that all or part of the steps of implementing the above-mentioned embodiments can be accomplished by hardware, and can also be accomplished by hardware related to program instructions, and the program can be stored in a computer-readable storage medium. The storage medium can be read-only memory, magnetic disk or optical disk, etc.

The above are only optional embodiments of the present application, and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the protection of the present application. within the range.

Claims (26)

1. A decoding method, characterized in that the method comprises:

   The codeword is stored in a plurality of storage addresses, and one storage address is used to store m*n data of m rows and n columns in the codeword, and both the m and the n are integers greater than or equal to 1;

   For any storage address in the plurality of storage addresses, read m*n data stored in the any storage address from the any storage address;

   Decode the m*n pieces of data read.
2. The method according to claim 1, wherein the decoding the read m*n data comprises:

   If the current decoding process is a row-wise decoding process, m row data groups are determined based on the positions of the read m*n data in the codeword, and one row data group includes the read data n pieces of data among the m*n pieces of data received, the n pieces of data are located in the same row in the codeword;

   The m row data sets are decoded.
3. The method according to claim 2, wherein the decoding the m row data groups comprises:

   The m line data groups are respectively sent to m first decoding nodes, and each first decoding node decodes the received line data groups.
4. The method according to claim 1, wherein the decoding the read m*n data comprises:

   If the current decoding process is a column-oriented decoding process, based on the positions of the read m*n data in the codeword, determine n column data groups, and one column data group includes the read data m data in the received m*n data, the m data are located in the same column in the codeword;

   The n column data sets are decoded.
5. The method according to claim 4, wherein the decoding the n column data groups comprises:

   The n column data sets are respectively sent to n second decoding nodes, and each second decoding node decodes the received column data sets.
6. The method according to any one of claims 1-5, characterized in that, after the read m*n pieces of data are decoded, the method further comprises:

   Modify m*n pieces of data stored in any of the storage addresses into m*n pieces of decoded data, and the m*n pieces of decoded data are the decoded pieces of the read m*n pieces of data. data.
7. The method according to any one of claims 1-6, wherein the storing the codeword in a plurality of storage addresses comprises:

   According to the granularity of m rows and n columns, the data in the codeword is divided into multiple data groups, and one data group includes m*n data of m rows and n columns in the codeword;

   The plurality of data groups are stored in the plurality of storage addresses, and one data group is stored in one storage address.
8. The method according to claim 7, wherein the plurality of storage addresses belong to the same storage node.
9. The method according to claim 7, wherein the plurality of storage addresses belong to a plurality of storage nodes;

   The storing the plurality of data groups at the plurality of storage addresses includes:

   combining the plurality of data sets into a plurality of data sets, and one data set includes a plurality of adjacent data sets in the plurality of data sets;

   The data groups in the multiple data sets are stored in the multiple storage addresses, wherein one storage address stores one data group, and multiple data groups in the same data set are stored in storage addresses in different storage nodes.
10. The method according to claim 9, wherein after storing the data groups in the plurality of data sets in the plurality of storage addresses, the method further comprises:

    For any data set in the plurality of data sets, a plurality of data groups of the any data set are read from the plurality of storage nodes.
11. The method according to claim 9, wherein the codeword includes multiple subcodes, and after the codeword is stored in multiple storage addresses, the method further comprises:

    For a plurality of adjacent target subcodes among the plurality of subcodes, at least one target data set related to the plurality of target subcodes is determined from the plurality of data sets, the at least one target data set including the at least one target data group related to the plurality of target subcodes, and one target data group includes data of part of the target subcodes in the plurality of target subcodes;

    reading target data groups in the at least one target data set from the plurality of storage nodes;

    Decode the read target data set.
12. The method according to claim 11, wherein the read target data group includes at least one odd target data group and at least one even target data group, and an odd target data group is an odd number located in the codeword The target data group of the row layer, an even target data group is the target data group located at the even-numbered row layer of the codeword;

    The decoding of the read multiple target data groups includes:

    sending the at least one odd target data group to a plurality of third decoding nodes, and decoding the received odd target data group by the plurality of third decoding nodes;

    The at least one even target data group is sent to a plurality of fourth decoding nodes, and the received even data group is decoded by the plurality of fourth decoding nodes.
13. A decoding device, characterized in that the device comprises:

    The storage module is used to store the codeword in a plurality of storage addresses, one storage address is used to store m*n data of m rows and n columns in the codeword, and both the m and the n are greater than or equal to 1 the integer;

    a reading module, configured to read m*n data stored in the any storage address from the any storage address for any storage address in the plurality of storage addresses;

    The decoding module is used to decode the read m*n data.
14. The apparatus according to claim 13, wherein the decoding module comprises:

    The first determination unit is configured to determine m row data groups based on the positions of the read m*n data in the codeword if the current decoding process is a row-wise decoding process, one row The data group includes n data in the read m*n data, and the n data are located in the same row in the codeword;

    The first decoding unit is used for decoding the m row data groups.
15. The apparatus according to claim 14, wherein the first decoding unit is used for:

    The m line data groups are respectively sent to m first decoding nodes, and each first decoding node decodes the received line data groups.
16. The apparatus according to claim 13, wherein the decoding module comprises:

    The second determining unit is configured to, if the current decoding process is a column-oriented decoding process, determine n column data groups, one column The data group includes m data in the read m*n data, and the m data are located in the same column in the codeword;

    The second decoding unit is used for decoding the n column data groups.
17. The apparatus according to claim 16, wherein the second decoding unit is used for:

    The n column data sets are respectively sent to n second decoding nodes, and each second decoding node decodes the received column data sets.
18. The device according to any one of claims 13-17, wherein the device further comprises:

    A modification module, configured to modify m*n pieces of data stored in any of the storage addresses into m*n pieces of decoded data, where the m*n pieces of decoded data are the read m*n pieces of data Data decoded data.
19. The device according to any one of claims 13-18, wherein the storage module is used for:

    According to the granularity of m rows and n columns, the data in the codeword is divided into multiple data groups, and one data group includes m*n data of m rows and n columns in the codeword;

    The plurality of data groups are stored in the plurality of storage addresses, and one data group is stored in one storage address.
20. The apparatus according to claim 19, wherein the plurality of storage addresses belong to the same storage node.
21. The apparatus of claim 19, wherein the plurality of storage addresses belong to a plurality of storage nodes;

    The storage module is used for:

    combining the plurality of data sets into a plurality of data sets, and one data set includes a plurality of adjacent data sets in the plurality of data sets;

    The data groups in the multiple data sets are stored in the multiple storage addresses, wherein one storage address stores one data group, and multiple data groups in the same data set are stored in storage addresses in different storage nodes.
22. The device according to claim 21, wherein the reading module is further used for:

    For any data set in the plurality of data sets, a plurality of data groups of the any data set are read from the plurality of storage nodes.
23. The apparatus of claim 21, wherein the apparatus further comprises:

    a determining module, configured to, for a plurality of adjacent target subcodes among the plurality of subcodes, determine at least one target data set related to the plurality of target subcodes from the plurality of data sets, the at least one target data set The data set includes at least one target data group related to the plurality of target subcodes, and one target data group includes data of part of the target subcodes in the plurality of target subcodes;

    The reading module is further configured to read the target data group in the at least one target data set from the plurality of storage nodes;

    The decoding module is also used for decoding the read target data group.
24. The apparatus according to claim 23, wherein the read target data group includes at least one odd target data group and at least one even target data group, and an odd target data group is an odd number located in the codeword The target data group of the row layer, an even target data group is the target data group located at the even-numbered row layer of the codeword;

    The decoding module is used for:

    sending the at least one odd target data group to a plurality of third decoding nodes, and decoding the received odd target data group by the plurality of third decoding nodes;

    The at least one even target data group is sent to a plurality of fourth decoding nodes, and the received even data group is decoded by the plurality of fourth decoding nodes.
25. A decoding device, characterized in that the decoding device includes a processor and a memory, and the memory stores at least one piece of program code, and the program code is loaded and executed by the processor to implement the method as claimed in claim 1 To the operations performed by the decoding method according to any one of claims 12.
26. A computer-readable storage medium, wherein at least one piece of program code is stored in the storage medium, and the program code is loaded and executed by a processor to implement the method according to any one of claims 1 to 12. The operations performed by the decoding method.

Images