CN102903378B - For the two-dimentional run length constraint coding/decoding device and method of elongated code check - Google Patents

Abstract

The present invention provides a two-dimensional run-length constrained coding and decoding device and method for variable-length code rates, the device includes an encoder and a decoder, and the encoder includes a one-dimensional data stream buffering and grouping module, a 2×2 Codeword generation module, two-dimensional word cell array construction module, 2×2 codeword writing two-dimensional word cell array module, and finally two-dimensional data array output through the 2×2 codeword writing two-dimensional word cell array module For a two-dimensional data recording device; the decoder includes a two-dimensional array buffer module, a two-dimensional word unit page construction module, a two-dimensional word unit reading module, a 2×2 codeword to one-dimensional data mapping module, and a one-dimensional data stream Assemble the module, and finally assemble the module output through the one-dimensional data flow therein. The invention can solve the technical problem of the two-dimensional data array structure in which the run length of the data 0 is at least equal to 1 and at most equal to 3 between adjacent data 1 in the horizontal and vertical directions of the two-dimensional data array.

Description

Apparatus and method for two-dimensional run-length constrained encoding and decoding for variable-length code rate

technical field

The invention belongs to the fields of data storage and communication. Specifically, when storing binary data on a plane composed of rectangular grids, whether in the horizontal direction or in the vertical direction, between adjacent data "1", the run length of data "0" is at least equal to 1, and at most No more than 3 codec methods and devices for their use. The present invention is a coding and decoding technology with a variable length code rate, that is, the ratio of the length of the output two-dimensional code word to the length of the input one-dimensional data changes with the change of the input condition, and the code rate changes.

Background technique

In current mainstream magnetic tape and optical disc storage technologies, data is recorded on a storage medium along a track (Track). In these memories, in order to improve the reliability of the storage system, necessary constraints are usually imposed on the binary data sequence. The most common restriction is the RLL (d, k) run length limited constraint (Runlength Limited Constraint), where d represents the minimum value of the run length of "0" between continuous data "1", and k represents its maximum value. By selecting the appropriate parameter d, the highest transmission frequency can be controlled to achieve the purpose of reducing intersymbol crosstalk; when reading data, the parameter k can ensure sufficient hopping frequency to meet the demand for clock synchronization control. The famous Miller code, that is, the improved frequency modulation code MFM (US Patent No. 3,108,261, published on October 22, 1963), is one of the representatives. In the MFM code, between adjacent data "1", the run length of "0" is at least equal to 1, and at most no more than 3.

In recent years, new types of memory that record data in two-dimensional pages have emerged, such as holographic memory, two-dimensional optical discs, and two-dimensional pattern records. This type of memory has a common feature: data is stored on the recording medium in the form of two-dimensional pages each time, and when reading data, it is also performed in the form of a full page. Figure 1 depicts the schematic layout of 8×8 binary bit data in this new type of two-dimensional memory. In this example, a square grid represents 1 bit of data information, that is, black represents "1" and white represents "0". Each binary data bit "0" or "1" has 4 bits of information and itself directly Adjacent (that is, commonly referred to as top, bottom, left, and right). Similar to the constraints on binary sequences in one-dimensional storage technology, in this two-dimensional page record memory, it is also necessary to constrain the two-dimensional data array in the row (horizontal) direction and column (vertical) direction limit. It is not difficult to see that the aforementioned modified frequency modulation code (MFM) is no longer effective in this two-dimensional situation, that is, the two-dimensional binary data array composed of data "0" and "1" is in the row direction and column direction. When d=1 and k=3 in the direction, the run length of the data "0" must be at least equal to 1 and not exceed 3 at the same time. Patent application 201110148840.9 proposes a method of coding along the diagonal of a two-dimensional grid with protected words. Based on the method of this patent, a binary rectangular data array composed of data "1" and data "0" can be constructed in rows (horizontally) A two-dimensional data array in which the minimum value of the run length of the data "0" in the two directions of direction and column (vertical) direction is equal to 1 and the maximum is no more than 3 at the same time. However, the technology provided by this patent is a fixed code rate, and the ratio of the code word length to the input data is a fixed value, that is, the code rate is equal to 0.25.

Contents of the invention

The technical problem to be solved by the present invention is to provide a codec device and method of use for two-dimensional run-length constraints of variable-length code rates. The formed two-dimensional data array is in the two directions of horizontal (row direction) and vertical (column direction), between adjacent data "1", the run length of data "0" is at least equal to 1, and the maximum is equal to 3 two-dimensional For the technical problem of data array construction, compared with the code rate given in patent 201110148840.9 which is equal to 0.25, the average code rate of the present invention has increased by about 12.3%, reaching about 0.2808.

In order to achieve the above-mentioned purpose of the invention, the technical solution adopted by the present invention is: a coding and decoding device for a variable-length code rate two-dimensional run length constraint, including an encoder and a decoder, and the encoder includes:

A one-dimensional data stream buffering and grouping module, which divides it into several one-dimensional data words according to 2 bits as a group while buffering and receiving the one-dimensional data stream;

2×2 codeword generation module, which generates corresponding 2×2 two-dimensional constraint codewords from several one-dimensional data words divided by the one-dimensional data stream buffer and grouping module according to the encoding and decoding rule table, and input them in the order of generation 2*2 code words are written into the 2*2 word unit buffer of the two-dimensional word unit array module;

Two-dimensional word cell array construction module, which divides the two-dimensional array into several two-dimensional word cell arrays composed of two-dimensional word cells with a size of 2×2, and juxtaposes each two-dimensional word cell array of the two-dimensional word cell array The unit write flag is equal to the writable flag;

2×2 codewords are written into the two-dimensional word unit array module, which reads out the 2×2 two-dimensional constrained codeword sequences in the 2×2 codeword generation module from the buffer of this module in sequence, and along the Diagonal rows of the two-dimensional word cell array, writing 2×2 two-dimensional constrained code words into the two-dimensional word cell array row by row;

The four modules of the above-mentioned encoder are connected in sequence, and finally the two-dimensional data array is output to the two-dimensional data recording device through the 2×2 code word therein written into the two-dimensional word cell array module;

The decoder includes:

A two-dimensional array buffer module, which is mainly used to temporarily store a two-dimensional data array with a size of 2N×2M read from a two-dimensional data recording device;

Two-dimensional word unit page construction module, which is mainly used to divide the 2N×2M two-dimensional data array in the two-dimensional array buffer module into N×M two-dimensional word unit arrays with 2×2 as the minimum read unit, juxtaposing each word unit read flag of the two-dimensional word unit array as a readable flag;

A two-dimensional word unit reading module, which is mainly used to read a suitable 2×2 two-dimensional constrained codeword from the two-dimensional word unit array, and temporarily store the generated 2×2 two-dimensional constrained codeword sequence;

2×2 codeword to one-dimensional data mapping module, this module is mainly used to read 2×2 two-dimensional constrained codeword sequence from two-dimensional word unit reading module, according to the encoding and decoding rule table, convert 2×2 two-dimensional The constrained codeword is decoded into a one-dimensional data word whose size is equal to 2 bits according to the sequence;

A one-dimensional data stream assembly module, which is mainly used to assemble the 2-bit one-dimensional data words generated by the 2×2 codeword to one-dimensional data mapping module into a one-dimensional data stream according to the sequence of generation;

The five modules of the above decoder are connected in sequence, and finally output through the one-dimensional data stream assembly module.

The encoding and decoding method of the two-dimensional run-length-limited encoding and decoding device used for variable-length code rate, realizes the encoding process through the encoder, and includes the following steps in turn:

S101) The one-dimensional data stream buffering and grouping module evenly divides the input binary one-dimensional data stream into a number of one-dimensional 2-bit data words in groups of 2 bits, and inputs the 2-bit data words into a 2×2 code The data buffer of the word generation module is temporarily stored;

S102) The 2×2 codeword generation module reads the one-dimensional 2-bit data words from the data buffer one by one, and selects the output scheme of the codec rule table to convert the one-dimensional 2-bit data words into 2×2 two-dimensional constrained codewords;

S103) The two-dimensional word unit array construction module divides a two-dimensional data array with a size of 2N×2M into groups of N×M word units according to the storage size of the two-dimensional data recording device. The two-dimensional word unit array of the same 2×2 two-dimensional word unit, juxtaposing each 2×2 two-dimensional word unit write flag as true, wherein N and M are positive integers;

S104) 2×2 codewords are written into the two-dimensional word unit array module. The 2×2 two-dimensional constrained codeword obtained in step S102 is based on the 2×2 two-dimensional word unit obtained in step S103, along the two-dimensional word unit The two-dimensional constrained codewords are sequentially written into the two-dimensional word units with a size equal to 2×2 from the upper right corner to the lower left corner in the diagonal direction of the array, and at the same time, the right and The directly adjacent two-dimensional word units below are selectively filled; if all the two-dimensional word units of the diagonal row are written, then enter the next step, and if all the two-dimensional constraint codewords have been filled, the encoding process is directly ended;

S105) Judging whether the next diagonal row has not been written, if so, go to step S104; if all 2×2 two-dimensional word units of the two-dimensional word unit page have been filled with two-dimensional constraint code words, then enter the next step;

S106) Judging whether there are unencoded one-dimensional data words, if so, go to step S102; otherwise, end the encoding process.

The encoding and decoding method for a codec device with a variable-length code rate two-dimensional run length constraint is implemented by a decoder to realize the decoding process, which includes the following steps in turn:

S201) Temporarily store the two-dimensional data array read from the two-dimensional data recording device in the two-dimensional data buffer module;

S202) The two-dimensional word unit page construction module reads a two-dimensional data array with a size of 2N×2M, divides the two-dimensional data array evenly with 2×2 as the minimum reading unit, and forms N×M identically sized A two-dimensional word unit array of 2×2 two-dimensional word units, juxtaposing each two-dimensional word unit of the two-dimensional word unit array to read the flag as a readable flag true, wherein N and M are positive integers;

S203) The two-dimensional word unit reading module takes the 2×2 two-dimensional word unit of the two-dimensional word unit array as a reference, and according to the encoding order of the encoder, along the diagonal row of the two-dimensional word unit array, from the upper right corner to the lower left corner Selectively read the two-dimensional constrained codewords in the two-dimensional word cell array according to the reading sign of each two-dimensional word unit in turn, if all the two-dimensional constrained codewords of this row are read, then enter the next step, if the current two-dimensional constrained codewords are all read The two-dimensional constrained codeword of the dimension word unit array has been read, then go to step S205;

S204) Judging whether there is a next diagonal row that has not been decoded, if so, go to step S203; if all the two-dimensional constraint codewords of all diagonal rows of the two-dimensional word unit page have been read, go to the next step;

S205) The 2×2 codeword-to-one-dimensional data mapping module decodes the 2×2 two-dimensional constraint codewords read from the two-dimensional word unit reading module according to the generation order according to the decoding table corresponding to the selected codec table Code into a corresponding one-dimensional 2-bit data word, and enter the next step after completion;

S206) The one-dimensional data stream assembly module assembles the obtained one-dimensional 2-bit data words according to the corresponding decoding sequence, obtains the corresponding one-dimensional binary data stream, and completes the decoding.

In the method described, the method of selectively filling the adjacent two-dimensional word units on the right and below of the currently written 2×2 two-dimensional word unit in step S104 is as follows:

a) If the two-dimensional constrained codeword written into the current 2×2 two-dimensional word unit does not belong to the diagonal matrix or sub-diagonal matrix, then judge the right side, The writing flag of the two-dimensional word unit below is true or false; if it is true, the two-dimensional constrained codeword that is filled in the two-dimensional word unit position corresponding to the flag bit is a diagonal matrix or a sub-diagonal matrix; if it is false or If not defined, the two-dimensional constraint codeword will not be filled;

After completing the above-mentioned writing process, set the current 2×2 two-dimensional word unit and the two-dimensional word unit writing flag filled with the two-dimensional constraint codeword as false, indicating that these positions cannot be allowed to write the two-dimensional constraint codeword again in the future;

b) If the two-dimensional constraint code word written into the current 2×2 two-dimensional word unit belongs to a diagonal matrix or a sub-diagonal matrix, after writing the two-dimensional constraint code word at the position of the current two-dimensional word unit, write it In flag is set to false.

5. The method according to claim 3, wherein the method for selectively reading the two-dimensional constraint codeword in the two-dimensional word unit page in step S203 is:

a) If the reading flag of the currently read two-dimensional word unit is true, read the two-dimensional constraint code word in the two-dimensional word unit, and set the reading flag of the two-dimensional word unit to false, that is, in the future Not allowed to read; if the currently read two-dimensional constraint code word does not belong to the diagonal matrix or sub-diagonal matrix, set the two-dimensional word unit adjacent to the right and below the current read word unit position to read The flag is false, that is, reading is not allowed;

b) If the reading flag of the currently read two-dimensional word unit is false, transfer to read the data in the next two-dimensional word unit in the diagonal row where the current two-dimensional word unit is located.

Technical effect of the present invention is as follows:

1) The encoder is a variable-length code rate. The code rate of the encoder is the ratio of the number of bits of the output codeword to the number of bits of the input codeword. The higher the code rate, the better the encoding effect. In the present invention, the data is encoded according to the context of the encoder. One-dimensional 2-bit data words output two-dimensional constrained codewords with different lengths. This variable-length codeword encoding method can fully utilize the correlation between adjacent data, so the code rate is higher than that of existing encoding methods . The code rate is 12.3% higher.

2) The compilation and decoding rule table is simple. In the present invention, the one-dimensional 2-bit data word can select different output methods according to the context and context of the data to meet the constraints. Through simple decoding equations, the decoding of one-dimensional 2-bit data words can be realized.

Description of drawings

Figure 1 is an 8×8 two-dimensional data array that satisfies (1,3) run-length constraints in both directions.

Figure 2 is a two-dimensional character unit composed of 2×2 square grids.

Fig. 3 is a schematic diagram of the coordinates of the two-dimensional word cell array and the numbering of diagonal rows.

FIG. 4 is a schematic diagram of the read/write sequence of data in the two-dimensional word cell array.

FIG. 5 is a schematic diagram of coordinates for calculating code rates.

Fig. 6 is a functional block diagram of an encoder with a variable code rate.

Fig. 7 is a functional block diagram of a decoder with a variable code rate.

FIG. 8 is an embodiment of the two-dimensional data array structure based on the first encoding and decoding rules in Tables 1 and 2.

FIG. 9 is a schematic diagram of the changes of the two-dimensional word unit array during the encoding process based on the first encoding and decoding rules in Tables 1 and 2.

Fig. 10 is a decoding embodiment of a two-dimensional word cell array based on the first encoding and decoding rules in Tables 1 and 2.

FIG. 11 is a schematic diagram of changes in the decoding process of the two-dimensional word cell array based on the first encoding and decoding rules in Tables 1 and 2.

FIG. 12 is an embodiment of the two-dimensional data array structure based on the second encoding and decoding rules in Tables 3 and 4.

FIG. 13 is a decoding embodiment of a two-dimensional word cell array based on the second encoding and decoding rules in Tables 3 and 4.

Figure 14 is the first adjacent type of two-dimensional constrained codewords along the diagonal.

Figure 15 is the second adjacent type of two-dimensional constrained codewords along the diagonal.

Figure 16 is the third adjacent type of two-dimensional constrained codewords along the diagonal.

Figure 17 is the fourth adjacent type of two-dimensional constrained codewords along the diagonal.

Detailed ways

Aiming at the problems raised by the background technology, the present invention provides two encoding and decoding methods with two-dimensional run-length constraints with variable-length code rates, which can make the binary bit information meet the pre-specified requirements in the environment of two-dimensional data storage and communication. Constraints. That is, when the binary data "0" and "1" are written in an array composed of rectangular grids, it is also possible to construct between adjacent data "1", whether in the row (horizontal) direction or the column (vertical) direction , a binary data array whose run length of the data "0" is at least equal to 1 and at most no more than 3. The difference from patent 201110148840.9 is that the present invention is a coding and decoding technology with a variable length code rate, that is, the ratio of the length of the output two-dimensional code word to the length of the input one-dimensional data changes with the input conditions, and the code rate is variable length. By adopting the technical solution given by the present invention, the relationship between the two-dimensional rectangular grid and the two-dimensional constraint codeword is fully utilized, and the relationship is maximized; the average coding rate is increased from 0.25 in patent 201110148840.9 to 0.2808 , the code rate has increased by about 12.3%. In other words, under the same conditions, the system storage capacity can be increased by about 12.3% by adopting the present invention.

The invention is suitable for a data storage system that records binary data according to two-dimensional square grids. Typical representatives of storage systems include holographic holographic memory, two-dimensional optical disk and other two-dimensional new memory. An object of the present invention is to provide two variable-length code rate two-dimensional run-length-constrained coding methods. Another object of the present invention is to provide two decoding methods with two-dimensional run length constraints corresponding to the two variable-length code rate encoding methods. Another object of the invention is to provide corresponding encoding and decoding means.

The two-dimensional run-length-limited (1,3) constrained encoding and decoding method of the variable-length code rate, taking the first encoding and decoding rule shown in Tables 1 and 2 as an example, includes the following steps:

Table 1 The coding table of the first coding rule

Table 2 Decoding table of the first encoding and decoding rules

Step 1: Divide the input one-dimensional binary data stream into several one-dimensional 2-bit data words according to a group of two bits. The value of each one-dimensional 2-bit data word has the following four possibilities: 00, 01, 10, 11. And input the one-dimensional 2-bit data word into the one-dimensional data buffer for temporary storage, and turn to step 2;

Step 2: Divide the two-dimensional data pages with a size equal to 2N×2M into a group of 2×2 into N×M two-dimensional word unit pages of the same size, and arrange each word unit of the two-dimensional word unit array Write flag is true, then go to step 3;

The third step: read the one-dimensional 2-bit data word sequence from the one-dimensional buffer of step one, and according to the output scheme three of the encoding and decoding rule table described in Table 1, 2 or Table 3, 4, the one-dimensional buffer of the buffer The 2-bit data word sequence is coded and mapped one by one into a 2×2 two-dimensional constrained codeword sequence, and then the 2×2 two-dimensional constrained codeword sequence is based on the diagonal behavior of the two-dimensional word unit array generated in step 2, along the For the 45° diagonal row of the two-dimensional word cell array, write the two-dimensional constraint code words of this 2×2 into the corresponding word units of the two-dimensional word cell array from the upper right corner to the lower left corner row by row, and turn to step 4;

Table 3 Encoding table of the second coding rule

Table 4 Decoding table of the second encoding and decoding rules

Step 4: On the basis of step (3), if the coordinates of the currently written two-dimensional word cell array are , and the write flag is true, and the 2×2 two-dimensional constraint codeword to be written at the same time is (If step (3) adopts the second encoding and decoding rules described in tables 3 and 4, then use instead), then in position to write time, and judge and The writing flag bit of the two-dimensional word unit position is true or false, if it is false, it is skipped, otherwise it is and Write two-dimensional constraint code words according to the situation (When step (3) adopts the second encoding and decoding rules described in Table 3 and 4, then use replace), and finally reset the writing flag of the corresponding coordinate position of each two-dimensional word unit in this step to false, and go to step five;

; ; (A)

; ; (B)

Step 5: If the coordinates of the currently written two-dimensional cell array are , and the 2×2 two-dimensional constrained codeword to be written is, directly at the position Write (if step (3) adopts table 3, 4 encoding and decoding rules, then the two places of step (5) Use evenly replace), and set the writing flag of this coordinate as false, indicating that data has been written at this coordinate, and go to step 6;

Step 6: Repeat step 4 until all 2×2 two-dimensional constraint words generated in step (2) are written into the N×M two-dimensional word cell array.

(Taking the first encoding and decoding rules described in Tables 1 and 2 as an example) the decoding steps are as follows:

Step 1: Read a binary data array whose size is equal to 2N×2M, divide the two-dimensional data array into N×M two-dimensional word unit arrays with 2×2 as a two-dimensional word unit, and place the two-dimensional word unit Each word unit read flag of the cell array is true;

Step 2: Read the two-dimensional constrained codeword row by row along the 45° diagonal of the two-dimensional word cell array from the upper right corner to the lower left corner;

Step 3: On the basis of Step 2, if the current coordinates If the read flag is true, read the corresponding 2×2 two-dimensional constraint codeword, and judge whether it is equal to , if yes, set the coordinates and The read flag of is false;

The fourth step: according to whether each word unit of the two-dimensional word unit array reads the true or false of the flag bit, judge whether all the two-dimensional word units of the two-dimensional word unit array are all read, if turn to step five, otherwise turn to step two;

Step 5: Decode the 2 × 2 two-dimensional constrained codewords obtained in step 3 into one-dimensional 2-bit data words according to the decoding tables in Tables 1 and 2 or Tables 3 and 4;

Step 6: Assemble the one-dimensional 2-bit data words obtained in step five into a one-dimensional data stream according to the order of generation and output it, and the decoding is completed.

In order to achieve the above object, the present invention also provides an encoder for implementing the above data encoding method, including:

A one-dimensional data stream buffering and grouping module, which divides the data into several one-dimensional 2-bit data words according to the size of the two-dimensional array while buffering the one-dimensional data stream;

2 × 2 code word generation module, this module generates several one-dimensional 2-bit data words of the one-dimensional data stream buffering and grouping module according to the output scheme three of the encoding and decoding rule table described in tables 1 and 2 or tables 3 and 4 The corresponding 2×2 two-dimensional constraint codewords are input to the 2×2 word unit buffer of the two-dimensional word unit array module according to the order of generation;

The two-dimensional word cell array construction module, which divides the two-dimensional array into several word cell arrays composed of 2×2 arrays, and juxtaposes the writing flag bit of each two-dimensional word cell of the two-dimensional word cell array to be equal to the writable flag "0";

2×2 codewords are written into the two-dimensional word unit array module, which reads out the 2×2 two-dimensional constrained codeword sequences in the 2×2 codeword generation module from the buffer of this module in sequence, and along the 45° diagonal rows of the two-dimensional word cell array, and write 2×2 two-dimensional constrained codewords into the two-dimensional array row by row;

In order to achieve the above object, the present invention also provides an encoder for implementing the above data decoding method, including:

Two-dimensional array buffer module, which is mainly used to temporarily store the 2N×2M two-dimensional data array read from the two-dimensional data recording device;

Two-dimensional word unit page construction module, which is mainly used to divide the 2N×2M two-dimensional data array in the two-dimensional array buffer module into N×M two-dimensional word unit arrays with 2×2 as the minimum read unit, The read flag bit of each word unit of the two-dimensional word cell array is arranged as a readable flag "0";

Two-dimensional word unit reading module, this module is mainly used to read suitable 2×2 two-dimensional constraint codewords from the current two-dimensional word unit array, and temporarily store all 2×2 two-dimensional code words generated by this two-dimensional word unit array. Dimensional constraint code word;

2×2 codewords to one-dimensional data mapping module, this module is mainly used to read 2×2 two-dimensional constrained codewords from the two-dimensional word unit reading module according to the sequence of generation, according to Table 1, 2 or Table 3, 4. The decoding table of the encoding and decoding rules, which decodes the 2×2 two-dimensional constraint codewords into one-dimensional data words whose size is equal to 2 bits according to the sequence;

A one-dimensional data stream assembly module, which is mainly used to assemble the 2-bit one-dimensional data words generated by the 2×2 codeword to one-dimensional data mapping module into a one-dimensional data stream according to the sequence of generation.

Below in conjunction with accompanying drawing and embodiment the present invention is described in further detail:

As shown in Figure 1, a two-dimensional data array produced according to the present invention is described, whether in the horizontal direction or in the vertical direction, the run length of data "0" between adjacent data "1" The length satisfies the constraints that the minimum is equal to 1 and the maximum is no more than 3. The left figure of Figure 1 is the recording mode of two-dimensional data in the holographic memory, and the right is the digital mode of the binary data array.

As shown in FIG. 2 , a two-dimensional word unit constructed of 2×2 square grids is the smallest unit of a codec two-dimensional data array.

As shown in FIG. 3 , a schematic diagram of coordinates of a two-dimensional word cell array and numbers of diagonal rows is described. Each basic unit of the array in the figure is a two-dimensional word unit, and the two-dimensional word unit is composed of a 2×2 square grid as shown in FIG. 2 .

As shown in FIG. 4, it describes the read/write sequence of data in the two-dimensional word cell array. That is, in the two-dimensional word unit, the reading/writing of data is carried out line by line along the 45° diagonal, and each line of the 45° diagonal is from the upper right corner to the lower left corner according to the Data is read/written sequentially.

As shown in Tables 1 and 2, Table 1 shows the coding table of the first encoding and decoding rule. In this coding table, the one-dimensional data words 00, 01, and 10 have three schemes for two-dimensional constraint code words according to the context. available for coding output; the one-dimensional data word 11 has only one kind of two-dimensional constraint code word for coding output. Table 2 shows the decoding rule table corresponding to the encoding rule based on Table 1. In the decoding rule table, if the 2-bit one-dimensional data word is used Indicates that the 2×2 two-dimensional constrained codeword uses express. Based on this encoding and decoding rule, the basic encoding equation of the 2×2 two-dimensional constrained codeword is as follows:

Among them, × ₁ to × ₂ are single bits of a one-dimensional data word. Y ₁₁ to Y ₂₂ are bits of the two-dimensional code word. Also according to the basic encoding equation, in the process of decoding, the decoding equation of the one-dimensional 2-bit data word is as follows:

;

As shown in Formula A, Formula A provides the classification of four 2×2 two-dimensional constrained codewords based on the first encoding and decoding rule table described in Tables 1 and 2. , , These three two-dimensional constraint words belong to one category, and are respectively used , and express; Separately classified into the second category, with Represents, for the convenience of description, referred to as a diagonal matrix.

Tables 3 and 4 show the second encoding and decoding rules. Table 3 shows the basic encoding table based on the second encoding and decoding rule. In this encoding rule, the one-dimensional data words 00, 01, and 10 adopt three other two-dimensional constrained codeword schemes for encoding output according to the context; The one-dimensional data word 11 is the same as the first scheme, and there is only one encoding output scheme to choose from. Table 4 shows the decoding table based on the coding rules described in Table 3. If a one-dimensional 2-bit data word is still used Indicates that the 2×2 two-dimensional constrained codewords still use express. Based on the second encoding and decoding rule, the basic encoding equation of the 2×2 two-dimensional constrained codeword is as follows:

According to the second basic encoding and decoding rules, the second decoding equation of the one-dimensional 2-bit data word is as follows:

;

As shown in formula B, what formula B provides is the classification of four 2×2 two-dimensional constrained codewords based on the second encoding and decoding rule table described in Tables 3 and 4. , , These three two-dimensional constraint words belong to one category, and are respectively used , and express; Separately classified into the second category, with Represents, for the convenience of description, referred to as subdiagonal matrix.

As shown in Figure 5, the figure describes how to calculate the average code rate of the two variable-length coding methods based on the two encoding and decoding rules based on Tables 1 and 2 and Tables 3 and 4. i.e. compute the coordinates in the cell array The probability of writing the two-dimensional constrained codeword for the occurrence of . Taking the first encoding and decoding method described in Tables 1 and 2 as an example, the average code rate of the specific calculation encoding method is as follows :

First, the one-dimensional binary data sequence is divided into groups of two by two, and then according to the one-dimensional to two-dimensional mapping relationship, the one-dimensional 2-bit data words are sequentially mapped into 2×2 two-dimensional word units, that is, the one-dimensional binary data The sequence is converted into a two-dimensional 2×2 word unit data sequence. In order to facilitate the description of the problem, take the first variable code rate method as an example, that is, divide the 2×2 two-dimensional word unit data into two types , that is, the two-dimensional 2×2 word unit data sequence consists of Consists of. here ,and ; .

In general, The probability is equal to ,and The probability is equal to . After the above mapping transformation, the one-dimensional binary data sequence is transformed into a two-dimensional word unit data sequence. Obviously, after the above transformation, the entropy of the array sequence composed of four 2×2 two-dimensional constrained codewords is equal to ,Right now . In a two-dimensional cell array, use represents the write value of the course cell, where . In a 2D cell array coordinates equal to The write value of express. At the same time, because the 2×2 two-dimensional constrained codewords are written into the two-dimensional word cell array line by row according to the 45° diagonal row, so in the Nth diagonal row, the 2×2 two-dimensional constrained codewords are written along the diagonal There are four adjacent types of lines as shown in Figure 14 to Figure 17. In the first type in Figure 14 , ; In the second type in Figure 15, ; Figure 16 in the third type , ; Figure 17 in the fourth type , where i=1,2,3. Obviously the average code rate of the encoder

in in any position The probability of writing a 2D constrained codeword. According to the order of writing two-dimensional unit data, the position Arbitrarily write two-dimensional constraint codewords , according to the constraints, that is, at position and location The cell data values on the should both be equal to . In conjunction with accompanying drawing 5 of the description, there are:

    

    

    

Combining equations (2), (3), (4), (5), (6), (7), and (8), we have the following equations:

     

in . Solving equations ,have

Substitute into equation (1), that is

Usually the binary data stream is independent and identically distributed, that is, the probability of "0" and "1" is equal to 1/2, then the binary data stream is directly converted into two dimensional constraint codeword sequence, the two-dimensional constraint codeword The probability of occurrence is equal to , according to the above calculation formula (11), so the average code rate of variable code rate . Apparently, the average code rate obtained by adopting the encoding and decoding rule method of the second table 3 and 4 is the same as that of the first method.

As shown in Figure 6, the encoding process of the encoder is as follows: first, the one-dimensional data stream is input to the one-dimensional data stream buffering and grouping module. While buffering the one-dimensional data stream, the module divides the data into several equal-length packets, and the packet size is equal to 2-bit data words. After the one-dimensional data stream passes through the grouping module, firstly, the data group whose size is equal to 2 bits is input to the 2×2 code word generation module, and then according to the basic coding table of the encoding and decoding rule table described in Table 1, 2 or Table 3, 4 , first generate 2×2 two-dimensional constrained codewords, and input the formed 2×2 two-dimensional constrained codeword sequence to the two-dimensional word unit cache of the 2×2 codeword write two-dimensional word unit array module. On the other hand, the two-dimensional word unit array construction module constructs a two-dimensional word unit according to 2×2 grids, and divides the two-dimensional array with a size equal to 2N×2M into N×M two-dimensional word unit arrays. The generation of the two-dimensional data array is completed by writing the two-dimensional code word into the array module. After the above operations are completed, the 2×2 codeword is written into the two-dimensional word cell array module to read the 2×2 constraint codeword from the cache of this module, and according to the basic coding rules in Tables 1 and 2 or Tables 3 and 4, The 2×2 constraint codewords are written row by row along the 45° diagonal rows of the two-dimensional word cell array in FIG. 3 . During the writing process of each diagonal row, the writing sequence is as described in FIG. 4 , that is, writing one by one from the upper left corner to the lower right corner of the current diagonal row. According to the classification of 2×2 two-dimensional constrained codewords by formula A or formula B, the corresponding two-dimensional constrained codewords are selected and written into the two-dimensional word units at the corresponding positions, so as to realize the encoding of one-dimensional data streams into two-dimensional data arrays. Taking the first encoding and decoding rules described in Tables 1 and 2 as an example, the encoding steps include:

(1) Divide the input one-dimensional data stream into several one-dimensional data words according to a group of 2 bits;

(2) Encode the one-dimensional 2-bit data word generated in step (1) into a 2×2 two-dimensional constraint code according to the output scheme 3 of the encoding and decoding rule table described in Table 1 and 2 Character;

(3) A two-dimensional word unit is formed with a 2×2 grid, and the two-dimensional 2N×2M array is divided into N×M two-dimensional word unit arrays with two-dimensional word units as units, and each of the two-dimensional word unit arrays is juxtaposed The word unit writable flag is true "0";

(4) On the basis of the two-dimensional word unit array formed in step (3), the 2×2 two-dimensional constrained codeword sequence obtained in step (2) is classified according to formula A to the 2×2 two-dimensional constrained codewords, Along the 45° diagonal row of the two-dimensional word cell array, data is written row by row from the upper right corner to the lower left corner;

(5) On the basis of step (4), if the coordinates of the currently written two-dimensional word cell array are , and the 2×2 two-dimensional constraint codeword to be written is , at position to write time, judgment and The writing flag of the two positions is true or false, if it is false, skip it, otherwise in and Write two-dimensional constraint codewords in the two positions according to the situation, and reset the writing flag of the corresponding coordinate position to false "1", indicating that the writing operation of these positions has been completed, and go to step (6);

(6) If the coordinates of the currently written two-dimensional word cell array are , and the 2×2 two-dimensional constrained codeword to be written is, directly at the position Write, and set the write flag of this coordinate to false "1", indicating that the write operation at this position is completed;

(7) Repeat step (4) until all 2×2 two-dimensional constraint words generated in step (2) are written into the N×M two-dimensional word cell array;

As shown in Figure 7, the decoding process of the encoder is as follows: first read the two-dimensional data array from the two-dimensional recording device, then read the data page by page from the two-dimensional array buffer module, and input it to the two-dimensional word unit page construction module, according to 2×2 is a two-dimensional word unit, divide the two-dimensional data array into two-dimensional word unit pages, and then input the divided two-dimensional word unit pages to the two-dimensional word unit reading module to read the 2×2 two-dimensional Constraint codewords, input two-dimensional constraint codewords to 2×2 codewords to one-dimensional data mapping module, obtain corresponding one-dimensional 2-bit data words, input one-dimensional 2-bit data words to one-dimensional data stream assembly module, The one-dimensional original data stream is assembled into a one-dimensional original data stream according to the order of the 2-bit data words, thereby completing the transformation from the two-dimensional constrained array to the one-dimensional data stream, and realizing the decoding process. Still taking the first encoding and decoding rules described in Tables 1 and 2 as an example, its decoding steps include:

(1) Read the two-dimensional data array from the two-dimensional recording device, and input the two-dimensional array to the two-dimensional array buffer module for storage;

(2) Divide the two-dimensional data array into corresponding two-dimensional word cell arrays according to 2×2 as a basic unit, and set the reading flag of the corresponding two-dimensional word cell array as true;

(3) Read the two-dimensional constrained codeword from the upper right corner to the lower left corner along the 45° diagonal line of the two-dimensional word unit array in an incremental manner;

(4) On the basis of step (3), if the current coordinate If the read flag is true, read the corresponding 2×2 two-dimensional constraint codeword, and judge whether it is equal to , if yes, set the coordinates and The reading flag of is false, otherwise continue to step (3) to read the two-dimensional constraint codeword of the next coordinate until the entire two-dimensional word cell array is read, and then go to step (5);

(5) Decode the 2×2 two-dimensional constrained codeword sequence obtained in step (4) into one-dimensional 2-bit data words one by one according to the decoding tables in Tables 1 and 2;

(6) Assemble the one-dimensional 2-bit data words obtained in step (5) into a one-dimensional data stream in sequence and output it, and the decoding is completed.

In order to illustrate the application method of the coding rules, Fig. 8 and Fig. 9 list the coding embodiment 1 based on the first coding and coding rules described in Tables 1 and 2 of the present invention. FIG. 10 and FIG. 11 illustrate the decoding embodiment 2 of the present invention based on the first encoding and decoding rules described in Tables 1 and 2. Both embodiment 1 and embodiment 2 are classified according to the encoding and decoding rule tables described in tables 1 and 2 and the two-dimensional constraint code words described in formula A. The size of the two-dimensional array is equal to 8×8, and the corresponding two-dimensional word cell array The size is equal to 4×4. According to Table 3, 4 and the second encoding and decoding rule table described in Formula B and the two-dimensional constraint code word analysis, its encoding and decoding embodiments are shown in Fig. 12 and Fig. 13 .

Coding Example 1:

Figure 8 shows the specific encoding process. The one-dimensional data stream 00110111111011001011 is combined and divided into 10 2-bit data words from right to left according to the order of 2 bits. The 10 one-dimensional 2-bit data words are as follows: 11, 01, 00, 11, 01 , 11, 11, 10, 11, 00. According to the output scheme three of the basic coding table described in Tables 1 and 2, encoding is performed to generate 10 2×2 two-dimensional constraint codeword sequences, and the generated 2×2 two-dimensional constraint codewords are arranged as follows according to the corresponding relationship:

, , , , , , , , ,

Then write the above-mentioned 10 2×2 two-dimensional constrained code words into the previously constructed 4×4 two-dimensional word cell array. Before implementation, each unit of the 4×4 two-dimensional word cell array is written into the flag position 0, that is, the unit flag bit is equal to 0, indicating that the unit at the current position can write data, and equal to 1, indicating that it cannot be written. Then the first 2×2 two-dimensional constraint codeword Write into the two-dimensional word unit at coordinates (0,0) along the 0th row of the diagonal line of the two-dimensional word unit array, and the written two-dimensional word unit array is shown in FIG. 9 (1). Then the second 2×2 two-dimensional constraint codeword Write to the position of the two-dimensional word unit whose coordinates are equal to (0,1) in the first row of the diagonal line, and according to formula A, belong , because the write flag at coordinates (1,1) and (0,2) is equal to 0, so write at the word unit position at coordinates (1,1) and (0,2) , and at the same time write the flag positions (1,1) and (0,2) to 1, that is, the one-dimensional data word 01 is encoded as . After writing the second one-dimensional 2-bit data word 01, the two-dimensional word cell array is shown in Figure 9 (2). The third 2×2 two-dimensional constraint codeword Continue to write to the corresponding word cell array along the first row along the 45° diagonal, that is, coordinates (1,0). because also belongs to , and the writing flag of the coordinate (1,1) is equal to 1, and the writing flag of the coordinate (2,0) is equal to 0, so according to the coding rules, the Write to the word cell array at coordinate (1,0), and write at coordinate (2,0) , on the other hand, since the write flag at coordinate (1,1) is equal to 1, skip at coordinate (1,1) in the current case. After the above process is completed, set the write flags of these two coordinates to 1. That is, when the one-dimensional 2-bit data word 00 is written into the array, according to the basic encoding table of the first coding rule described in Tables 1 and 2, according to the output scheme 2, 00 is encoded into , that is, after writing the third one-dimensional 2-bit data word 00, the two-dimensional word cell array is shown in Figure 9 (3). Since there is no free space in the second row of the diagonal of the array, the fourth 2×2 two-dimensional constraint codeword The coordinates of the writing position in the two-dimensional word cell array are (0,3) in the third row of the diagonal array. because belong , so the one-dimensional 2-bit data word 11 is encoded into , that is, when the fourth one-dimensional 2-bit data word 00 is encoded and written into the two-dimensional word cell array, the two-dimensional word cell array becomes as shown in Figure 9 (4). Repeat the above process until the 10th 2-bit data word 00 is written into (3, 3) of the 6th diagonal row of the two-dimensional word array unit. The output scheme three of the coding rule basic coding table of table 1, 2, is coded into , the 2×2 two-dimensional constraint codeword Write to the array (3,3), that is to complete the two-dimensional constrained array encoding process for one-dimensional data stream to one page.

Example 2:

Figure 10 shows the specific decoding process. First read data from the two-dimensional recording device to the two-dimensional array buffer module. In this embodiment, the 8×8 two-dimensional data array is as follows:

Then the two-dimensional data array is divided into two-dimensional word cell arrays whose size is equal to 4×4 with 2×2 as a basic unit as follows:

Then read the 2×2 two-dimensional constraint codeword row by row along the diagonal rows of the two-dimensional word cell array. Before reading data, set the read flag bit for each word unit of the 4×4 two-dimensional word unit array , the flag bit "0" indicates the coordinates of the two-dimensional word unit The 2×2 two-dimensional constraint codeword at can be read, and the flag bit “1” means it cannot be read, here .

First read the 2×2 two-dimensional constraint codeword from the coordinate (0,0) of the 0th row of the diagonal , the read two-dimensional word cell array is shown in Figure 11(1), and then read the two-dimensional constrained codeword at row 1 (0,1) ,because belong , so the word unit coordinates of the array are respectively equal to (0,2) and (1,1) read flag position 1, that is, , the read two-dimensional word cell array is shown in Figure 11 (2), and the blank in the figure indicates that the two-dimensional constraint codeword has been read. Since there is still data in the first row of the diagonal row that has not been read, continue to read the two-dimensional constraint codeword of the word unit (1,0) of the diagonal row from the upper right corner to the lower left corner ,because belong , so set the reading flag position of unit coordinates (2,0) and (1,1) to 1, since the previous , so only , Fig. 11(3) describes the reading of two-dimensional constrained codewords after the situation. The above process of reading the two-dimensional array is repeated until all the data in the two-dimensional array is read. The two-dimensional constrained codewords read above are arranged as follows according to the reading order:

, , , , , , , , ,

In this embodiment, the read two-dimensional data word is decoded into a one-dimensional 2-bit data word according to the decoding table of the encoding and decoding rules of Tables 1 and 2, namely: ; ; ; . Finally, the one-dimensional 2-bit data words 11, 01, 00, 11, 01, 11, 11, 10, 11, 00 are assembled into a user data stream 00110111111011001011 and output to complete the two-dimensional decoding of user data.

Based on the second encoding and decoding rules in Tables 3 and 4, a similar encoding and decoding work can also be completed. See the encoding example in FIG. 12 and the decoding example in FIG. 13 for specific embodiments. Although the disclosed embodiments of the present invention have been described in detail, it should be understood that there are many changes, substitutions and modifications in the specific embodiments of the present invention without departing from the two coding rules based on Tables 1 and 2 and Tables 3 and 4.

Claims (7)

1. A two-dimensional run-length-limited codec device for variable-length code rates, characterized in that it includes an encoder and a decoder, and the encoder includes: A one-dimensional data stream buffering and grouping module, which divides it into several one-dimensional data words according to 2 bits as a group while buffering and receiving the one-dimensional data stream; 2×2 codeword generation module, which generates corresponding 2×2 two-dimensional constraint codewords from several one-dimensional data words divided by the one-dimensional data stream buffer and grouping module according to the encoding and decoding rule table, and input them in the order of generation 2*2 code words are written into the 2*2 word unit buffer of the two-dimensional word unit array module; Two-dimensional word cell array construction module, which divides the two-dimensional array into several two-dimensional word cell arrays composed of two-dimensional word cells with a size of 2×2, and juxtaposes each two-dimensional word cell array of the two-dimensional word cell array The unit write flag is writable; 2×2 codewords are written into the two-dimensional word unit array module, which reads out the 2×2 two-dimensional constrained codeword sequences in the 2×2 codeword generation module from the buffer of this module in sequence, and along the Diagonal rows of the two-dimensional word cell array, writing 2×2 two-dimensional constrained code words into the two-dimensional word cell array row by row; The four modules of the above-mentioned encoder are connected in sequence, and finally the two-dimensional data array is output to the two-dimensional data recording device through the 2×2 code word therein written into the two-dimensional word cell array module; The decoder includes: A two-dimensional array buffer module, which is mainly used to temporarily store a two-dimensional data array with a size of 2N×2M read from a two-dimensional data recording device; Two-dimensional word unit page construction module, which is mainly used to divide the 2N×2M two-dimensional data array in the two-dimensional array buffer module into N×M two-dimensional word unit arrays with 2×2 as the minimum read unit, juxtaposing each word unit read flag of the two-dimensional word unit array as a readable flag; A two-dimensional word unit reading module, which is mainly used to read a suitable 2×2 two-dimensional constrained codeword from the two-dimensional word unit array, and temporarily store the generated 2×2 two-dimensional constrained codeword sequence; 2×2 codeword to one-dimensional data mapping module, this module is mainly used to read 2×2 two-dimensional constrained codeword sequence from two-dimensional word unit reading module, according to the encoding and decoding rule table, convert 2×2 two-dimensional The constrained codeword is decoded into a one-dimensional data word whose size is equal to 2 bits according to the sequence; A one-dimensional data stream assembly module, which is mainly used to assemble the 2-bit one-dimensional data words generated by the 2×2 codeword to one-dimensional data mapping module into a one-dimensional data stream according to the sequence of generation; The five modules of the above decoder are connected in sequence, and finally output through the one-dimensional data stream assembly module. 2. A two-dimensional run-length limited constraint encoding and decoding method for variable length code rate, characterized in that the encoding process is realized by an encoder, comprising the following steps successively: S101) The one-dimensional data stream buffering and grouping module divides the input binary one-dimensional data stream evenly into a plurality of one-dimensional 2-bit data words according to each 2-bit group, and inputs the 2-bit data words into a 2×2 code The data buffer of the word generation module is temporarily stored; S102) The 2×2 codeword generation module reads the one-dimensional 2-bit data word one by one from the data buffer, and selects the output scheme of the codec rule table to convert the one-dimensional 2-bit data word into a 2×2 two-dimensional constraint codeword; S103) The two-dimensional word unit array construction module divides a two-dimensional data array with a size of 2N×2M into 2×2 word units according to the storage size of the two-dimensional data recording device, and divides it into N×M size The two-dimensional word unit array of the same 2×2 two-dimensional word unit, juxtaposing each 2×2 two-dimensional word unit write flag as true, wherein N and M are positive integers; S104) 2 × 2 code words are written into the two-dimensional word unit array module: the 2 × 2 two-dimensional constrained code word obtained in step S102 is based on the 2 × 2 two-dimensional word unit obtained in step S103, along the two-dimensional word In the diagonal direction of the cell array, two-dimensional constrained codewords are sequentially written into two-dimensional word units with a size equal to 2×2, and at the same time, the right side of the currently written two-dimensional word unit is written according to the codec rule table. Selectively fill with the directly adjacent two-dimensional word units below; if all the two-dimensional word units in the diagonal row are written, enter the next step, and if all the two-dimensional constraint codewords have been filled, the encoding process will end directly ; S105) Judging whether there is a next diagonal line that has not been written, if so, then go to step S104; if all 2*2 two-dimensional word units of the two-dimensional word unit page have been filled with two-dimensional constraint codewords, then enter the next step step; S106) Judging whether there are unencoded one-dimensional data words, if so, go to step S102; otherwise, end the encoding process. 3. The encoding and decoding method according to claim 2, characterized in that the decoding process is realized by a decoder, comprising the following steps successively: S201) Temporarily store the two-dimensional data array read from the two-dimensional data recording device in the two-dimensional data buffer module; S202) The two-dimensional word unit page construction module reads a two-dimensional data array with a size of 2N×2M, divides the two-dimensional data array evenly with 2×2 as the minimum reading unit, and forms N×M identically sized A two-dimensional word unit array of 2×2 two-dimensional word units, juxtaposing each two-dimensional word unit of the two-dimensional word unit array to read the flag as a readable flag true, wherein N and M are positive integers; S203) The two-dimensional word unit reading module takes the 2×2 two-dimensional word unit of the two-dimensional word unit array as a reference, according to the encoding sequence of the encoder, along the diagonal row of the two-dimensional word unit array, from the upper right corner to the lower left corner Selectively read the two-dimensional constrained codewords in the two-dimensional word cell array according to the reading sign of each two-dimensional word unit in turn, if all the two-dimensional constrained codewords of this row are read, then enter the next step, if the current two-dimensional constrained codewords are all read The two-dimensional constrained codeword of the dimension word unit array has been read, then go to step S205; S204) judging whether there is a next diagonal row that has not yet been decoded, if so, go to step S203; if the two-dimensional constraint codewords of all diagonal rows of the two-dimensional word unit page have all been read, enter the next step; S205) The 2×2 codeword-to-one-dimensional data mapping module decodes the 2×2 two-dimensional constrained codewords read from the two-dimensional word unit reading module according to the generation order according to the decoding table corresponding to the selected codec table Code into a corresponding one-dimensional 2-bit data word, and enter the next step after completion; S206) The one-dimensional data stream assembling module assembles the obtained one-dimensional 2-bit data words according to the corresponding decoding sequence, obtains the corresponding one-dimensional binary data stream, and completes the decoding. 4. The encoding and decoding method according to claim 2, characterized in that the method for selectively filling the right and lower adjacent two-dimensional word units currently written in the 2×2 two-dimensional word unit in step S104 is as follows: a) If the two-dimensional constrained codeword written into the current 2×2 two-dimensional word unit does not belong to a diagonal matrix or a sub-diagonal matrix, then judge the right side, directly adjacent to the current 2×2 two-dimensional word unit, The writing flag of the two-dimensional word unit below is true or false; if it is true, the two-dimensional constrained codeword that is filled in the two-dimensional word unit position corresponding to the flag bit is a diagonal matrix or a sub-diagonal matrix; if it is false or If not defined, the two-dimensional constraint codeword will not be filled; After completing the above-mentioned writing process, set the current 2×2 two-dimensional word unit and the two-dimensional word unit writing flag filled with the two-dimensional constraint codeword as false, indicating that these positions cannot be allowed to write the two-dimensional constraint codeword again in the future; b) If the two-dimensional constraint code word written into the current 2×2 two-dimensional word unit belongs to a diagonal matrix or a sub-diagonal matrix, after writing the two-dimensional constraint code word at the position of the current two-dimensional word unit, write it In flag is set to false. 5. The encoding and decoding method according to claim 3, wherein the method for selectively reading the two-dimensional constraint codeword in the two-dimensional word unit page in step S203 is: a) If the reading flag of the currently read two-dimensional word unit is true, then read the two-dimensional constraint code word in the two-dimensional word unit, and set the read flag of the two-dimensional word unit as false, that is, in the future Not allowed to read; if the currently read two-dimensional constraint code word does not belong to the diagonal matrix or sub-diagonal matrix, set the two-dimensional word unit adjacent to the right and below the current read word unit position to read The flag is false, that is, reading is not allowed; b) If the reading flag of the currently read two-dimensional word unit is false, transfer to read the data in the next two-dimensional word unit in the diagonal row where the current two-dimensional word unit is located. 6. The encoding and decoding method according to any one of claims 2 to 5, characterized in that, the encoding and decoding rule tables are shown in Tables 1 and 2 below: Table 1 Encoding table of the first coding rule Table 2 Decoding table of the first encoding and decoding rules 7. The encoding and decoding method according to any one of claims 2 to 5, characterized in that, the encoding and decoding rule tables are shown in Tables 3 and 4 below: Table 3 Encoding table of the second coding rule Table 4 Decoding table of the second encoding and decoding rules