CN106849956A - Compression method, decompression method, device and data processing system - Google Patents

Abstract

The present application discloses a compression method, a decompression method, a device and a data processing system, and belongs to the field of data processing. The decompression method comprises: obtaining (a*N+b)/a sub-code segments according to a code segment with a code length of a*N+b read from compressed data in the xth clock cycle; concurrently decompressing (a*N+b)/a sub-code segments to obtain (a*N+b)/a characters; and determining the target decompressed data of the xth clock cycle from (a*N+b)/a characters according to a variable-length coding table and the code segment read in the xth clock cycle. The present application concurrently decompresses multiple sub-code segments to obtain multiple characters, and then determines the target decompressed data of the xth clock cycle from the multiple characters. There is no need to wait for a code word to be decompressed before decompressing the next code word according to the length of the code word, and the decompression speed is high.

Description

Compression method, decompression method, device and data processing system

technical field

The present application relates to the field of data processing, in particular to a compression method, decompression method, device and data processing system.

Background technique

Variable length coding (English: Variable Length Coding) is a coding method used when compressing data. In this coding method, the characters in the data to be compressed can be represented by codewords of different lengths (the codewords are several binary codes). Indicates that characters with a higher probability of occurrence are usually represented by a shorter codeword (such as 01), and characters with a lower probability of occurrence are represented by a longer codeword (such as 111001), and the character and codeword The corresponding relationship can be recorded in the variable-length coding table.

At present, the data obtained by using variable-length coding to compress data is called compressed data packaging. The compressed data packaging may include: compressed data and variable-length coding tables, etc. When decompressing the compressed data The data is encapsulated and analyzed to obtain a variable-length code table, and then, starting from the first bit of the compressed data, the compressed data is sequentially decoded according to the variable-length code table to obtain decompressed data.

Since the length of each codeword in the compressed data in the above-mentioned compressed data package is unknown before decompression, when the compressed data is decompressed, the starting position of the next codeword can only be determined after decompressing a codeword. Therefore, it is necessary to decompress one codeword after another according to the order of the compressed data, and the decompression speed is low.

Contents of the invention

In order to solve the problem of low decompression speed during decompression, embodiments of the present invention provide a compression method, decompression method, device and data processing system. Described technical scheme is as follows:

The decompression method provided by the embodiment of the present invention can be executed by a decompression engine, and the decompression engine can be set on a processor of x86 architecture (English: The X86 architecture) or an advanced reduced instruction set processor (English: Advanced RISC Machines; abbreviation: ARM ) architecture processors.

In a first aspect, an embodiment of the present invention provides a decompression method, the method comprising:

The decompression engine reads the code segment with a code length of a*N+b from the compressed data according to the xth clock cycle, and obtains (a*N+b)/a sub-code segments; where x is greater than or equal to 1 Integer; the code length of each subcode segment in (a*N+b)/a subcode segments is c, and (a*N+b)/a subcode segments include the first subcode segment, and the starting point of the first subcode segment The start bit coincides with the start bit of the code segment read in the xth clock cycle, and the code length between the start bits of two adjacent sub-code segments in (a*N+b)/a sub-code segments is a -1, the N is an integer greater than 0.

The decompression engine decompresses (a*N+b)/a sub-code segments concurrently based on the variable-length code table to obtain (a*N+b)/a characters, wherein the decompression engine decompresses each sub-code segment able to get a character.

The variable-length coding table may include a plurality of codewords, and each character in the (a*N+b)/a characters corresponds to a codeword in the variable-length coding table; at least two codewords in the multiple codewords correspond to The code lengths of different code words are different, the code length corresponding to the code word with the longest code length among multiple code words is c, the code length corresponding to the code word with the shortest code length among multiple code words is a, and a is multiple codes The greatest common divisor of the code length corresponding to each code word in the word, where a is an integer greater than or equal to 2, c is an integer greater than a, and the difference between c and a is b.

The decompression engine determines the target decompressed data of the xth clock cycle from (a*N+b)/a characters according to the variable-length code table and the code segment read by the xth clock cycle, wherein the target of the xth clock cycle In the decompressed data, the position of the last code element of the code word corresponding to one valid character among the two adjacent valid characters and the start of the code word corresponding to the other valid character in the code segment read in the xth clock cycle The positions of the symbols in the code segment read in the xth clock cycle are adjacent. The target decompressed data of the xth clock cycle includes a plurality of valid characters arranged in a specific order.

In the decompression method provided by the embodiment of the present invention, when decompressing the compressed data, the (a*N+b)/a subcode segments are decompressed concurrently to obtain (a*N+b)/a characters, and then determine the target decompressed data of the xth clock cycle from (a*N+b)/a characters according to the variable-length code table and the code segment read in the xth clock cycle. There is no need to wait for a codeword to be decompressed before decompressing the next codeword according to the length of the codeword. During decompression, decompression can be started from multiple locations at the same time, and the decompression speed is relatively high.

Optionally, the last b bits of the code segment with a code length of a*N+b read in the xth clock cycle are in phase with the first b bits of the code segment with a code length of a*N+b read in the x+1th clock cycle coincide.

In the decompression method provided by the embodiment of the present invention, the decompression engine avoids omission of Uncompressed code segment.

Optionally, when x is equal to 1, the decompression engine determines the xth clock cycle from (a*N+b)/a characters according to the variable-length code table and the code segment read by the xth clock cycle The target decompression data, including:

The decompression engine determines the target decompressed data of the xth clock cycle from (a*N+b)/a characters, and the starting code element and The starting symbols of the code segments read in the xth clock cycle are coincident.

In the decompression method provided by the embodiment of the present invention, when x is equal to 1, the target decompressed data of the first clock cycle is determined by the position of the start symbol of the target decompressed data.

Optionally, when x is an integer greater than or equal to 2, the decompression engine determines from (a*N+b)/a characters The target decompressed data of the xth clock cycle, including:

The decompression engine determines the c/a group of candidate decompression data from (a*N+b)/a characters; each group of candidate decompression data includes multiple characters, and every two adjacent characters, the code corresponding to one character The position in the code segment read by the last code element of the word is adjacent to the position in the code segment read by the code word corresponding to another character in the x clock cycle; The position of the start symbol of the code word corresponding to the first character in each group of candidate decompressed data is different in the code segment read in the xth clock cycle, and the first character in each group of candidate decompressed data corresponds to The starting symbol of the code word is the W*a-th symbol in the code segment read in the xth clock cycle, and W is an integer greater than or equal to 0 and less than or equal to b/a.

The decompression engine determines the target decompressed data of the xth clock cycle from the candidate decompressed data of group c/a; in the compressed data, the last symbol of the codeword corresponding to the last character in the target decompressed data of the x-1th clock cycle The start symbol of the codeword corresponding to the first character in the target decompressed data of the xth clock cycle is adjacent, and the target decompressed data of the x-1th clock cycle refers to the code read in the x-1th clock cycle segment corresponding to the target decompressed data.

In the decompression method provided by the embodiment of the present invention, when x is an integer greater than or equal to 2, by first obtaining possible c/a candidate decompression data of the data read in the x-th period, and then according to the x-th The target decompression data of 1 clock cycle selects the target decompression data of the xth clock cycle, so that the decompression engine can start to obtain the candidate decompression data before the decompression of the data of the previous clock cycle is completed, and the speed of decompression data is improved. When x is equal to 1, the target decompressed data of the x-1th clock cycle is a plurality of valid characters arranged in a specific order obtained by decompressing the code segment read in the first clock cycle.

Optionally, after the decompression engine determines the target decompressed data of the xth clock cycle, the method further includes:

The decompression engine splices the obtained decompressed data of multiple clock cycles according to the compressed data, and obtains the decompressed data corresponding to the compressed data. The position of the last symbol of the code word in the compressed data is adjacent to the position of the start symbol of the code word corresponding to the first character in the target decompressed data of another clock cycle in the compressed data.

In the decompression method provided by the embodiment of the present invention, after obtaining the decompressed data of multiple clock cycles, these results are spliced together to obtain the decompressed data of the compressed data, and the decompression speed is relatively high.

In a second aspect, an embodiment of the present invention provides a compression method, which can be executed by a compression engine, and the method includes:

The compression engine determines the target common divisor of the code length corresponding to each codeword in the variable-length coding table; the variable-length coding table includes multiple codewords, the target common divisor is an integer greater than or equal to 2, and each character in the data to be compressed Corresponding to a codeword in the variable-length coding table;

The compression engine generates a variable-length code table according to each character in the data to be compressed and the target common divisor; the code lengths of at least two codewords in the variable-length code table are different, and the code corresponding to the character with a higher probability of appearing in the data to be compressed The code length of word is less than the code length of the corresponding code word of character with lower probability of occurrence, and the greatest common divisor of the code length of code word in the variable-length coding table is target common divisor;

The compression engine compresses the data to be compressed according to the variable-length code table to obtain the compressed data;

The compression engine generates compressed data packages based on the compressed data and the variable-length code table.

The compression method provided by the embodiment of the present invention generates a variable-length code table according to each character in the data to be compressed and the target common divisor, and compresses the data to be compressed according to the variable-length code table, so that the decompression engine decompresses the compressed data , can start decompression concurrently from multiple locations, instead of waiting for a codeword to be decompressed and then decompress the next codeword according to the length of the codeword, and the decompression speed is relatively high.

Optionally, the decompression engine determines the target common divisor of the code length corresponding to each codeword in the variable-length coding table, including:

The decompression engine can first determine at least one common divisor, and compress the data to be compressed according to the variable-length coding table corresponding to each common divisor in the at least one common divisor. Each common divisor in the number can generate a corresponding variable-length code table;

Afterwards, the decompression engine may determine the target common divisor from the at least one common divisor, and the value of the target common divisor is greater than the value of each common divisor in the at least one common divisor except the target common divisor.

The greater the greatest common divisor a, the higher the decompression speed of the data compressed by the compression method provided by the embodiment of the present invention, but the compression rate may increase when the greatest common divisor becomes larger. In the compression method provided by the embodiment of the present invention, by pre-determining at least one common divisor whose corresponding compression rate is smaller than the preset value, and selecting the largest common divisor as the target common divisor, it is ensured that both the compression rate and the decompression speed are within a relatively low high range.

In the third aspect, the embodiment of the present invention provides a decompression device, the decompression device includes a plurality of units, and the plurality of units are used to implement the above-mentioned first aspect or any one of the possible implementations of the first aspect. Unzip method.

In a fourth aspect, an embodiment of the present invention provides a compression device, the compression device includes a plurality of units, and the plurality of units are used to implement the compression method provided in the second aspect or any possible implementation manner of the second aspect .

In a fifth aspect, an embodiment of the present invention provides a data processing system, the system includes the decompression device provided in the third aspect and the compression device provided in the fourth aspect, the compression device is used to compress data to be compressed to obtain compressed data; The decompression device is used for decompressing the compressed data.

In a sixth aspect, a decompression device is provided, the decompression device includes: a base plate and a central processing unit (English: Central Processing Unit; CPU for short) and memory arranged on the base plate, the base plate is connected with a bus interface, the The bus interface may be connected with a decompression engine, and the decompression engine is used to execute the decompression method provided in the first aspect.

In a seventh aspect, a decompression device is provided, the decompression device includes: a processor chip, a memory connected to the processor chip, and a CPU and a bus interface arranged on the processor chip, and the decompression engine can be integrated in the processor On the chip and connected to the CPU through the bus interface, the decompression engine is used to execute the decompression method provided by the first aspect.

In an eighth aspect, a compression device is provided, the compression device includes: at least one processor, at least one network interface, memory, and at least one bus, the memory and the network interface are respectively connected to the processor through the bus; the processor is configured to execute Instructions stored in the memory; the processor implements the compression method provided by the second aspect by executing the instructions.

In summary, the decompression method provided by the embodiment of the present invention, when decompressing compressed data, decompresses (a*N+b)/a subcode segments concurrently to obtain (a*N+ b)/a characters, and then determine the target decompressed data of the xth clock cycle from (a*N+b)/a characters according to the variable-length code table and the code segment read in the xth clock cycle. There is no need to wait for a codeword to be decompressed before decompressing the next codeword according to the length of the codeword. The problem of low decompression speed in the related technology is solved. When decompressing, it can be decompressed from multiple locations at the same time, and the decompression speed is relatively high.

Description of drawings

FIG. 1A is a structural block diagram of a decompression device provided by an embodiment of the present invention;

FIG. 1B is a structural block diagram of another decompression device provided by an embodiment of the present invention;

Fig. 1C is a structural block diagram of a compression device provided by an embodiment of the present invention;

FIG. 2A is a flowchart of a compression method shown in an embodiment of the present invention;

Fig. 2B is a flow chart of determining the common divisor of the target in the embodiment shown in Fig. 2A;

FIG. 3A is a flowchart of a decompression method shown in an embodiment of the present invention;

Fig. 3B is a schematic diagram of the corresponding relationship between codewords and characters in the data read in the xth clock cycle in the embodiment shown in Fig. 3A;

Fig. 3C is a bar graph of chip area occupied by the decompression engine in the embodiment shown in Fig. 3A;

Fig. 4 is a structural block diagram of a compression device provided by an embodiment of the present invention;

FIG. 5A is a structural block diagram of a decompression device provided by an embodiment of the present invention;

FIG. 5B is a structural block diagram of a decompression device provided by an embodiment of the present invention;

FIG. 6 is a schematic structural diagram of a data processing system provided by an embodiment of the present invention.

detailed description

In order to make the purpose, technical solutions and advantages of the present disclosure clearer, the present disclosure will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some of the embodiments of the present disclosure, not all of them. .

Please refer to FIG. 1A, which shows a structural block diagram of a decompression device provided by an embodiment of the present invention. The decompression device may include: a base plate 11, a CPU 12 and a memory 13 arranged on the base plate 11, and the base plate 11 is connected with Bus interface 14, this bus interface 14 can be connected with decompression engine 15, and decompression engine 15 can be expansion card, and expansion card can be Field-Programmable Gate Array (English: Field-Programmable Gate Array; Abbreviation: FPGA) or application-specific integration Circuit (English: Application Specific Integrated Circuit; Abbreviation: ASIC). The bus interface 14 may be PCIe (a bus interface).

Please refer to FIG. 1B, which shows a structural block diagram of another decompression device provided by an embodiment of the present invention. The decompression device may include: a processor chip 16, a memory 17 connected to the processor chip 16, and a The CPU 161 and the bus interface 162 on the processor chip 16 and the decompression engine 163 may be integrated on the processor chip 16 and connected to the CPU 161 through the bus interface 162 . Wherein, the memory 17 may be a double rate synchronous dynamic random access memory (English: Double Data Rate SDRAM; abbreviation: DDR), and the bus interface 162 may be an advanced extensible interface (English: AdvancedeXtensible Interface; abbreviation: AXI).

Please refer to FIG. 1C, which shows a structural block diagram of a compression device provided by an embodiment of the present invention. The compression device 20 may include: at least one processor 21, at least one network interface 22, memory 23, and at least one bus 24, The memory 23 and the network interface 22 are respectively connected to the processor 21 through the bus 24 ; the processor 21 is configured to execute instructions 231 stored in the memory 23 . The memory 23 may include a high-speed random access memory (English: Random Access Memory; RAM for short), and may also include a non-volatile memory (English: non-volatile memory), such as at least one disk memory.

FIG. 2A is a flow chart of a compression method shown in an embodiment of the present invention. This embodiment is described by taking the compression method applied to compress data to be compressed as an example. The compression method may include the following steps:

Step 201, the compression engine obtains the probability distribution of various characters in the data to be compressed.

When using the compression method provided by the embodiment of the present invention, the compression engine can analyze the data to be compressed to obtain the probability distribution of various characters in the data to be compressed. The data to be compressed may be composed of a plurality of different characters, and the occurrence probability of different characters may be different, and the probability distribution may include the occurrence probability of each character. Exemplarily, the occurrence probability of the character "A" may be 4% ( percent sign), the occurrence probability of the character "B" may be 6%. Wherein, the characters in the data to be compressed refer to the uncompressed characters stored in the storage device. When the storage device stores data, it usually uses binary bits (English: bit) as the unit. According to different encoding methods, each character The number of binary bits occupied is different. For example, in data to be compressed that uses American Standard Code for Information Interchange (English: American Standard Code for Information Interchange; ASCII for short) as the encoding method, each character can occupy 8 binary bits.

The compression engine in the embodiment of the present invention may be integrated in the compression device shown in FIG. 1C through software or hardware.

Step 202, the compression engine determines the target common divisor of the code length corresponding to each codeword in the variable-length coding table.

The variable-length coding table may include multiple codewords, the target common divisor is an integer greater than or equal to 2, and each character in the data to be compressed corresponds to a codeword in the variable-length coding table.

The target common divisor is the greatest common divisor of the code length corresponding to each codeword in the coding table, and the coding table is subsequently generated by the compression engine. It should be noted that a codeword is composed of several code elements, which are usually expressed as several bits of binary data in computer communication. A symbol can be a binary number.

As shown in Figure 2B, the process of the compression engine determining the target common divisor may include the following two sub-steps:

Sub-step 2021, the compression engine determines at least one common divisor, and compresses the data to be compressed according to the variable-length coding table corresponding to each common divisor in the at least one common divisor, and the obtained compression ratios are all lower than the preset value.

Wherein, according to each common divisor in at least one common divisor, a corresponding variable-length coding table can be generated.

Substep 2021 may include:

1) The compression engine can first set multiple common divisors in advance, and these multiple common divisors are all integers greater than or equal to 2 and less than S, where S is the number of binary bits occupied by each character in the data to be compressed (this is Because if the common divisor is greater than or equal to the number of binary bits occupied by characters, when the common divisor is the greatest common divisor of the code length of each codeword in the compressed data, the data volume of the compressed data will be greater than the data to be compressed, which is difficult for the purpose of compression). Exemplarily, the number of bits occupied by characters in the data to be compressed is 8, and the common divisor can be preset as 6 numbers from 2 to 7.

2) After setting multiple common divisors, the compression engine can estimate the compression ratio (English: Compression ratio) corresponding to each common divisor among the multiple common divisors. The compression ratio is the difference between the size of the data after compression and the size before compression. Compare. Exemplarily, if the size of the data to be compressed is 100 megabytes (English: Megabytes; M for short) before compression, and the size after compression is 10M, the compression rate is 10%.

When the compression engine estimates the compression rate of each common divisor, it can sample the data to be compressed to obtain sampled data, and then generate a variable-length code table according to each common divisor (the process of generating a code table according to the common divisor can refer to step 203), And compress the sampling data according to each variable-length coding table to obtain the compression rate corresponding to each common divisor. Among them, in the coding table generated according to any of the multiple common divisors, the corresponding relationship between each character in the sampling data and the code word corresponding to the character is recorded, and at the same time, each character in the variable-length coding table The code length of the corresponding code word is a positive integer multiple of any common divisor.

3) Selecting at least one common divisor whose corresponding compression ratio is smaller than a preset value from multiple common divisors.

The preset value can be preset for the operator, and the preset value can consider the requirement for decompression speed when setting it. When the requirement for decompression speed is high, a higher preset value can be set, and when the requirement for decompression speed is low , you can set a smaller preset value. This is because generally the smaller the compression ratio, the slower the decompression speed will be.

In sub-step 2022, the compression engine determines a target common divisor from at least one common divisor, and the value of the target common divisor is greater than the value of each common divisor in the at least one common divisor except the target common divisor.

The compression engine determines a target common divisor, where the target common divisor refers to a common divisor having a maximum value among at least one common divisor whose corresponding compression rate is smaller than a preset value.

It should be understood that, in the aforementioned "at least one common divisor", when the number of the "at least one common divisor" is 1, that is, the "at least one common divisor" refers to a common divisor, then the common divisor is the objective common divisor. Only when the number of "at least one common divisor" is greater than or equal to 2, the target common divisor refers to the common divisor that is greater than the value of each common divisor in the at least one common divisor except the target common divisor.

In the decompression method corresponding to the compression method provided by the embodiment of the present invention (refer to the embodiment shown in Figure 3A), the greater the greatest common divisor of the codeword in the variable-length coding table, the greater the data read by the decompression engine per clock cycle The more there are, the higher the decompression speed is. Therefore, in consideration of the decompression speed, the largest common divisor among the common divisors whose compression rate is smaller than the preset value can be determined as the target common divisor.

In addition, since the greater the greatest common divisor, the longer the code length of the code word representing the character will be, and the larger the data volume of the compressed data (compressed data is composed of each code word) used to represent the data to be compressed will be. Larger, so the greater the greatest common divisor, the greater the compression rate. In the case of considering the compression ratio, the smallest common divisor among the common divisors whose compression ratio is smaller than the preset value may also be determined as the target common divisor.

Step 203, the compression engine generates a variable-length code table according to each character in the data to be compressed and the target common divisor.

The code lengths of at least two code words in the variable-length coding table are different, and the code length of the code word corresponding to the character with a higher probability of occurrence in the data to be compressed is smaller than the code length of the code word corresponding to the character with a lower probability of occurrence. The greatest common divisor of the code lengths of the codewords in the table is the target common divisor.

The variable-length coding table is a mapping table that records the correspondence between characters and codewords in the data to be compressed, and the process of generating the variable-length coding table is the process of determining the correspondence between characters and codewords in the data to be compressed. Wherein, each codeword in the variable-length coding table corresponds to a character in the data to be compressed, the codewords corresponding to the same character are the same, and the codewords corresponding to different characters are different. For "different characters correspond to different codewords", one situation is that the number of digits (or "code length") of codewords corresponding to different characters is different; another situation is that although the codewords corresponding to different characters have The number of bits is the same, but the arrangement order of the binary data in the codeword is different. Referring to Table 1, the codeword corresponding to character A is 000, and the codeword corresponding to character B is 001. Although the codeword 000 corresponding to character A and the codeword 001 corresponding to character B have the same number of digits, the binary data in the codeword are sorted in a different order. Further referring to Table 1, the code corresponding to character A is 000, and the code corresponding to character F is 101000, so the code 000 corresponding to character A and the code 101000 corresponding to character F have different digits.

When generating the variable-length code table, you can assign codewords with lengths from small to large for each character in the descending order of the occurrence probability of each character in the data to be compressed, that is, the higher the probability of a character, the more A character corresponds to a codeword with a shorter length, so that the data to be compressed can be expressed with a smaller amount of data.

Exemplarily, when the target common divisor is 3, the variable-length coding table may be as shown in Table 1:

Table 1

In Table 1, in the character column, the occurrence probability of a character corresponding to a code word with a shorter code length is greater than that of a character corresponding to a longer code word. The code length on the same line as the character represents the code length of the code word corresponding to the character, or the number of code elements contained in the code word corresponding to the character, for example, the code length 6 on the same line as the character "P" represents It is the code length of the codeword "110010" corresponding to the character "P". The codeword located on the same row as the character represents the codeword corresponding to the character, for example, the codeword “110011” located on the same row as the character “Q” represents the codeword corresponding to the character “Q”. The arrangement order of the characters in the data to be compressed is consistent with the arrangement order of the codewords in the compressed data.

The variable length encoding in the embodiment of the present invention may be an entropy encoding (English: entropy encoding), and the entropy encoding is an encoding method used in lossless compression.

Step 204, the compression engine compresses the data to be compressed according to the variable-length coding table to obtain compressed data.

After the variable-length coding table is obtained, the data to be compressed can be compressed according to the variable-length coding table to obtain compressed data. Exemplarily, the variable-length coding table is Table 1, and after compressing the data "ABC" to be compressed according to Table 1, the obtained compressed data is "000 001 010".

Step 205, the compression engine generates a compressed data package based on the compressed data and the variable-length coding table.

After obtaining the variable-length encoding table and the compressed data, the compression engine generates compressed data encapsulation based on the compressed data and the variable-length encoding table. The compressed data encapsulation may also include other data, such as checksum (English: CHECKSUM) and the like.

The compression rate of the compression method provided by the embodiment of the present invention is about 30% lower than that of the compression method based on LZ4 (a coding method).

To sum up, the compression method provided by the embodiment of the present invention generates a variable-length coding table with the largest common divisor in the length of the codeword, and compresses the data to be compressed according to the variable-length coding table, so that the decompression engine decompresses the compressed data. Data can be decompressed concurrently from multiple locations, instead of waiting for a codeword to be decompressed before decompressing the next codeword according to the length of the codeword, and the decompression speed is relatively high.

FIG. 3A is a flow chart of a decompression method shown in an embodiment of the present invention. In this embodiment, the decompression method is applied to decompress the compressed data package generated by the compression method provided in the embodiment shown in FIG. 2A as an example. The decompression method can be implemented by the decompression engine in FIG. 1A or FIG. 1B . The decompression method may include the following steps:

Step 301, the decompression engine analyzes the compressed data package, and obtains the variable-length code table and the compressed data.

When using the decompression method provided by the embodiment of the present invention, the decompression engine can first analyze the compressed data package to obtain the variable-length code table and compressed data. Wherein, the compressed data package may be a compressed data package compressed by the compression method provided in the embodiment shown in FIG. 2A , and the compressed data package may include a variable-length code table, compressed data and other data.

This step is an optional step, that is, the decompression engine can also directly obtain variable-length coded and compressed data.

Step 302, the decompression engine reads the code segment with a code length of a*N+b from the compressed data in each clock cycle according to the generation sequence of the compressed data, and the code segment read at the xth clock cycle is a*N+b The last b bits of the code segment coincide with the first b bits of the code segment whose code length is a*N+b read in the x+1th clock cycle.

After obtaining the variable-length encoding table, the decompression engine can read the code length from the compressed data in each clock cycle according to the generation sequence of the compressed data (that is, the first-generated data is read first, and the second-generated data is read later). The code segment of a*N+b (a*N+b is greater than or equal to the code length c of the longest code word in the compressed data), the code length read in the xth clock cycle is the code segment of a*N+b The last b bits coincide with the first b bits of the code segment whose code length is a*N+b read in the x+1th clock cycle, where N is an integer greater than 0, and a represents the target common divisor mentioned in the above embodiment , the value of N can be preset according to the hardware situation, b is the difference between the code length c of the longest code word in the compressed data and the code length a of the shortest code word (the code length of the shortest code word is equal to the target common divisor a), a *N may represent the decompression speed of the decompression method provided by the embodiment of the present invention, that is, the larger a and N are, the higher the decompression speed of the decompression method provided by the embodiment of the present invention. Exemplarily, when a=3, c=12, b=c-a=9, the decompression engine reads 3N+9 bits of data every clock cycle.

In the compressed data, values such as the greatest common divisor a of the code length of the code word, the difference b between the code length of the longest code word and the code length of the shortest code word in the compressed data, etc., can be known in advance by the decompression engine before decompression , Exemplarily, these values may be written into the decompression engine when the decompression engine is set. Or, values such as the greatest common divisor a of the code length of the code word, the difference b between the code length of the longest code word and the code length of the shortest code word in the compressed data can also be included in the variable-length coding table, and the decompression engine can Get these values when parsing a variable-length coded table.

A clock cycle (English: Clock Cycle) is also called an oscillation cycle, and a clock cycle is a unit of time in a computer. In one clock cycle, the CPU completes a basic action.

In step 302, the decompression engine reads the code segment of fixed code length from each clock cycle in the compressed data, so the data can be transmitted to the decompression engine with a stable bandwidth, and the decompression engine has a higher utilization rate of the bandwidth. The present invention The decompression method provided by the embodiment can be applied to bandwidth compression scenarios, such as the online decompression application of neural network parameters. In the related art, since the code length of the code word decompressed in each clock cycle may be different, the code segment of the code segment read by the decompression engine The length is also different, and it is difficult to apply to bandwidth compression scenarios.

In the decompression method provided by the embodiment of the present invention, the decompression engine can adjust the amount of data read in each clock cycle by adjusting the value of N, thereby adjusting the decompression speed.

Since there may be undecompressed multi-bit symbols at the end of the code segment read in the previous clock cycle of two adjacent clock cycles, in order to avoid the omission of these symbols, when the decompression engine reads the code segment in each clock cycle, The last b bits of the code segment read in the previous clock cycle are repeatedly read, because the code length of the entire compressed data is an integer multiple of a, and the number of symbols that may not be decompressed in the previous clock cycle is b( In the code segment read in the previous clock cycle, when the last code word has the maximum length c, the number of undecompressed symbols missed in the previous clock cycle is the maximum number b), and the least possible uncompressed code in the previous clock cycle The number of elements is 0. It should be noted that, in fact, the number of uncompressed symbols in the previous clock cycle is the product of an integer in the interval [0, K] and a, where K is equal to b/a.

Step 303, the decompression engine reads the code segment with a code length of a*N+b from the compressed data according to the xth clock cycle, and obtains (a*N+b)/a sub-code segments.

Wherein, x is an integer greater than or equal to 1; the code length of each subcode segment in (a*N+b)/a subcode segments is c, and includes the first subcode segment in (a*N+b)/a subcode segments Code segment, the start bit of the first sub-code segment coincides with the start bit of the code segment read in the xth clock cycle, and the start of two adjacent sub-code segments in (a*N+b)/a sub-code segments The code length of the interval between bits is a-1.

Step 304, the decompression engine decompresses (a*N+b)/a subcode segments concurrently based on the variable-length code table to obtain (a*N+b)/a characters, wherein each subcode is decompressed segment gets one character. When x is equal to 1, execute step 305; when x is an integer greater than or equal to 2, execute step 306.

In the a*N+b-bit data read in the x-th clock cycle, the decompression engine can decompress multiple sub-code segments concurrently, starting from the code word corresponding to a character obtained by decompressing each sub-code segment The initial symbol is the initial symbol of each subcode segment.

Among them, the decompression engine is based on the variable-length coding table for the multiple sub-code segments whose start bit is the N*ath bit to the (N+K-1)*a bit in the code segment whose code length is a*N+b When decompressing, if the number of remaining symbols in the code segment with a code length of a*N+b is less than c, the decompression engine can use a preset value (the preset The value can be set arbitrarily, such as all 0 or all 1, or 0 and 1 alternately, etc.) fill the code length to c, and then decompress these subcode segments to obtain multiple characters, and then each subcode can be detected The code length of the corresponding code word in the variable-length code table of the character obtained by segment decompression, such as the code length of the code word corresponding to the character obtained by decompressing a certain sub-code segment in the variable-length code table is greater than the code length of the sub-code segment before filling If the code length is long, the characters obtained by decompressing the sub-code segment can be deleted. After the characters are deleted, there are some code elements that have not been decompressed in the a*N+b bit data.

Exemplarily, the code segment read in the xth clock cycle is "000 001 010 011 100", a=3, c=6, according to the variable-length coding table shown in Table 1, and the sub-code segment "000 001" , "001 010", "010 011", "011 100" and "100000" are decompressed, and the obtained characters are the character "A" corresponding to the code word 000, the character "B" corresponding to the code word 001, and the code word 010 The corresponding character "C", the character "D" corresponding to the code word 011, and the character "E" corresponding to the code word "100". Among them, "100 000" is a sub-code segment filled with 6 bits, and the code length of the code word corresponding to the character "E" obtained by decompressing the 6-bit sub-code segment is 3, which is not greater than the length of "100 000" before filling, Thus, the character "E" may not be deleted.

It should be noted that there is a one-to-one correspondence between the codewords in the compressed data and the characters in the target decompressed data obtained by decompressing the compressed data. The Kth codeword in the compressed data is corresponding to the Kth character in the target decompressed data, and the start bit of a certain codeword is determined, and the codeword can be determined (read at least a bit code elements, up to c bit code elements, can obtain a codeword represented by these code elements from the variable-length coding table). In the prior art, the code length of each codeword in the compressed data is not the same. Before a codeword is decompressed out of the order in the compressed data, the decompression engine cannot know the code length of the codeword, and thus cannot know the next codeword. The start bit of the codeword. But because the code length of the code word in the compressed data in the decompression method that the embodiment of the present invention provides has the greatest common divisor a, thereby the start bit of the code word corresponding to each character in the decompressed data is included in (a*N+ In the start bit of each subcode segment in b)/a subcode segments, it can be said that the code word corresponding to each character in the target decompressed data obtained by decompressing the compressed data (from the variable-length coding table, it can be known that each The start bit of the codeword corresponding to the character) is the correct start bit in the start bits of (a*N+b)/a subcode segments (the correct start bit is based on the order of the bit in the entire compressed data Determined, if the start bit of the codeword corresponding to the first character in the decompressed data is the first bit of the entire compressed data, then the first bit in the entire compressed data is the codeword corresponding to the first character in the decompressed data The correct start bit), and the other bits except the correct start bit among the start bits of (a*N+b)/a subcode segments are called wrong start bits. In step 304, multiple characters obtained by decompressing (a*N+b)/a subcode segments concurrently include valid characters decompressed from the correct start bit and invalid characters decompressed from the wrong start bit .

Step 305, when x is equal to 1, the decompression engine determines the target decompressed data of the xth clock cycle from the (a*N+b)/a characters, and the first valid character in the target decompressed data of the xth clock cycle corresponds to The starting symbol of the code word coincides with the starting symbol of the code segment read in the xth clock cycle.

Among them, in the target decompressed data of the xth clock cycle, the position of the last code element of the codeword corresponding to one valid character in every two adjacent valid characters is the same as the other effective The starting symbols of the codewords corresponding to the characters are adjacent to each other in the code segment read in the xth clock cycle. The target decompressed data of the xth clock cycle includes a plurality of valid characters arranged in a specific order.

After the decompression engine obtains (a*N+b)/a characters, it knows the code length of the code word corresponding to each character in the (a*N+b)/a characters. Because these multiple characters include invalid characters decompressed from the wrong start bit, multiple valid characters can be selected from them for splicing to obtain the target decompressed data of the xth clock cycle, and the basis for selection can be to first determine The correct start bit of the first code word in the code segment whose code length is a*N+b read in the xth clock cycle, and then determine the second code word according to the correct start bit and the code length of the first code word The correct start bit of the first codeword (the correct start bit of the second codeword is the next bit after the last bit of the first codeword), and so on, the decompression engine can obtain the code corresponding to each valid character The correct start bit of the word, and then just be able to select a plurality of valid characters, these multiple valid characters are in order of the a*N+b bit data read in the code word corresponding to each valid character in the xth clock cycle By splicing in sequence, the target decompressed data of the data read in the xth clock cycle is obtained.

When x=1, the start bit of the code segment whose code length is a*N+b read in the xth clock cycle is the start bit of the compressed data, and the start bit is the code read in the xth clock cycle The correct start bit of the first codeword in the code segment of length a*N+b.

Exemplarily, as shown in Figure 3B, it reads the first 24 code elements in the code segment whose code length is a*N+b for the xth clock cycle, and in step 304, the value of each codeword can be obtained The code length and the characters corresponding to each code word, that is, a code word with a code length of 6 composed of 0 to 5 bits corresponds to character A, a code word with a code length of 6 composed of 3 to 8 bits corresponds to character B, and 6 to 14 bits The formed code length is the corresponding character C of the code word of 9, the code word corresponding character D of the code length of 6 is the code word formed by the code length of 9 to 14 bits, and the corresponding character E of the code word of the code length of 12 formed by the 12 to 23 bits, 15 to 17 The code length that the bit constitutes is that the code word corresponding character F of 3, wherein, the code length of code word and the character corresponding to code word can be as shown in Table 2:

Table 2

In Table 2, the order of each character from top to bottom, and the start bit of the code word corresponding to each character in the code segment with code length a*N+b read in the xth clock cycle is consistent. The mode of determining valid characters from the 24-bit code element shown in Figure 3B and the corresponding characters of the 24-bit code element can be as follows: the start bit of the 24-bit code element is used as the start bit of the first code word (the start bit is the correct start bit of the codeword corresponding to the first valid character), the code length of the first codeword is determined to be 6, and the character corresponding to the first codeword is A, because the first The code length of a codeword is 6, thus the correct start bit of the second codeword is the 6th bit of the 24-bit symbol (the first 0 to 5 bits are occupied by the first codeword), and the 24-bit The 6th bit of the code element is the character corresponding to the code word of the starting position is C, so character C can be used as the second effective character, because the code length of the code word corresponding to the second effective character is 9, so the third The correct start bit of a code word is the 15th bit of the 24-bit code element (the first 0 to 14 bits are occupied by the first code word and the second code word), and the code word with the 15th bit as the starting position The corresponding character is F, so the obtained decompressed data is A C F.

Step 306, when x is an integer greater than or equal to 2, the decompression engine determines c/a group of candidate decompression data from (a*N+b)/a characters.

Wherein, each group of candidate decompressed data includes a plurality of characters, and in every two adjacent characters, the position of the last code element of the code word corresponding to a character in the code segment read in the xth clock cycle and the position of another character The position of the start symbol of the corresponding codeword in the code segment read in the xth clock cycle is adjacent; the start symbol of the codeword corresponding to the first character in each two groups of candidate decompressed data is at the xth clock cycle The position in the code segment read by the clock cycle is different, and the starting code element of the code word corresponding to the first character in each group of candidate decompressed data is the W*th in the code segment read by the xth clock cycle a symbols, W is an integer greater than or equal to 0 and less than or equal to b/a.

When x is an integer greater than or equal to 2, since the data read in the x-1th clock cycle has not been decompressed, it is difficult for the decompression engine to know the number of undecompressed symbols in the data read in the x-1th clock cycle Therefore, it is currently difficult for the decompression engine to determine the correct start bit of the first codeword in the read code segment with a code length of a*N+b. At this point, the decompression engine can determine all possible correct start bits (all possible correct start bits are the H*a bit in the code segment whose code length is a*N+b read in the xth clock cycle, H gets each integer from 0 to K, K=b/a), then respectively use these all possible correct start bits as the correct start bits of the first code word, and obtain c/ in the manner in step 305 a candidate decompressed data.

Wherein, since the first b bits of the code segment read in the x-th clock cycle are repeated with the last b bits of the code segment read in the x-1 clock cycle, and the first few bits in the repeated b bits may have the first A part of the last code word in the code segment read by x-1 clock cycle, the correct start bit of the first code word in the code segment read by x-1 clock cycle is the code segment read by x-1 clock cycle The next digit of the last digit of the last codeword in .

Since the greatest common divisor of the code length of the code word in the compressed data is a, the length of the code segment read in the x-1th clock cycle that exists in the first b bits of the code segment read in the xth clock cycle can be from 0 to The product of any integer in K and a (when the length is 0*a, it indicates that the last bit of the code segment read in the x-1 clock cycle does not exist in the first b bits of the code segment read in the xth clock cycle Part of the code word), so the H*a bit in the first b-bit symbols of the code segment read in the xth clock cycle may be the correct start of the first code word in the code segment read in the xth clock cycle bit, and the number of candidate decompressed data depends on the number of integers from 0 to K, and this number is equal to K+1=b/a+1=(a+b)/a=c/a, wherein, for K 1 is added because 0 also counts as an integer from 0 to K.

Exemplarily, c=9, a=3, the first 9 digits of the code segment read in the xth clock cycle are "000 001 010", then the 0th, 3rd and 6th digits are possible correct Start bit, the decompression engine can respectively use the 0th bit, the 3rd bit and the 6th bit as the start bit of the codeword corresponding to the first character in the 3 candidate decompressed data, and obtain 3 in the manner in step 305 respectively candidate decompressed data.

Step 307, the decompression engine determines the target decompressed data of the xth clock cycle from the c/a group of candidate decompressed data.

In the compressed data, the last code element of the codeword corresponding to the last character in the target decompressed data of the x-1th clock cycle and the start code of the codeword corresponding to the first character in the target decompressed data of the xth clock cycle elements are adjacent, and the target decompressed data of the x-1th clock cycle refers to the target decompressed data corresponding to the code segment read in the x-1th clock cycle. When x is equal to 1, the target decompressed data of the x-1th clock cycle is a plurality of valid characters arranged in a specific order obtained by decompressing the code segment read in the first clock cycle.

After the decompression engine obtains the c/a candidate decompressed data of the xth clock cycle, it also completes the decompression of the code segment read in the x-1th clock cycle, and obtains the target decompressed data of the x-1th clock cycle , the decompression engine can determine the correct start bit of the first codeword in the data read in the xth clock cycle according to the target decompressed data of the x-1th clock cycle, and select the xth from the c/a candidate decompressed data Clock cycle target to decompress data.

Exemplarily, the decompressed data of the x-1 clock cycle is "ABCD", the corresponding code word is "000 001 010011", and the code segment read in the x-1 clock cycle is "000 001 010 011 101", b =6, a=3, c=9, N=4, a*N+b=15, the code segment read in the xth clock cycle is "011 101000 101001", where the first 6 bits are the x-1th clock cycle The last 6 bits of the read code segment, the last three "101" in the code segment read in the x-1th clock cycle are the first three bits of a 6-bit code word, which will not be corrected in the manner in step 304 Three bits are decompressed, and these three bits of data are code segments not decompressed, and the starting code element of these three bits of code segments can be determined as the starting code word corresponding to the first character in the target decompressed data of the xth clock cycle start code unit.

It should be noted that, in the embodiment of the present invention, the code segment read in each clock cycle is processed concurrently, that is, after the decompression engine reads the code segment in the x-1th clock cycle, the decompression engine processes the When the code segment read in the x-1 clock cycle is decompressed, the decompression engine reads the code segment in the xth clock cycle and starts to decompress the code segment read in the xth clock cycle at the same time. The code segment read in the x-1th clock cycle is read one clock cycle earlier than the code segment read in the xth clock cycle, so the decompression engine performs any step from step 303 to step 307 on the code segment read in the xth clock cycle Each process is one step slower than the decompression engine performing the same process on the code segment read by x-1 clock cycles. Therefore, when the decompression engine is processing the code segment read in the x-th clock cycle, when step 306 is executed, the decompression engine has completed the decompression of the code segment read in the x-1 clock cycle, and the x-th clock cycle is obtained. -1 clock cycle target to decompress data.

Through steps 302 to 307, all the data in the compressed data can be decompressed, and the target decompressed data of each clock cycle can be obtained.

Step 308, the decompression engine splices the obtained decompressed data of multiple clock cycles according to the compressed data, and obtains decompressed data corresponding to the compressed data.

Wherein, in two adjacent clock cycles after splicing, the position of the last code element of the code word corresponding to the last character in the target decompressed data of one clock cycle in the compressed data and the first position in the target decompressed data of another clock cycle The positions of the starting code elements of the code word corresponding to a character in the compressed data are adjacent.

As shown in Figure 3C, it is the bar of the chip area occupied by the decompression engine when applying the decompression method provided by the embodiment of the present invention, when applying the decompression method based on Huffman coding, and when applying the decompression method based on LZ4. In the graph, the vertical axis represents the area of the chip, the unit of the area of the chip is square millimeter, and the chip refers to a chip of 16 nanometer (nm) process. Wherein, the bar 31 represents the chip area occupied by the decompression engine when the decompression method based on Huffman coding is applied, and the area is 0.05 square millimeters. The bar 32 represents that the greatest common divisor is 3. When the decompression method provided by the embodiment of the present invention is applied, the chip area occupied by the decompression engine is 0.035 square millimeters. Bar 33 represents that the greatest common divisor is 2. When the decompression method provided by the embodiment of the present invention is applied, the chip area occupied by the decompression engine is 0.025 square millimeters. Bar 34 represents the chip area occupied by the decompression engine when the LZ4-based decompression method is applied, and the area is 0.16 square millimeters. It can be seen that the decompression method provided by the embodiment of the present invention occupies less chip area than the decompression method in the related art.

In addition, the decompression speed of the decompression method provided by the embodiment of the present invention can reach about 10 times the decompression speed of the compressed data package generated based on Huffman coding, and is basically the same as the decompression speed of the compressed data package generated based on LZ4.

In summary, the decompression method provided by the embodiment of the present invention, when decompressing compressed data, decompresses (a*N+b)/a subcode segments concurrently to obtain (a*N+ b)/a characters, and then determine the target decompressed data of the xth clock cycle from (a*N+b)/a characters according to the variable-length code table and the code segment read in the xth clock cycle. There is no need to wait for a codeword to be decompressed before decompressing the next codeword according to the length of the codeword. The problem of low decompression speed in the related technology is solved. During decompression, decompression can be started from multiple locations at the same time, and the decompression speed is relatively high.

Fig. 4 is a structural block diagram of a compression device provided by an embodiment of the present invention, and the compression device may be part or all of a terminal or a server. The compression device 400 may include:

The common divisor determination unit 410 is configured to execute step 202 in the above embodiment.

The coding table generating unit 420 is configured to execute step 203 and step 201 in the above-mentioned embodiment.

The compression unit 430 is configured to execute step 204 in the above embodiment.

The package generation unit 440 is configured to execute step 205 in the above embodiment.

To sum up, the compression device provided by the embodiment of the present invention generates a variable-length coding table whose codeword length has the greatest common divisor, and compresses the data to be compressed according to the variable-length coding table, so that the decompression engine decompresses the compressed data. Data can be decompressed concurrently from multiple locations, instead of waiting for a codeword to be decompressed before decompressing the next codeword according to the length of the codeword, and the decompression speed is relatively high.

Fig. 5A is a structural block diagram of a decompression device provided by an embodiment of the present invention. The decompression device 500 may include:

The subcode segment determining unit 510 is configured to execute steps 301, 302 and 303 in the above-mentioned embodiment.

The decompression unit 520 is configured to execute step 304 in the above embodiment.

The decompressed data determining unit 530 is configured to execute steps 305, 306, and 307 in the above-mentioned embodiments.

Optionally, as shown in FIG. 5B, the decompression device 500 may also include:

The splicing unit 540 is configured to execute step 308 in the foregoing embodiment.

In summary, the decompression device provided by the embodiment of the present invention, when decompressing compressed data, decompresses (a*N+b)/a sub-code segments concurrently to obtain (a*N+ b)/a characters, and then determine the target decompressed data of the xth clock cycle from (a*N+b)/a characters according to the variable-length code table and the code segment read in the xth clock cycle. There is no need to wait for a codeword to be decompressed before decompressing the next codeword according to the length of the codeword. The problem of low decompression speed in the related technology is solved. During decompression, decompression can be started from multiple locations at the same time, and the decompression speed is relatively high.

Referring to accompanying drawing 6, it is a schematic structural diagram of a data processing system provided by an embodiment of the present invention, the data processing system includes a compression device 61 and a decompression device 63, the compression device 61 can be the compression device described in the foregoing embodiment, and the decompression The device 63 may be the decompression device described in the foregoing embodiments. The compression device 61 is used to compress the data to be compressed to obtain compressed data; the decompression device 63 is used to decompress the compressed data. The data processing system can be applied to a terminal or a server. Wherein, the terminal may include a smart phone, a tablet computer, a computer and a laptop computer, and the server may be a server, or a server cluster composed of several servers, or a cloud computing service center.

In the several embodiments provided in this application, it should be understood that the disclosed devices and methods may be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

Those of ordinary skill in the art can understand that all or part of the steps for implementing the above embodiments can be completed by hardware, and can also be completed by instructing related hardware through a program. The program can be stored in a computer-readable storage medium. The above-mentioned The storage medium mentioned may be a read-only memory, a magnetic disk or an optical disk, and the like.

The above descriptions are only optional embodiments of the embodiments of the present invention, 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 present application within the scope of protection.

Claims (15)

1. a kind of decompression method, it is characterised in that methods described includes：

The code length read from compressed data according to the xth clock cycle is the code section of a*N+b, (a*N+b) for obtaining/a subcode Section；Wherein, x is the integer more than or equal to 1；The code length of each subcode section is c in (a*N+b)/a subcode section, described (a*N+b)/a subcode section includes first subcode section, and the start bit of the first subcode section reads with the xth clock cycle The start bit of code section overlap, interval between the start bit of two neighboring subcode section in and (a*N+b)/a subcode section Code length is a-1, and the N is the integer more than 0；

Based on variable length coding table, decompression processing concomitantly is carried out to (a*N+b)/a subcode section, obtain (a*N+b)/a Character, wherein, decompressing each subcode section can obtain a character；

The variable length coding table includes multiple code words, and each character corresponds to the elongated volume in (a*N+b)/a character A code word in code table；The corresponding code length of at least two code words is different in the multiple code word, correspondence in the multiple code word The corresponding code length of code length code word most long is c, and the most short corresponding code length of code word of correspondence code length is a in the multiple code word, and a It is the greatest common divisor of the corresponding code length of each code word in the multiple code word, wherein, a is the integer more than or equal to 2, and c is Integer more than a, the difference of c and a is b；

According to the code section that the variable length coding table and the xth clock cycle read, determine from (a*N+b)/a character Go out the target decompression data of xth clock cycle, wherein, in the target decompression data of the xth clock cycle, per adjacent two Position of the last bit symbols of the corresponding code word of significant character in the code section that the xth clock cycle reads in significant character Position of the start element of code word corresponding with another significant character in the code section that the xth clock cycle reads is adjacent.

2. method according to claim 1, it is characterised in that

The code length that the xth clock cycle reads is that the rear b code length read with the (x+1)th clock cycle of the code section of a*N+b is B coincides before the code section of a*N+b.

3. method according to claim 1 and 2, it is characterised in that described according to the change in the case of being equal to 1 in x The code section that long codes table and the xth clock cycle read, the xth clock cycle is determined from (a*N+b)/a character Target decompression data, specifically include：

The target decompression data of xth clock cycle, the target of the xth clock cycle are determined from (a*N+b)/a character The starting of the code section that the start element of the corresponding code word of first significant character and the xth clock cycle read in decompression data Code element coincides.

4. method according to claim 1 and 2, it is characterised in that in the case where x is the integer more than or equal to 2, institute The code section read according to the variable length coding table and the xth clock cycle is stated, is determined from (a*N+b)/a character The target decompression data of xth clock cycle, specifically include：

Determine that c/a groups candidate decompresses data from (a*N+b)/a character；Every group of candidate's decompression packet includes multiple characters, And per in two adjacent characters, the code section that the last bit symbols of the corresponding code word of character read in the xth clock cycle In the position of the start element in code that the xth clock cycle reads section of position code word corresponding with another character be Adjacent；The start element of the corresponding code word of first character is read in the xth clock cycle in every two groups of candidates decompression data Position in the code section for taking differs, and in every group of candidate decompression data the corresponding code word of first character start element The W*a code element in the code section read for the xth clock cycle, W is more than or equal to 0 and whole less than or equal to b/a Number；

Determine that the target of xth clock cycle decompresses data from c/a groups candidate decompression data；In the compressed data In, the last bit symbols and the xth clock of the corresponding code word of last character in the target decompression data of the clock cycle of xth -1 The start element of the corresponding code word of first character is adjacent, the clock cycle of xth -1 in the target decompression data in cycle Target decompression data refer to the corresponding target decompression data of the code section for reading clock cycle of xth -1.

5. method according to claim 1, it is characterised in that the target for determining the xth clock cycle decompress data it Afterwards, methods described also includes：

According to the compressed data, the decompression data of the multiple clock cycle for obtaining are spliced, obtained the compressed data Corresponding decompression data, in two clock cycle adjacent after splicing, in the target decompression data of clock cycle last The target decompression number of position and another clock cycle of the last bit symbols of the corresponding code word of individual character in the compressed data It is adjacent according to position of the start element of the corresponding code word of interior first character in the compressed data.

6. a kind of compression method, it is characterised in that methods described includes：

Determine the target common divisor of the corresponding code length of each code word in variable length coding table；The variable length coding table includes multiple codes Word, the target common divisor is the integer more than or equal to 2, and each character in data to be compressed corresponds to the variable-length encoding A code word in table；

Each character and the target common divisor in the data to be compressed, generate the variable length coding table；The change The code length of at least two code words is different in long codes table, the probability of occurrence corresponding code word of character higher in the data to be compressed Code length of the code length less than the relatively low corresponding code word of character of probability of occurrence, the maximum of the code length of code word in the variable length coding table Common divisor is the target common divisor；

The data to be compressed are compressed according to the variable length coding table, obtain compressed data；

Based on the compressed data and variable length coding table generation compressed data encapsulation.

7. method according to claim 6, it is characterised in that the corresponding code of each code word in the determination variable length coding table Target common divisor long, including：

Determine at least one common divisor, according to the variable length coding table corresponding to each common divisor at least one common divisor, The data to be compressed are compressed, the compression ratio for obtaining respectively less than preset value, wherein, according at least one common divisor In each common divisor, can generate a corresponding variable length coding table；

Determine the target common divisor from least one common divisor, the numerical value of the target common divisor more than it is described extremely The numerical value of each common divisor in a few common divisor in addition to the target common divisor.

8. a kind of decompressing device, it is characterised in that the decompressing device includes：

Subcode section determining unit, the code length for being read from compressed data according to the xth clock cycle is the code section of a*N+b, is obtained (a*N+b) for arriving/a subcode section；Wherein, x is the integer more than or equal to 1；Per height in (a*N+b)/a subcode section The code length of code section is c, and (a*N+b)/a subcode section includes first subcode section, the start bit of the first subcode section and The start bit of the code section that the xth clock cycle reads overlaps, and two neighboring subcode section in (a*N+b)/a subcode section Start bit between be spaced code length be a-1, the N is the integer more than 0；

Decompression units, for based on variable length coding table, concomitantly carrying out decompression processing to (a*N+b)/a subcode section, obtain To (a*N+b)/a character, wherein, decompressing each subcode section can obtain a character；The variable length coding table includes multiple Code word, each character corresponds to a code word in the variable length coding table in (a*N+b)/a character；The multiple code The corresponding code length of at least two code words is different in word, and the correspondence code length corresponding code length of code word most long is c in the multiple code word, The most short corresponding code length of code word of correspondence code length is a in the multiple code word, and a is each code word correspondence in the multiple code word Code length greatest common divisor, wherein, a is the integer more than or equal to 2, and c is the integer more than a, and the difference of c and a is b；

Decompression data determination unit, for the code section read according to the variable length coding table and the xth clock cycle, from described (a*N+b) the target decompression data of xth clock cycle are determined in/a character, wherein, the target solution of the xth clock cycle In pressure data, per the last bit symbols of the corresponding code word of significant character in two adjacent significant characters in the xth clock The start element of the position code word corresponding with another significant character in the code section that the cycle reads is read in the xth clock cycle Position in the code section for taking is adjacent.

9. decompressing device according to claim 1, it is characterised in that

The code length that the xth clock cycle reads is that the rear b code length read with the (x+1)th clock cycle of the code section of a*N+b is B coincides before the code section of a*N+b.

10. decompressing device according to claim 8 or claim 9, it is characterised in that in the case of being equal to 1 in x, decompresses number According to determining unit, specifically for：

The target decompression data of xth clock cycle, the target of the xth clock cycle are determined from (a*N+b)/a character The starting of the code section that the start element of the corresponding code word of first significant character and the xth clock cycle read in decompression data Code element coincides.

11. decompressing device according to claim 8 or claim 9, it is characterised in that in the feelings that x is the integer more than or equal to 2 Under condition, the decompression data determination unit is specifically included：

Determine that c/a groups candidate decompresses data from (a*N+b)/a character；Every group of candidate's decompression packet includes multiple characters, And per in two adjacent characters, the code section that the last bit symbols of the corresponding code word of character read in the xth clock cycle In the position of the start element in code that the xth clock cycle reads section of position code word corresponding with another character be Adjacent；The start element of the corresponding code word of first character is read in the xth clock cycle in every two groups of candidates decompression data Position in the code section for taking differs, and in every group of candidate decompression data the corresponding code word of first character start element The W*a code element in the code section read for the xth clock cycle, W is more than or equal to 0 and whole less than or equal to b/a Number；

Determine that the target of xth clock cycle decompresses data from c/a groups candidate decompression data；In the compressed data In, the last bit symbols and the xth clock of the corresponding code word of last character in the target decompression data of the clock cycle of xth -1 The start element of the corresponding code word of first character is adjacent, the clock cycle of xth -1 in the target decompression data in cycle Target decompression data refer to the corresponding target decompression data of the code section for reading clock cycle of xth -1.

12. decompressing devices according to claim 8, it is characterised in that the decompressing device also includes：

Concatenation unit, for according to the compressed data, the decompression data of the multiple clock cycle for obtaining being spliced, obtains The corresponding decompression data of the compressed data, in two clock cycle adjacent after splicing, the target decompression of clock cycle Position of the last bit symbols of the corresponding code word of last character in the compressed data and another clock cycle in data Target decompression data in the position of the start element in the compressed data of the corresponding code word of first character be adjacent.

13. a kind of compression sets, it is characterised in that the compression set includes：

Common divisor determining unit, the target common divisor for determining the corresponding code length of each code word in variable length coding table；The change Long codes table includes multiple code words, and the target common divisor is the integer more than or equal to 2, each character in data to be compressed Corresponding to a code word in the variable length coding table；

Coding schedule generation unit, for each character in the data to be compressed and the target common divisor, generates institute State variable length coding table；The code length of at least two code words is different in the variable length coding table, probability of occurrence in the data to be compressed The code length of the code length of the corresponding code word of the character higher character corresponding code word relatively low less than probability of occurrence, the variable-length encoding The greatest common divisor of the code length of code word is the target common divisor in table；

Compression unit, for being compressed to the data to be compressed according to the variable length coding table, obtains compressed data；

Encapsulation generation unit, for based on the compressed data and variable length coding table generation compressed data encapsulation.

14. compression sets according to claim 13, it is characterised in that the common divisor determining unit, are used for：

Determine at least one common divisor, according to the variable length coding table corresponding to each common divisor at least one common divisor, The data to be compressed are compressed, the compression ratio for obtaining respectively less than preset value, wherein, according at least one common divisor In each common divisor, can generate a corresponding variable length coding table；

Determine the target common divisor from least one common divisor, the numerical value of the target common divisor more than it is described extremely The numerical value of each common divisor in a few common divisor in addition to the target common divisor.

15. a kind of data handling systems, it is characterised in that the system includes compression set and decompressing device,

The compression set includes the compression set described in claim 13 or 14, and the compression set is used to treat compressed data It is compressed and obtains compressed data；

The decompressing device includes any described decompressing devices of claim 8-12；The decompressing device is used for institute Stating compressed data carries out decompression processing.