WO2011150810A1

WO2011150810A1 - Data encoding method, decoding method, encoder and decoder

Info

Publication number: WO2011150810A1
Application number: PCT/CN2011/074930
Authority: WO
Inventors: 马鸿飞; 周佳; 钟政良; 刘怡; 夏雨
Original assignee: 华为技术有限公司
Priority date: 2010-09-30
Filing date: 2011-05-30
Publication date: 2011-12-08
Also published as: CN101980464B; CN101980464A

Abstract

A data encoding method, decoding method, encoder and decoder are provided by the present invention. In the encoding method provided by the embodiments of the present invention, high bit data of communication data represent the main information of the communication data and have the characteristic of non-uniform distribution, and low bit data represent the secondary information of the communication data and are non-uniformly distributed in the whole dynamic range and more uniformly distributed in a local dynamic range. The embodiments of the present invention perform entropy encoding to the high bit data of each sampling point, and perform the entropy encoding or quantification encoding to the low bit data, which fully utilize the different distribution characteristics of the high bit data and the low bit data and have the higher encoding efficiency and higher compression ratio by selecting corresponding encoding means.

Description

Data encoding method, decoding method, encoder and decoder

This application claims priority to Chinese Patent Application No. 201010505926.8, entitled "Data Encoding Method, Decoding Method, Encoder and Decoder", filed on September 30, 2010, the entire contents of which are hereby incorporated by reference. Combined in this application.

TECHNICAL FIELD The present invention relates to the field of communications technologies, and in particular, to a compression encoding method, a decoding decompression method, an apparatus, and a communication system. BACKGROUND OF THE INVENTION With the increasing types of communication services and traffic, the amount of data that needs to be transmitted in a communication network is drastically increased, and thus more channel resources and storage space are required.

The signals transmitted in the existing communication network usually include voice, image and video signals, etc. These multimedia signals have high correlation and statistical characteristics, have good mathematical and physical models, and thus have better data encoding methods. .

The inventor found in the process of studying the prior art that the existing data compression algorithm is mainly for multimedia data. Since the communication data generally does not have the high correlation and statistical characteristics of the multimedia data, the data encoding method is directly applied to the communication data. When compressed, the compression ratio is not high enough. SUMMARY OF THE INVENTION The present invention provides a data encoding method, a decoding method, an encoder, and a decoder that can efficiently compress communication data.

In order to solve the above technical problem, the embodiment of the present invention is implemented by the following technical solutions: On the one hand, a data encoding method is provided, including:

Obtaining the positive and negative polarity of each sample in the data block;

Obtaining the most significant bit of the sample with the largest absolute value among the samples of the data block;

Obtaining n-bit high-bit data starting from the most significant bit in each sample of the data block, and performing first-class entropy coding on the high-bit data of each sample, where n is greater than 1 and less than The number of bits of valid data of the sample with the largest absolute value;

Obtaining the data from the most significant bit in each sample of the data block except the high The low bit data other than the bit data is subjected to the first type of quantization coding or the second type of entropy coding for each of the low bit data of each sample;

And outputting the encoded code frame, where the coded frame includes the most significant bit, a positive and negative codeword of each sample carrying the data block, and a first type of entropy coding of each sample of the data block The subsequent high bit data and the first type of quantized code or the second type of entropy encoded low bit data.

In another aspect, a data decoding method is provided, including:

Obtaining a codeword carrying positive and negative polarities of each sample in the received coded frame, and obtaining positive and negative polarities of each sample according to the codeword;

Decoding high-bit data of the entropy-encoded data in the encoded frame to obtain high-specificity data of each sample;

Decoding the quantized or entropy encoded low bit data in the encoded frame to obtain low bit data of each sample;

The original data block is obtained by assembling the positive and negative polarities of each sample, the high bit data of each sample, and the low bit data according to the most significant bits contained in the received encoded frame.

In another aspect, a data encoding method is provided, including:

Obtaining the positive and negative polarity of each sample in the data block;

Obtaining the most significant bit p of the sample with the largest absolute value among the samples of the data block;

Obtaining the most significant bit d of each sample of the data block, and shifting the effective data bits of each sample to the left by p-d bits;

Performing a fourth type of entropy coding on the left shift number p-d of each sample of the data block;

Performing a fifth type of entropy coding or a second type of quantization coding on the shifted valid data bits in each sample of the data block;

And outputting the encoded code frame, wherein the coded frame includes the most significant bit P, a codeword carrying positive and negative polarity of each sample, and a fourth type of entropy coding of each sample of the data block. The left shift number and the fifth type of entropy coding or the second type of quantized coded valid data.

In another aspect, a data decoding method is provided, including:

Obtaining a codeword carrying positive and negative polarities of each sample in the encoded frame, and obtaining positive and negative polarities of each sample according to the codeword; Decoding an entropy-encoded left shift number in the encoded frame to obtain a left shift number of valid data of each sample point;

Decoding the quantized or entropy encoded valid data in the encoded frame to obtain the shifted valid data of each sample;

The positive and negative polarities of each sample, the number of left shifts of valid data of each sample, and the valid data after shifting are assembled according to the most significant bit of the sample having the largest absolute value included in the encoded frame. Raw data block.

In another aspect, an encoder is provided, including:

a first pre-processing unit, configured to obtain positive and negative polarities of each sample in the data block, and a most significant bit of the sample with the largest absolute value among the samples of the data block;

a high-bit coding unit, configured to acquire n-bit high-bit data from each of the most significant bits in each sample in the data block, and perform first-class entropy coding on each of the high-bit data of each sample The n is greater than 1, and is less than the number of bits of valid data of the sample having the largest absolute value;

a low bit coding unit, configured to acquire low bit data other than the high bit data in data starting from the most significant bit in each sample in the data block, low for each sample The bit data is respectively subjected to the first type of quantization coding or the second type of entropy coding;

a first coding output unit, configured to output an encoded code frame, where the coded frame includes the most significant bit, a codeword carrying positive and negative polarity of each sample in the data block, and each of the data blocks The first type of entropy encoded high bit data of the sample and the encoded frame composed of the first type of quantized code or the second type of entropy coded low bit data.

In another aspect, a decoder is provided, including:

a first polarity acquiring unit, configured to acquire a codeword carrying positive and negative polarities of each sample in the received encoded frame, and obtain positive and negative polarities of each sample according to the codeword;

a high bit decoding unit, configured to decode the entropy encoded high bit data in the encoded frame to obtain high bit data of each sample;

a low bit decoding unit, configured to decode the quantized or entropy encoded low bit data in the encoded frame to obtain low bit data of each sample;

a first data assembling unit for using the most significant bit included in the received encoded frame The positive and negative polarities, high bit data, and low bit data of each sample are assembled to obtain the original data block.

In another aspect, an encoder is provided, including:

a second pre-processing unit, configured to obtain positive and negative polarities of each sample in the data block and a most significant bit p of the sample with the largest absolute value among the samples of the data block;

a shifting unit, configured to obtain a most significant bit d of each sample in the data block, and shift the valid data bits of each sample to the left by p-d bits;

a shift information encoding unit, configured to perform a fourth type of entropy encoding on each of the left shift numbers p-d of each sample in the data block;

a valid data encoding unit, configured to perform a fifth type of entropy coding or a second type of quantization coding on the shifted valid data bits in each sample in the data block;

a second code output unit, configured to output the encoded coded frame, where the coded frame includes the most significant bit P, a codeword carrying positive and negative polarity of each sample in the data block, and each of the data blocks The fourth type of entropy-coded left shift number of the samples and the fifth type of entropy coding or the second type of quantized coded valid data.

In another aspect, a decoder is provided, including:

a second polarity acquiring unit, configured to acquire a codeword carrying positive and negative polarities of each sample in the encoded frame, and obtaining positive and negative polarities of each sample according to the codeword;

a left shift number decoding unit, configured to decode the entropy encoded left shift number in the encoded frame, to obtain a left shift number of valid data of each sample point;

An effective data decoding unit, configured to decode the quantized or entropy encoded valid data in the encoded frame to obtain the shifted valid data of each sample;

a second data assembling unit, configured to: positive and negative polarity of each sample, left shift number of valid data of each sample, and shift according to a most significant bit of a sample having the largest absolute value included in the encoded frame The valid data after the bit is assembled to obtain the original data block.

In the above technical solution, the high-bit data of the communication data represents the main information of the communication data, and has the characteristics of non-uniform hook distribution, and the low-bit data represents the secondary information of the communication data, and is within the entire dynamic range. Non-uniform distribution, the division is relatively uniform within the local dynamic range, the present invention The embodiment entropy encodes the high bit data of each sample, entropy encodes or quantizes the low bit data, and fully utilizes different distribution characteristics of the high bit data and the low bit data, by selecting the corresponding code. In this way, it can have higher coding efficiency and higher compression. BRIEF DESCRIPTION OF THE DRAWINGS In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the prior art and the embodiments will be briefly described below. Obviously, the drawings in the following description are only Some embodiments of the invention may also be used to obtain other figures from these figures without departing from the art.

1 is a flowchart of a data encoding method according to Embodiment 1 of the present invention;

2 is a flowchart of a method for decoding a data according to Embodiment 2 of the present invention;

3 is a schematic structural diagram of original data before encoding in an embodiment of the present invention;

4 is a schematic structural diagram of a data block before encoding in an embodiment of the present invention;

FIG. 5 is a schematic diagram of dividing a data block in a data encoding method according to Embodiment 1 of the present invention; FIG.

6 is a probability distribution diagram of original data in an embodiment of the present invention;

7 is a probability distribution diagram of high bit data of original data in the embodiment of the present invention; FIG. 8 is a probability distribution diagram of low bit data of original data in the embodiment of the present invention; FIG. 9 is a valid bit in the embodiment of the present invention; a probability distribution map of low bit data of raw data having a bit number Q of 6;

Figure 10 is a probability distribution diagram of low bit data of original data of which the effective bit number Q is 7 in the embodiment of the present invention;

11 is a schematic structural diagram of an encoded frame in Embodiment 1 of the present invention;

FIG. 12 is a flowchart of a method for data encoding according to Embodiment 3 of the present invention; FIG.

13 is a flowchart of a method for decoding a data provided by Embodiment 4 of the present invention;

14 is a schematic diagram of dividing a data block in an improved data encoding method according to Embodiment 3 of the present invention;

15 is a probability distribution diagram of a left shift number of original data in an embodiment of the present invention;

16 is a probability distribution diagram of valid data of original data in an embodiment of the present invention; 17 is a probability distribution diagram of valid data of original data with a most significant bit number P of 7 in the embodiment of the present invention;

18 is a probability distribution diagram of valid data of original data with a most significant bit number P of 8 in the embodiment of the present invention;

19 is a schematic structural diagram of an encoded frame according to Embodiment 3 of the present invention;

20 is a schematic structural diagram of an encoder according to Embodiment 5 of the present invention;

21 is a schematic block diagram of an encoder according to Embodiment 5 of the present invention;

FIG. 22 is a schematic structural diagram of a decoder according to Embodiment 6 of the present invention; FIG.

23 is a schematic block diagram of a decoder provided in Embodiment 6 of the present invention;

24 is a schematic structural diagram of an encoder according to Embodiment 7 of the present invention;

25 is a schematic block diagram of an encoder provided in Embodiment 7 of the present invention;

26 is a schematic structural diagram of a decoder according to Embodiment 8 of the present invention;

27 is a schematic block diagram of a decoder provided in Embodiment 8 of the present invention;

28 is an application scenario diagram of a decoder and an encoder according to an embodiment of the present invention;

FIG. 29 is still another application scenario diagram of a decoder and an encoder provided by an embodiment of the present invention. The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. example. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without the creative work are all within the scope of the present invention.

The present invention provides a data encoding method, a decoding method, a decoder and an encoder. In order to better understand the technical solutions of the present invention, the embodiments provided by the present invention will be described in detail below with reference to the accompanying drawings.

Referring to FIG. 1, FIG. 1 is a flowchart of a data encoding method according to Embodiment 1 of the present invention.

In the embodiment of the present invention, the communication data that needs to be encoded may be in units of data blocks. The number of samples included in the data block can be determined according to the compression ratio of the code and the delay requirement. For example, the lower the delay requirement, the larger the number of samples included in the data block, and the higher the compression ratio, the data block. The more samples are included in the sample. The data encoding method provided in the first embodiment of the present invention may include: 101. Obtain the positive and negative polarity of each sample in the data block.

Specifically, the encoder can obtain the positive and negative polarities of each sample in the data block that needs to be encoded. Among them, the data block is composed of L samples, and the size of L is adjustable. Each sample in the data block can be represented in the form of a complement.

The encoder can obtain the positive and negative polarity of each sample in the data block according to the sign bit of each sample.

102. Acquire a most significant bit of the sample with the largest absolute value among the samples of the data block.

Specifically, the encoder can take an absolute value for each of the L samples of the data block, and then obtain the sample with the largest absolute value, and obtain the most significant bit of the sample with the largest absolute value. In the embodiment of the present invention, the highest bit of the bit of the absolute value of the sample is referred to as the most significant bit of the sample. In other words, the most significant bit, the first non-zero bit from the right to left of the non-sign bit is the most significant bit. It will be appreciated that the most significant bit may also be suitably higher than the non-zero bit.

103. Obtain n-bit high-bit data starting from the most significant bit in each sample, and perform first-class entropy coding on the high-bit data of each sample, where n is greater than 1, and is smaller than the absolute maximum. The number of bits of valid data for the point.

In the embodiment of the present invention, after obtaining the most significant bit of the data block, the encoder divides the right number of bits from the most significant bit into high-bit data, and the remaining bits are divided into low-bit data, where n An integer that is greater than 1, less than the number of significant digits of the valid data for the sample with the largest absolute value. Wherein, the size of the n value can be determined according to the statistical rule of the high bit data of different communication data. In the embodiment of the present invention, bits other than the sign bit, the sign extension bit, and the padded zero in each sample are referred to as valid data. For example, -5 , expressed in the form of complement, is 10000101, the first digit is a sign bit, with "1" for negative polarity, "000" for filled zero, and "101" for valid data.

Specifically, the encoder performs first-class entropy coding on the high-bit data of each sample, for example, performing Huffman coding or arithmetic coding.

104. Obtain low-bit data other than the high-bit data in the data starting from the most significant bit in each sample, and perform first-class quantization coding or second on each low-bit data of each sample. Class entropy coding.

Specifically, the encoder obtains low-bit data other than the high-bit data in the data starting from the most significant bit in each sample, and performs the first-class quantization encoding on the low-bit data of each sample respectively. Or a second type of entropy coding, for example, the second type of entropy coding may be Huffman coding, the first type of quantization coding It can be non-equalized and quantized.

In the embodiment of the present invention, the types of the first type of entropy coding, the second type of entropy coding, and the third type of entropy coding are entropy coding, but the specific entropy coding type may be determined according to specific statistical rules of data, such as Huffman coding, arithmetic coding and run length coding.

105. Output an encoded code frame, where the coded frame includes a most significant bit, a codeword carrying positive and negative polarity of each sample, a first type of entropy coded high bit data of each sample, and a first A type of quantized coding or a second type of entropy encoded low bit data.

Specifically, after encoding the high bit data and the low bit data, the encoder outputs the encoded code frame, where the code frame includes the most significant bit, and the code word carrying the positive and negative polarity of each sample. a first type of entropy encoded high bit data of each sample and a coded frame composed of a first type of quantized code or a second type of entropy coded low bit data, wherein the plurality of coded frames constitute an encoded bit stream, Data encoding process. It will be appreciated that the most significant bit can be carried in the header of the encoded frame.

In the embodiment of the present invention, the high-bit data of the communication data has the characteristics of a non-uniform hook distribution, and the low-bit data is non-uniformly distributed in the entire dynamic range, and the segments are relatively uniform in the local dynamic range. The first type of entropy coding is performed on the high bit data of each sample, and the second type entropy coding or the first type of quantization coding is performed on the low bit data, which fully utilizes different distributions of high bit data and low bit data. The feature, by selecting the corresponding coding mode, can have higher coding efficiency and higher compression.

Further, in the embodiment of the present invention, the code characters carrying the positive and negative polarities of each sample may be the positive and negative polarities of each sample without entropy coding, or the third type of entropy of each sample. Positive and negative polarity after encoding. Correspondingly, the coded frame outputted after coding includes the most significant bit, the third type of entropy coded positive and negative polarity of each sample, the first type of entropy coded high bit data, and the first type of quantization code. Or a second type of entropy encoded low bit data. In the embodiment of the present invention, the third type of entropy coding for the positive and negative polarities of each sample can further improve the coding efficiency.

In the embodiment of the present invention, when the high-bit data and the low-bit data are encoded, a preset encoding mode and an encoding parameter may be selected, and the encoding mode and the encoding parameter are collectively referred to as encoding information in the embodiment of the present invention. In addition, when the encoder performs the encoding of the high-bit data and the low-bit data, the encoding method and the encoding parameter can be flexibly selected according to different types of communication data. The encoding method provided by the embodiment of the present invention further includes: acquiring The first class of entropy coding for the high bit data of each sample Code information and low bit data The first type of quantized code or the second type of entropy coded coded information. After obtaining the encoded information, the encoder transmits the encoded information through the encoded frame, so that the decoder performs corresponding decoding according to the encoded information. In the embodiment of the present invention, the encoded frame further includes a first type of entropy encoded coding information and a low bit data second type entropy coding or a first type of quantized coding information of the high bit data of each sample.

In the embodiment of the present invention, by adding coding information to the coded frame, the encoder can select more coding modes by performing entropy coding or quantization coding, without being limited by the preset coding mode, and the adaptation range is wider. . Embodiment 2

Referring to FIG. 2, FIG. 2 is a flowchart of a method for decoding a data according to Embodiment 2 of the present invention. In the embodiment of the present invention, after the decoder receives the encoded bit stream sent by the encoder, the decoder decodes the encoded bit stream in units of frames. The data encoding method provided in the second embodiment of the present invention may include:

201. Acquire a codeword carrying positive and negative polarity of each sample in the received coded frame, and obtain positive and negative polarity of each sample according to the codeword.

Specifically, the decoder obtains the code words carrying the positive and negative polarities of each sample in the encoded frame, and obtains the positive and negative polarities of each sample according to the code words. Among them, the code characters carrying the positive and negative polarities of each sample can be the positive and negative polarity of each sample, and the decoder can obtain the positive and negative polarity of each sample without decoding.

202. Decode the high-bit data after entropy coding in the encoded frame to obtain high-bit data of each sample.

Specifically, the decoder can decode the entropy-encoded high-bit data in the encoded frame according to the decoding manner corresponding to the encoding mode according to the number of samples included in the data block, and obtain high-bit data of each sample. .

The coding mode used by the encoder and the decoding mode used by the decoder may be preset coding modes and decoding modes.

203. Decode low-bit data after quantization coding or entropy coding in the encoded frame to obtain low-bit data of each sample. Specifically, the decoder may decode the quantized or entropy-coded low-bit data in the encoded frame according to the decoding mode corresponding to the encoding mode used by the encoder according to the number of samples included in the data block, thereby obtaining each Low bit data of samples.

204. Assemble the positive and negative polarities, the high bit data, and the low bit data of each sample according to the most significant bits included in the received encoded frame to obtain the original data block.

In the embodiment of the present invention, each sample is stored in the form of a complement, and the number of bits of the sample is, for example, 16 bits, 32 bits, or the like. After obtaining the positive and negative polarity of each sample, the decoded high bit data, and the decoded low bit data, the decoder will each sample in the order of the sign bit, the high bit data, and the low bit data. The positive and negative polarities of the points, the corresponding high bit data, and the low bit data are assembled to obtain the original data block, and the data decoding process is completed. Wherein, in the process of data assembly, it is necessary to determine the number of zeros in the sign extension bits of each sample according to the most significant bits contained in the received coded frame.

The decoding method provided by the embodiment of the present invention provides support for the encoding method provided in the first embodiment. Since the high bit data of the communication data represents the main information of the communication data, it has a non-uniform distribution characteristic, and the low bit data represents The secondary information of the communication data is non-uniformly distributed over the entire dynamic range, and the segments are relatively uniform within the local dynamic range. In the embodiment of the present invention, the high-bit data of each sample is entropy encoded, and the low-bit is used. The bit data is entropy coded or quantized, and the different distribution characteristics of the high bit data and the low bit data are fully utilized. By selecting the corresponding coding mode, the coding efficiency can be high and the data compression is relatively high.

Further, in the embodiment of the present invention, the positive and negative polarities of each sample are obtained according to the codeword, and the positive and negative polarity codewords of each sample are separately entropy decoded to obtain each sample point. Positive and negative polarity. The decoding method for entropy decoding the positive and negative polarity code words corresponds to the coding method for the third type entropy coding for the positive and negative polarities of each sample.

Further, the encoded frame sent by the encoder may further include high bit data entropy encoded coding information and low bit data entropy coding or quantized coding information of each sample, where the coding information includes an encoding mode and an encoding parameter. The coding parameters may include a quantization table used for coding. It can be understood that the encoded information can be carried in the header of the encoded frame.

In the case that the encoded frame carries the encoded information, the step of decoding the high-bit-encoded high-bit data in the encoded frame to obtain the high-bit data of each sample is as follows: The high-bit data entropy-encoded coding information of the sample entropy-decodes the entropy-encoded high-bit data in the encoded frame to obtain high-bit data of each sample. The decoding method for entropy decoding the entropy-encoded high-bit data is determined according to the encoding information, that is, the entropy-encoded high-bit data is decoded by using the encoding method and the corresponding decoding method.

In addition, in the case that the encoded information is carried in the encoded frame, the step of decoding the low-bit data of the quantized or entropy-encoded data in the encoded frame to obtain the low-bit data of each sample is specifically: The quantized or entropy encoded low bit data in the encoded frame is decoded according to the low bit data entropy encoded or quantized encoded information of each sample to obtain low bit information of each sample. The decoding method for decoding the quantized or entropy encoded low-bit data in the encoded frame is determined according to the encoding information, that is, the encoding and the encoding method and the corresponding decoding manner are low after entropy encoding or quantization encoding. The bit data is decoded.

Further, in the embodiment of the present invention, the step of decoding the low-bit data of the quantized or entropy-encoded data in the encoded frame to obtain the low-bit data of each sample may include:

1) Entropy decoding the entropy-encoded low-bit data in the encoded frame to obtain low-bit data of each sample.

2) Perform inverse quantization processing on the quantized low-bit data in the encoded frame to obtain low-bit data of each sample, thereby recovering the low-bit data with distortion. For a more detailed understanding of the data encoding method provided in Embodiment 1 of the present invention and the data decoding method provided in Embodiment 2, a specific application scenario of the embodiment of the present invention is further given below.

Referring to FIG. 3, FIG. 3 is a schematic structural diagram of original data before encoding in an embodiment of the present invention.

Where S denotes the bit occupied by the sign bit, a total of m bits, and D denotes the bit occupied by the valid data, a total of n bits. In the figure, B _H represents the highest bit of the data, B _L represents the lowest bit of the data, and B _M represents the highest bit of the data valid bit.

Referring to FIG. 4, FIG. 4 is a schematic structural diagram of a data block before encoding according to an embodiment of the present invention.

The encoder encodes the original data in units of data blocks. According to the distribution characteristics of the original data, the S area and the D area of each sample in the data block are random.

Referring to FIG. 5, FIG. 5 is a schematic diagram of dividing a data block in an embodiment of the present invention. The encoder divides the data block to obtain a sign bit, a high bit, and a low bit, each of which occupies n0, n1, and n2 bits, respectively. When the encoder divides the data block, it first extracts the sign bit Ss of each sample point in the data block, and obtains the positive and negative polarity of each sample according to the sign bit. The encoder can take an absolute value for each sample, obtain the sample with the largest absolute value among the L samples, and determine the most significant bit of the sample. After determining the most significant bit, the n1 bits from the most significant bit in each sample are divided into high bit data S _h , and the remaining n 2 bits in the valid data are divided into low bit data S .

The embodiment of the present invention can use the states 0 and 1 to indicate the positive and negative polarity of each sample in each frame of data. For example, 0 indicates that the sample is negative, 1 indicates that the sample is positive; or 0 indicates a sample. The value is positive and 1 means the sample value is negative. After the sign bit of each sample is processed, a binary code word S _s c carrying the positive and negative polarities of each sample is obtained.

Figure 6 shows an example of the probability distribution of the original data, and the probability distribution of the corresponding high-bit data is given by Figure 7. It can be seen that the probability of occurrence of the high bit data S _h is unequal, which provides a basis for data compression using entropy coding. Therefore, the embodiment of the present invention uses the entropy coding method to perform high bit data _Sh Undistorted entropy coding yields high bit data entropy coding S _h e. Entropy coding may employ Huffman coding, arithmetic coding or other unequal length coding algorithms, or a combination thereof. The coding information such as the entropy coding and the coding parameters for entropy coding of the high bit data may be stored in the frame header information, and the frame header information is stored in the coding frame for output. And there are two methods of distortion coding (quantization coding). In the distortion coding mode, in order to keep the encoder low algorithm complexity, a simple scalar quantization method can be used. When performing quantization coding, it is necessary to acquire the dynamic range of the low bit data, determine the scalar quantization table based on the dynamic range, and then perform encoding according to the scalar quantization table.

In practical applications, the effective data is non-uniformly distributed over the entire dynamic range, but distributed evenly within the local dynamic range. Figure 6 shows an example of the probability distribution of the original data S. The probability distribution of the low-bit data corresponding to it is given by Figure 8. It can be seen that the low-bit data is non-uniformly distributed as a whole, and the dynamic range is large. . In the case where the effective bit number Q of the low bit data is known, classification can be performed, and FIG. 9 and FIG. 10 respectively show low bit data of the effective bit number Q of 6 and 7. The probability distribution shows that the distribution is relatively uniform and the dynamic range is small. If the two parts of data are separately quantified, the quantization is more accurate. Therefore, the compression coding algorithm for low-bit data can use the classified quantization coding algorithm, for example, the Lloyd-Max non-uniform scalar quantization algorithm. The coding information such as the entropy coding or the quantization coding of the low bit data and the coding information such as the coding parameters may be stored in the frame header information, and the frame header information is output through the coding frame.

In the embodiment of the present invention, the encoder combines the positive and negative polarity encoded codeword S _s c , the high bit code S _h c and the low bit code S _{lC obtained by} encoding, and combines the frame header information to form a complete The coded frame, FIG. 11 is a schematic structural diagram of the coded frame in the coded bit stream B1 provided by the embodiment of the present invention. The different coded frames are arranged in chronological order of the output to form an encoded bit stream B1 output by the encoder, as shown in FIG. The frame header information may further include the number of samples in the data block, the most significant bit in the data block, and the like. It should be noted that, if the encoder and the decoder use the preset decoding type and encoding type, and the length of each sample in the data block and the number of samples are preset values, the encoded frame may be No need to carry frame header information.

At the decoder, the encoded bit stream is decoded in units of coded frames, the decoder can read the frame header information and save, and then the positive and negative polarities of each sample are determined based on the 0, 1 state of the binary bit stream. Secondly, the decoder performs distortion-free entropy decoding to recover high-bit data and save it. The specific method of entropy decoding corresponds to the encoder. In the case where the encoder obtains the entropy encoding of the low bit data according to the frame header information, the decoder determines the effective bit number of the low bit data through the frame header information, and performs entropy decoding on the entropy encoded low bit data. Further, distortion-free decoding of low-bit data is achieved. When there is distortion coding, the decoder performs inverse quantization processing on the encoded low-bit data, and the specific process includes: recovering the quantized subscript value of the low-bit data from the input bit stream, and then determining the low-bit according to the frame header information. The dynamic range of the data is used to determine the scalar quantization code table, perform inverse quantization processing, search for the corresponding quantized value in the quantization table by the subscript index, and realize distortion decoding of the low bit data bits. Finally, the decoder assembles the positive and negative polarities, high-bit data and low-bit data of each sample obtained by decoding, obtains the original data block and outputs, completes decoding of one encoded frame, and prepares to proceed to the next frame. Decoding until the end of decoding. Embodiment 3

Referring to FIG. 12, FIG. 12 is a flowchart of a method for encoding a data according to Embodiment 3 of the present invention. The data encoding method provided by the embodiment of the present invention may include:

In the embodiment of the present invention, the communication data that needs to be encoded may be in units of data blocks. The number of samples included in the data block can be determined according to the compression ratio of the code and the delay requirement, that is, the lower the delay requirement, the more the number of samples included in the data block, the higher the compression ratio, the data block The more samples are included in the sample. The data encoding method provided in Embodiment 3 of the present invention may include:

301. Obtain the positive and negative polarity of each sample in the data block.

Specifically, the encoder can obtain the positive and negative polarities of each sample in the data block that needs to be encoded. Among them, the data block is composed of L samples, and the size of L is adjustable. Each sample in the data block is represented in the form of a complement.

302. Acquire a most significant bit p of the sample with the largest absolute value among the samples of the data block.

Specifically, the encoder can take an absolute value for each of the L samples of the data block, and then obtain the sample with the largest absolute value, and obtain the most significant bit of the sample with the largest absolute value. In the embodiment of the invention, the most significant bit of the absolute value of the sample is referred to as the most significant bit of the sample.

303. Obtain the most significant bit d of each sample, and shift the valid data bits of each sample to the left by p-d bits.

In the embodiment of the present invention, after obtaining the most significant bit of the data block, the encoder further obtains the most significant bit d of each sample, and shifts the effective data bits of each sample to the left by p-d bits. Wherein, the valid data bits of the sample point are data bits other than the sign bit in the two's complement of the sample.

304. Perform a fourth type of entropy coding on the left shift number p-d of each sample.

Specifically, the encoder performs a fourth type of entropy coding on the left shift number distribution of each sample, for example, performing Huffman coding or arithmetic coding.

305. Perform a fifth type of entropy coding or a second type of quantization coding on the shifted valid data bits in each sample.

Specifically, the encoder performs a fifth type of entropy coding or a second type of quantized coding on the shifted valid data bits in each sample, and the fifth type of entropy coding may be Huffman coding, etc., and the second type of quantization coding It can be non-equalized and quantized.

306. Output an encoded code frame, where the coded frame includes a most significant bit p, a codeword carrying positive and negative polarities of each sample, a fourth type of entropy coded left shift number, and a second type of quantization code. Or Five types of valid data after entropy coding.

Specifically, after encoding the left shift number and the shifted valid data of each sample, the encoder outputs the encoded code frame, where the coded frame includes the most significant bit p, and carries each sample. The positive and negative polarity of the code word, the fourth type of entropy coded left shift number of each sample, and the second type of quantized code or the fifth type of entropy coded valid data, multiple codes output by the encoder The frame constitutes an encoded bit stream, and the data encoding process is completed.

In the embodiment of the present invention, the left shift number generated after the shift of the communication data and the valid data generated after the shift reflect the distribution law of the original data. Wherein, the left shift number represents the main information of the communication data, and has a non-uniform distribution characteristic, and the shifted valid data represents the secondary information of the communication data, which is non-uniformly distributed throughout the dynamic range, in the local In the dynamic range, the segments are relatively uniform. In the embodiment of the present invention, the left shift number of each sample is entropy encoded, and the shifted valid data is entropy encoded or quantized, and the left shift number and shift are fully utilized. After the different distribution characteristics of the valid data, by selecting the corresponding coding mode, it is possible to have higher coding efficiency and higher compression.

Further, in the embodiment of the present invention, the code characters carrying the positive and negative polarities of each sample may be the positive and negative polarities of each sample without entropy coding, or the sixth entropy of each sample. Positive and negative polarity after encoding. Correspondingly, the coded frame outputted after encoding includes a sixth type of entropy coded positive and negative polarity of each sample, a fourth type of entropy coded left shift number, and a second type of quantized code or a fifth type of entropy. Encoded valid data. In the embodiment of the present invention, entropy coding of the positive and negative polarities of each sample can further improve the coding efficiency.

In the embodiment of the present invention, the types of the fourth type of entropy coding, the fifth type of entropy coding, and the sixth type of entropy coding are entropy coding, but the specific entropy coding type may be determined according to specific statistical rules of data. The types of the first type of quantization coding and the second type of quantization coding are both quantization codes, but the specific quantization coding type can be determined according to specific statistical rules of data.

In the embodiment of the present invention, a preset encoding mode and an encoding parameter may be selected when encoding the left shift number and the shifted valid data, and the encoding parameter may include a quantization table, an encoding method, and an encoding used in encoding. The parameters are collectively referred to as coding information in the embodiments of the present invention. In addition, when the encoder performs the encoding of the left shift number and the shifted valid data, the encoding method and the encoding parameter may be flexibly selected according to different types of communication data. The encoding method provided by the embodiment of the present invention further includes : Obtaining the left shift number of each sample, the fourth type of entropy coded coding information and the shifted effective data, the second class Quantization coding or coding information of the fifth type of entropy coding. After obtaining the encoded information, the encoder transmits the encoded information through the encoded frame, so that the decoder performs corresponding decoding according to the encoded information. In the embodiment of the present invention, the coded frame further includes a left shift number of each sample, and a fourth type of entropy coded coding information and a shifted effective data, a fifth type of entropy coding or a second type of quantization coded coding. information.

In the embodiment of the present invention, by adding coding information to the coded frame, the encoder can select more coding modes by performing entropy coding or quantization coding, without being limited by the preset coding mode, and the adaptation range is wider. . Embodiment 4

Referring to FIG. 13, FIG. 13 is a flowchart of a method for decoding a data according to Embodiment 4 of the present invention. In the embodiment of the present invention, after the decoder receives the encoded bit stream sent by the encoder, the decoder decodes the encoded bit stream in units of frames. The data encoding method provided in the fourth embodiment of the present invention may include:

401. Obtain a codeword carrying a positive and negative polarity of each sample in the encoded frame, and obtain positive and negative polarities of each sample according to the codeword.

402. Decode the entropy-encoded left shift number in the encoded frame to obtain a left shift number of the valid data of each sample.

Specifically, the decoder may decode the entropy-encoded left shift number in the encoded frame according to the decoding manner corresponding to the encoding manner according to the number of samples in the data block, and obtain the left shift number of the valid data of each sample point. .

403. Decode the quantized or entropy encoded valid data in the encoded frame to obtain the shifted valid data of each sample. Specifically, the decoder can decode the quantized or entropy-encoded valid data in the encoded frame according to the decoding manner corresponding to the encoding mode used by the encoder according to the number of samples in the data block, thereby obtaining the shift of each sample. Valid data after the bit.

404. Accomplish the positive and negative polarity of each sample, the left shift number of the valid data, and the valid data after the shift according to the most significant bit of the sample with the largest absolute value included in the encoded frame to obtain the original data block. .

In the embodiment of the present invention, each sample is stored in the form of a complement, and the number of bits of the sample is, for example, 16 bits, 32 bits, or the like. After obtaining the positive and negative polarity of each sample, the number of left shifts of the valid data, and the valid data after shifting, the decoder assembles the positive and negative polarities and the valid data of the sample to obtain the original data block, and completes Data decoding process. In the data assembly, the position of the shifted valid data needs to be shifted to the right by the corresponding number of bits, and the number of bits shifted to the right is equal to the number of left shifts. In the process of data assembly, it is necessary to determine the number of zeros in the sign extension bits of each sample according to the highest significant ratio of the sample having the largest absolute value contained in the received coded frame. The decoding method provided in the fourth embodiment of the present invention provides support for the decoding method provided in the third embodiment. The left shift number generated after the shift of the communication data and the valid data generated after the shift reflect the distribution rule of the original data. Wherein, the left shift number has a non-uniform hook distribution characteristic, and the shifted effective data is non-uniformly distributed in the entire dynamic range, and the partial dynamic range is relatively uniform in the local dynamic range, and the embodiment of the present invention is applicable to each sample point. The left shift number is entropy coded, and the shifted effective data is entropy coded or quantized, and the left shift number and the different distribution characteristics of the shifted effective data are fully utilized, and the corresponding coding mode may be selected. Higher coding efficiency and higher compression.

Further, in the embodiment of the present invention, the positive and negative polarities of each sample are obtained according to the codeword, and the positive and negative polarity codewords of each sample are separately entropy decoded to obtain each sample point. Positive and negative polarity. Among them, the decoding method for entropy decoding the positive and negative polarity code words corresponds to the coding method for the entropy coding of the positive and negative polarities of each sample at the transmitting end.

Further, the encoded frame sent by the encoder may further include encoding information of the left shift number entropy coding of each sample and encoding information of the shifted effective data entropy encoding or quantization encoding, where the encoding information includes an encoding method and Encoding parameters, encoding parameters may include a quantization table used for encoding. When the encoded information is carried in the encoded frame, the entropy encoded left shift number in the encoded frame is decoded, The step (402) of obtaining the left shift number of the valid data of each sample is specifically: performing entropy-coded left shift number in the encoded frame according to the encoding information of the left shift number entropy coding of each sample point Entropy decoding, obtaining the number of left shifts of valid data for each sample. The specific decoding mode for entropy decoding the entropy-encoded left shift number is determined according to the coding information, that is, the entropy-encoded left shift number is decoded by using a decoding method corresponding to the coding mode, for example, an encoding method. For entropy coding, the corresponding decoding mode is entropy decoding.

In addition, when the encoded information is carried in the encoded frame, the step of decoding the quantized or entropy encoded valid data in the encoded frame to obtain the shifted valid data for each sample is specifically: The quantized or entropy encoded low bit data in the encoded frame is decoded according to the low bit data entropy encoded or quantized encoded information of each sample to obtain low bit information of each sample. The specific decoding manner for decoding the quantized or entropy encoded low-bit data in the encoded frame is determined according to the encoding information, that is, the entropy encoding or the quantized low-bit encoding by using the encoding method and the corresponding decoding manner. The bit data is decoded.

Further, in the embodiment of the present invention, the step of decoding the valid data after the quantization coding or the entropy coding in the coded frame, and obtaining the valid data after the shift of each sample, may specifically include:

1) Entropy decoding the effective data after entropy coding in the encoded frame to obtain low bit data of each sample.

2) Perform inverse quantization processing on the quantized and validated data in the encoded frame to obtain the shifted valid data of each sample, thereby recovering the shifted valid data with distortion. For a more detailed understanding of the data encoding method provided in Embodiment 3 of the present invention and the data decoding method provided in Embodiment 4, a specific application scenario of the embodiment of the present invention is further given below.

Referring to FIG. 14, FIG. 14 is a schematic diagram of dividing a data block in an improved data encoding method according to Embodiment 3 of the present invention.

The encoder encodes the original data in units of data blocks. According to the distribution characteristics of the original data, the S region and the D region of each sample in the data block have randomness. The encoder divides and shifts the data block to obtain the sign bit, the left shift number, and the shifted valid data, each of which occupies n3, L, and n4 bits, respectively. Where n3 includes the number of bits of the sign bit and the sign extension bit, and the sign bit indicates The positive and negative polarity of the point, the specific number of bits of the left shift number L can be determined according to the number of bits of the sample, for example, the number of bits of the sample is 32, then the number of left shifts L can be represented by 5 bits, the sample point The number of bits is 16, and the number of left shifts L can be represented by 4 bits. The number of bits of the sample is 8, and the number of left shifts L can be represented by 3 bits.

When the encoder divides the data block, it first extracts the sign bit Ss of each sample point in the data block, and obtains the positive and negative polarities of each sample according to the sign bit. The encoder can take an absolute value for each sample, obtain the sample with the largest absolute value among the L samples, and obtain the most significant bit 卩 of the sample. After determining the most significant bit P of the sample with the largest absolute value, the next most significant bit di of each sample is sequentially determined, and each sample is shifted to the left by the P-di bit, that is, the effective after the shift is obtained. data. Where i takes the value of 11丄.

In the embodiment of the present invention, the state 0 and 1 can be used to indicate the positive and negative polarity of each sample in each frame of data, for example, 0 indicates that the sample is negative, 1 indicates that the sample is positive; or 0 indicates a sample value. Positive, and 1 means the sample value is negative. After the sign bit of each sample is processed, a binary code word S _s c carrying the positive and negative polarity codes of each sample is obtained.

Figure 6 shows an example of the probability distribution of the original data. The probability distribution of the corresponding left shift number is given by Figure 15. It can be seen that the probability of occurrence of the left shift number is unequal, which provides a basis for data compression using entropy coding. Therefore, the embodiment of the present invention uses the entropy coding method to perform distortion-free entropy on the left shift number. Encoding gives the shift information code S _m c. Entropy coding may employ Huffman coding, arithmetic coding or other unequal length coding algorithms, or a combination thereof. In the embodiment of the present invention, the encoding information such as the encoding method and the encoding parameter for entropy encoding the left shift number may be stored in the frame header information, and the frame header information is stored in the encoded frame for output. There are two methods of distortion coding (quantization coding). In the distortionless coding mode, the encoder can obtain the effective number of bits of valid data and directly output the corresponding number of bits to achieve distortion-free coding. In the distortion coding mode, in order to keep the encoder low algorithm complexity, a simple scalar quantization method can be used. In practical applications, the effective data is non-uniformly distributed over the entire dynamic range, but distributed evenly within the local dynamic range. The probability distribution of the valid data of the original data S is given by Fig. 16. It can be seen that the effective data is non-uniformly distributed as a whole, and the dynamic range is large. In the case where the most significant bit in the data block is known, the valid data can be classified, Figure 17 and Figure 18, respectively. The probability distribution of the effective data with the most significant bits P of 7 and 8 is given. It can be seen that the distribution is relatively uniform and the dynamic range is small. If the two parts of data are separately quantized, the quantization is more accurate. Therefore, the compression coding algorithm for effective data can use the classified quantization coding algorithm, for example, the Lloyd-Max non-uniform scalar quantization algorithm. The coding information such as the coding mode and the coding parameter for entropy coding or quantization coding of the valid data may be stored in the frame header information, and the frame header information is output through the coding frame.

In the embodiment of the present invention, the encoder combines the encoded positive and negative polarity coded word S _s c , the shift information code S _m c and the shifted effective data code S _e c , and combines the frame header information. , to form a complete coded frame, Figure 19 shows the structure of the coded frame in the coded bitstream. The different coded frames are arranged in chronological order of the output to form an encoded bit stream B1 output by the encoder, as shown in FIG. The frame header information may further include the number of samples in the data block, the most significant bit in the data block, and the like. It should be noted that, if the encoder and the decoder use the preset decoding type and the encoding type, and the length of each sample in the data block and the number of samples are preset values, the encoded frame may be No need to carry frame header information.

At the decoder, the encoded bit stream is decoded in units of coded frames, the decoder can read the frame header information and save, and then the positive and negative polarities of each sample are determined based on the 0, 1 state of the binary bit stream. Secondly, the decoder performs distortion-free entropy decoding to recover the left shift number of the valid data and saves it. The specific decoding method used for entropy decoding corresponds to the encoding mode used by the encoder.

In the case where the encoder obtains the entropy encoding of the valid data according to the frame header information, the decoder determines the effective bit number of the valid data by the most significant bit of the current frame and the shift information, and sequentially reads the corresponding quantity from the input bit stream. The bits are entropy decoded to achieve distortion-free decoding of the valid data. When there is distortion coding, the decoder performs inverse quantization processing on the encoded valid data, and the specific process includes: recovering the quantized subscript value of the valid data from the input bit stream, and then determining the scalar quantization code table according to the most significant bit of the current frame. Performing an inverse quantization process, searching for a corresponding quantized value in the quantization table by using a subscript index, and implementing distortion decoding of the valid data. Finally, the decoder assembles the positive and negative polarities and the shifted valid data of each sample obtained by decoding, obtains the original data block and outputs, completes decoding of one encoded frame, and prepares to decode the next frame. Until the end of the decoding. A device corresponding to an embodiment of the method of the present invention. Embodiment 5

Referring to FIG. 20, FIG. 20 is a schematic structural diagram of an encoder according to Embodiment 5 of the present invention.

The encoder provided in Embodiment 13 of the present invention includes:

a first pre-processing unit 1101, configured to obtain positive and negative polarity of each sample in the data block, and a most significant bit of the sample with the largest absolute value among the samples of the data block;

a high-bit coding unit 1102, configured to acquire n-bit high-bit data from each of the most significant bits in each sample, and perform first-class entropy coding on each of the high-bit data of each sample. The number of bits of valid data where n is greater than 1 and less than the absolute maximum value;

a low-bit coding unit 1103, configured to acquire low-bit data other than the high-bit data in data starting from the most significant bit in each sample, and low bits for each sample The data is respectively subjected to the first type of quantization coding or the second type of entropy coding;

a first code output unit 1104, configured to output a codeword carrying positive and negative polarities of each sample, a first type of entropy coded high bit data of each sample, and a first type of quantized code or a second class Encoded frame composed of entropy encoded low bit data.

The encoder provided in the fifth embodiment of the present invention can be used in the data encoding method provided in the foregoing corresponding embodiment. For the detailed execution process, refer to the foregoing method embodiment, and the description is not repeated here.

The encoder provided in Embodiment 5 of the present invention further includes:

a first polarity encoding unit 1105, configured to respectively perform a third type of entropy encoding on the positive and negative polarities of each sample;

The code words carrying the positive and negative polarities of each sample in the coded frame output by the first code output unit 1104 are specifically the third type of entropy coded positive and negative polarities of each sample.

Referring to FIG. 21, FIG. 21 is a schematic block diagram of an encoder according to Embodiment 5 of the present invention.

As shown, a frame of raw data S _{m is} input to the first pre-processing module 1201 for data analysis and processing. The pre-processing includes extracting the sign bit, detecting the most significant bit of one frame, separating the high bit data and the low bit. data. The basic idea of the first pre-processing module 1201 is that the signal distribution to be encoded has randomness, and the statistical distribution of the high-bit area and the low-bit area in the effective bit area is different, and different encoding methods can be used for processing. Among them, the high bit data has a non-homogeneous distribution characteristic, and the low bit data is distributed in a non-homogeneous distribution within the entire dynamic range, but distributed uniformly in the local dynamic range, which is low bit data. The compression coding process provides the basis. The symbol bit data S _s , the high bit data S _h and the low bit data obtained by the first preprocessing module 1201 are respectively sent to the sign bit encoding module 1202, the high bit encoding module 1203 and the low bit encoding module 1204, The sign bit code S _s c , the high bit code S _h c and the low bit code S c . The Sign Bit Encoding Module 1202 uses the states 0, 1 to indicate the positive and negative polarity of each sample, and the final encoding generates a set of binary bit streams. According to the uneven distribution characteristic of the high bit data, the high bit coding module 1203 uses the distortion-free entropy coding. Entropy coding may employ Huffman coding, arithmetic coding, or a combination thereof. According to the distribution characteristics of the low bit data, the low bit coding module 1204 classifies it, and the classification process more closely reflects the distribution characteristics of each frame of data. In the distortionless coding state, the encoder determines the effective number of bits of the low bit data according to the most significant bit of the current frame, and sequentially writes the corresponding number of bits into the output buffer to implement distortionless coding output. In the distortion coding state, the low bit coding module 1204 classifies the low bit data according to the most significant bit of the current frame, and selects a corresponding quantization table to quantize the low bit data to implement distortion compression coding.

Finally, the first frame encapsulating module 1205 combines the sign bit code S _s c , the high bit code S _h e and the low bit code S _lC , and combines the frame header information H _{l to} obtain a complete coded frame output to form an encoded bit stream. :^. Referring to FIG. 22, FIG. 22 is a schematic structural diagram of a decoder according to Embodiment 6 of the present invention.

The decoder provided in Embodiment 6 of the present invention includes:

The first polarity acquiring unit 1301 is configured to acquire a codeword carrying positive and negative polarities of each sample in the received encoded frame, and obtain positive and negative polarities of each sample according to the codeword;

The high bit decoding unit 1302 is configured to decode the entropy encoded high bit data in the encoded frame to obtain high bit data of each sample;

a low bit decoding unit 1303, configured to decode the quantized or entropy encoded low bit data in the encoded frame to obtain low bit data of each sample;

The first data assembling unit 1304 is configured to assemble the positive and negative polarities, the high bit data, and the low bit data of each sample according to the most significant bits included in the received encoded frame to obtain the original data block. The decoder provided in the sixth embodiment of the present invention may be used in the data decoding method provided in the foregoing corresponding embodiment 2. For the detailed process, refer to the foregoing method embodiment, and the description is not repeated here.

Referring to FIG. 23 and FIG. 23, FIG. 23 is a schematic block diagram of a decoder provided in Embodiment 6 of the present invention.

If shown, after receiving the encoded bit stream, the first frame deblocking module 1401 decomposes the information in the encoded bit stream into frame header information, symbol bit code S _s c , high bit code S _h e and low. Bit code Sic

Combining the frame header information H _l sign bit code S _s c is decoded by the sign bit decoding module 1402 to obtain the sign bit S _s , and the sign bit decoding module 1402 determines the current frame by determining the 0, 1 state of each bit in the input bit stream. Positive and negative polarity of samples.

The high bit code S _h e is decoded by the high bit decoding module 1403 to obtain high bit data S _h corresponding to the encoding end, and the high bit decoding module 1403 performs entropy decoding processing.

The low bit code S _lC is decoded by the low bit decoding module 1404 to obtain low bit data. In the distortionless decoding state, the low bit decoding module determines the effective bit number of each low bit data according to the frame header information, and The corresponding number of bits are read from the input bit stream for distortion-free decoding. In the distortion-decoded state, the low-bit decoding module 1404 classifies the current frame according to the frame header information, and performs inverse quantization processing by using a corresponding method. It recovers low-bit data with distortion and realizes distortion decoding.

Finally, in combination with the frame header information, the first inverse pre-processing module 1405 analyzes the symbol bit data S _s , the high bit data S _h and the low bit data to determine the symbol and sign extension bit of the original frame of one frame, and the high bit. The bit and the low bit, the reconstructed data Sr is obtained and output. Referring to FIG. 24, FIG. 24 is a schematic structural diagram of an encoder according to Embodiment 7 of the present invention.

The encoder provided in Embodiment 7 of the present invention includes:

a second pre-processing unit 1501, configured to obtain positive and negative polarities of each sample in the data block, and a most significant bit p of the sample with the largest absolute value among the samples of the data block;

The shifting unit 1502 is configured to obtain the most significant bit d of each sample, and shift the effective data bits of each sample to the left by p-d bits;

a shift information encoding unit 1503, configured to perform a fourth type of entropy encoding on each of the left shift numbers pd of each sample; The effective data encoding unit 1504 is configured to respectively perform a fifth type of entropy coding or a second type of quantization coding on the shifted valid data bits in each sample;

The second code output unit 1505 is configured to output the encoded code frame, where the coded frame includes the most significant bit P, the codeword carrying the positive and negative polarity of each sample, and the fourth type of entropy coded left shift A coded frame consisting of a number and a fifth type of entropy coding or a second type of quantized coded valid data.

The encoder provided in the seventh embodiment of the present invention can be used in the data encoding method provided in the foregoing corresponding third embodiment. For the detailed process, refer to the foregoing method embodiment, and the description is not repeated here.

Further, the encoder provided by the embodiment of the present invention further includes:

a second polarity encoding unit 1506, configured to perform a sixth type of entropy coding on the positive and negative polarities of each sample respectively;

The code words carrying the positive and negative polarities of each sample in the coded frame output by the second code output unit 1505 are the sixth type of entropy coded positive and negative polarities of each sample.

Referring to FIG. 25 and FIG. 25, FIG. 25 is a schematic block diagram of an encoder according to Embodiment 7 of the present invention.

As shown, a frame of raw data S _{m is} input to a second pre-processing module 1601 for data analysis and processing. The pre-processing includes extracting sign bits, detecting the most significant bits of a frame, and detecting the most significant bits of each sample. Bit, shifts each sample and saves the shift information and the shifted valid data. The basic idea of the second pre-processing module 1601 is that the signal distribution to be encoded has randomness, and the newly generated shift information and the new data generated after shifting respectively reflect the distribution law of the original data, and different encodings can be used. The method is processed. Among them, the shift information data has a non-uniform hook distribution characteristic, and the distribution characteristics of the shifted effective data are: a non-uniform hook distribution within the entire dynamic range, but a uniform distribution within the local dynamic range, which is effective The compression encoding process of the data provides a basis.

The symbol bit data S _s , the left shift number S _m and the valid data S _e obtained by the second pre-processing module 1601 are sent to the sign bit encoding module 1602, the shift information encoding module 1603 and the valid data encoding module 1604, respectively, to obtain symbols. The bit code S _s c, the shift information code S _m c and the shifted valid data code S _e c. The Sign Bit Encoding Module 1602 uses states 0, 1 to indicate the positive and negative polarity of each sample, and the final encoding generates a set of binary bit streams. According to the uneven distribution characteristic of the left shift number, the shift information encoding module 1603 employs distortion-free entropy coding. Entropy coding may employ Huffman coding, arithmetic coding, or a combination thereof. According to the distribution characteristics of the effective data, the effective data encoding module 1604 classifies the data, and the classification processing more finely reflects the distribution characteristics of each frame of data. In the form of distortionless coding In the state, the encoder determines the effective number of valid data according to the most significant bit of the current frame and the shift information, and sequentially writes the corresponding number of bits into the output buffer to implement distortionless coding output. In the distortion coding state, the effective data encoding module 1604 classifies the valid data according to the most significant bit of the current frame, and selects a corresponding quantization table to quantize and encode the valid data to implement distortion compression coding. Referring to FIG. 26, FIG. 26 is a schematic structural diagram of a decoder according to Embodiment 8 of the present invention.

The decoder provided in Embodiment 8 of the present invention includes:

The second polarity acquiring unit 1701 is configured to obtain a codeword carrying positive and negative polarities of each sample in the encoded frame, and obtain positive and negative polarities of each sample according to the codeword;

The left shift number decoding unit 1702 is configured to decode the entropy encoded left shift number in the encoded frame to obtain a left shift number of valid data of each sample point;

The effective data decoding unit 1703 is configured to decode the quantized or entropy encoded valid data in the encoded frame to obtain the shifted valid data of each sample;

The second data assembling unit 1704 is configured to: positive and negative polarity of each sample, left shift number of valid data, and effective after shifting according to the most significant bit of the sample with the largest absolute value included in the encoded frame. The data is assembled to obtain the original data block.

The decoder provided in Embodiment 8 of the present invention may be used in the data decoding method provided in the foregoing corresponding Embodiment 4. For the detailed process, refer to the foregoing method embodiment, and the description is not repeated herein.

Referring to FIG. 27, FIG. 27 is a schematic block diagram of a decoder according to Embodiment 8 of the present invention.

As shown, after receiving the encoded bit stream, the second frame deblocking module 1801 decomposes the information in the encoded bit stream into frame header information H ₁ sign bit code S _s c , and shift information code S _m c And valid data encoding S _e c.

Combining the frame header information H _l sign bit code S _s c is decoded by the sign bit decoding module 1802 to obtain the sign bit S _s , and the sign bit decoding module 1802 determines the current frame by determining the 0, 1 state of each bit in the input bit stream. Positive and negative polarity of samples.

The shift information encoding S _m c is decoded by the shift information decoding module 1803 to obtain a left shift number S _m corresponding to the encoding end, and the shift information decoding module 1803 performs entropy decoding processing. The valid data encoding S _e c is decoded by the valid data decoding module 1804 to obtain valid data S _e . In the distortion-free decoding state, according to the frame header information and the shift information, the effective data decoding module 1804 determines the effective number of bits of each valid data sample, and reads a corresponding number of bits from the input bit stream to perform distortion-free. decoding. In the distortion decoding state, the effective data decoding module 1804 classifies the current frame according to the frame header information, performs inverse quantization processing by using a corresponding method, recovers valid data with distortion, and implements distortion decoding.

Finally, in combination with the frame header information, the second inverse pre-processing module 1805 analyzes the symbol bit data S _s , the shift information data S _m and the valid data S _e to determine a symbol and a sign extension bit of the original signal of one frame, The information and the significant bits are shifted, and the reconstructed data Sr is obtained and output. Referring to FIG. 28 and FIG. 29, FIG. 28 and FIG. 29 are respectively schematic diagrams of a data transmission system to which a data encoding method and a data decoding method according to an embodiment of the present invention are applicable.

The data encoding method provided in Embodiments 1 and 3 of the present invention can be used in the encoder in the figure. The data decoding method provided in Embodiments 2 and 4 of the present invention can be used in the decoder in the figure, and the decoding used in the decoder. The method corresponds to the encoding method.

The encoder and the decoder provided by the embodiments of the present invention can be applied to remote data communication (such as data communication, file transmission, voice communication, audio communication, etc.) of various channels and networks, short-range and data communication within the system (for example, general-purpose Data transmission between the IQ data in the public wireless interface, remote access, etc.), in-device or on-board modules (such as data exchange between analog-to-digital/digital-to-analog converters and other components).

In addition, the encoder and the decoder provided by the embodiments of the present invention can also be applied to a system for storage applications, such as efficient storage of data, digital media storage, digital library, and the like.

It should be noted that the information interaction, the execution process, and the like between the units in the encoder and the decoder are based on the same concept as the method embodiment of the present invention. For details, refer to the description in the method embodiment of the present invention. I won't go into details here.

A person skilled in the art can understand that all or part of the process of implementing the above embodiments can be completed by a computer program to instruct related hardware, and the program can be stored in a computer readable storage medium. The flow of an embodiment of the methods as described above may be included. The storage medium may be a magnetic disk, an optical disk, or a read-only memory (ROM). Or random access to Random Access Memory (RAM).

The data encoding method, the data decoding method, the encoder and the decoder of the present invention are described in detail above. For those skilled in the art, according to the idea of the embodiment of the present invention, the specific implementation manner and the application range may be changed. The contents of this specification are not to be construed as limiting the invention.

Claims

Rights request

A data encoding method, comprising:

Obtaining the positive and negative polarity of each sample in the data block;

Obtaining a most significant bit of the sample having the largest absolute value among the samples of the data block; acquiring n-bit high-bit data from the most significant bit in each sample of the data block, for each The high bit data of each sample is respectively subjected to a first type of entropy coding, wherein n is greater than 1, and is less than the number of bits of valid data of the sample with the largest absolute value;

Obtaining low-bit data other than the high-bit data in the data starting from the most significant bit in each sample of the data block, respectively performing low-bit data for each sample a type of quantization coding or a second type of entropy coding;

Outputting the encoded coded frame, wherein the coded frame includes the most significant bit, a codeword carrying positive and negative polarities of each sample of the data block, and a sample of each sample of the data block A class of entropy encoded high bit data and a first type of quantized code or a second type of entropy encoded low bit data.

The method according to claim 1, wherein the method further comprises: performing a third type of entropy coding on the positive and negative polarities of each sample of the data block;

The codeword carrying the positive and negative polarities of each sample is the third type of entropy coded positive and negative polarity of each sample of the data block.

3. A method according to claim 1 or 2, characterized in that

The encoded frame further includes high-bit data of each sample of the data block, first-class entropy encoded coding information, and low-bit data second-class entropy coding or first-class quantized coding information.

4. A data decoding method, comprising:

Decoding the quantized or entropy encoded low bit data in the encoded frame to obtain low bit data of each sample; Performing positive and negative polarity of each sample, high bit data of each sample, and low bit data of each sample according to the most significant bit included in the received encoded frame Assemble to get the original data block.

5. The method according to claim 4, wherein the obtaining positive and negative polarities of each sample according to the codeword comprises:

The code words carrying the positive and negative polarities of each sample in the encoded frame are separately entropy decoded to obtain the positive and negative polarities of each sample.

6. A method according to claim 4 or 5, characterized in that

The coded frame further includes high bit data entropy coded coding information and low bit data entropy coded or quantized coded information of each sample;

Decoding the high-bit data of the entropy-encoded data in the encoded frame to obtain high-specificity data of each sample, specifically including:

Entropy decoding the entropy-encoded high-bit data in the encoded frame according to the high-bit data entropy-encoded coding information of each sample to obtain high-bit data of each sample;

Decoding the low-bit data of the quantized or entropy-encoded data in the encoded frame to obtain the low-bit data of each sample, specifically including:

The quantized or entropy encoded low bit data in the encoded frame is decoded according to the low bit data entropy encoded or quantized encoded information of each sample to obtain low bit information for each sample.

The method according to claim 4 or 5, wherein the decoding the low-bit data of the quantized or entropy-encoded data in the encoded frame to obtain the low-bit data of each sample includes:

Entropy decoding the entropy-encoded low-bit data in the encoded frame to obtain low-specificity data of each sample; or

The quantized low-bit data in the encoded frame is subjected to inverse quantization processing to obtain low-bit data of each sample.

8. A data encoding method, comprising:

Obtaining the positive and negative polarity of each sample in the data block;

Obtaining the most significant bit p of the sample with the largest absolute value among the samples of the data block; Obtaining the most significant bit d of each sample of the data block, and shifting the valid data bits of each sample to the left by a pd bit;

Performing a fourth type of entropy coding on the left shift number pd of each sample of the data block; performing a fifth type of entropy coding on the shifted valid data bits in each sample of the data block The second type of quantization code;

And outputting the encoded coded frame, where the coded frame includes the most significant bit P, a positive and negative polarity codeword carrying each sample of the data block, and each sample of the data block The fourth type of entropy-coded left shift number and the fifth type of entropy coding or the second type of quantized coded valid data.

9. The method according to claim 8, further comprising:

Performing a sixth type of entropy coding on the positive and negative polarities of each sample of the data block;

The positive and negative polarity codewords of each sample carrying the data block are the sixth type of entropy encoded positive and negative polarities of each sample of the data block.

10. A method according to claim 8 or 9, characterized in that

The coded frame further includes a left shift number of each sample of the data block, a fourth type of entropy coded coding information, and a shifted effective data, a second type of quantized code or a fifth type of entropy coded coded information. .

11. A data decoding method, comprising:

Obtaining a codeword carrying positive and negative polarities of each sample in the encoded frame, and obtaining positive and negative polarities of each sample according to the codeword;

Decoding an entropy-encoded left shift number in the encoded frame to obtain a left shift number of valid data of each sample point;

The method according to claim 11, wherein the obtaining positive and negative polarities of each sample according to the codeword is specifically:

The code words of the positive and negative polarities of each sample of the encoded frame are separately entropy decoded to obtain the positive and negative polarities of each sample.

13. Method according to claim 11 or 12, characterized in that

The coded frame further includes code information of left shift number entropy coding of each sample and coded information of the shifted limited data entropy coding or quantization coding;

Decoding the left shift number after the entropy coding in the encoded frame, and obtaining the left shift number of the valid data of each sample, specifically:

And entropy decoding the entropy-encoded left shift number in the encoded frame according to the left shift number entropy coded coding information of each sample of the encoded frame, to obtain a left shift number of valid data of each sample point;

Decoding the quantized or entropy encoded valid data in the encoded frame to obtain the shifted valid data of each sample, specifically:

Decoding the quantized or entropy encoded valid data in the encoded frame according to the entropy encoded or quantized encoded information of the shifted valid data of each sample in the encoded frame to obtain a shift of each sample After valid data.

The method according to claim 11 or 12, wherein the decoding of the quantized or entropy encoded valid data in the encoded frame to obtain the shifted valid data of each sample comprises:

Entropy decoding the entropy-encoded valid data in the encoded frame to obtain the shifted valid data of each sample; or

The quantized and validated data in the encoded frame is inverse quantized to obtain the shifted valid data for each sample.

15. An encoder, comprising:

a high-bit coding unit, configured to acquire n-bit high-bit data from the most significant bit in each sample in the data block, and perform a first class on each of the high-bit data of each sample Entropy coding, wherein n is greater than 1, and is less than the number of bits of valid data of the sample having the largest absolute value;

a low bit coding unit, configured to acquire low bit data other than the high bit data in data starting from the most significant bit in each sample in the data block, for each sample point The low bit data is respectively subjected to the first type of quantization coding or the second type of entropy coding; a first code output unit, configured to output an encoded code frame, where the coded frame includes the most significant bit, a codeword carrying positive and negative polarity of each sample of the data block, and the data A first type of entropy encoded high bit data of each sample of the block and an encoded frame of the first type of quantized code or the second type of entropy encoded low bit data.

The encoder according to claim 15, further comprising:

a first polarity encoding unit, configured to perform a third type of entropy encoding on the positive and negative polarities of each sample of the data block;

The code words of the positive and negative polarities of each sample carrying the data block in the encoded frame output by the first code output unit are the third type of entropy coded positive of each sample of the data block Negative polarity.

17. A decoder, comprising:

a first data assembling unit, configured to assemble the positive and negative polarities of each sample, the high bit data of each sample, and the low bit data according to the most significant bit included in the received encoded frame to obtain the original data block.

18. An encoder, comprising:

a shifting unit, configured to acquire a most significant bit d of each sample in the data block, and shift a valid data bit of each sample to a left by a p-d bit;

An effective data encoding unit, configured to respectively perform a fifth type of entropy coding or a second type of quantization coding on the shifted valid data bits in each sample in the data block; a second code output unit, configured to output the encoded coded frame, where the coded frame includes the most significant bit P, a codeword carrying positive and negative polarities of each sample of the data block, The fourth type of entropy-coded left shift number of each sample of the data block and the fifth type of entropy coding or the second type of quantized coded valid data.

The encoder according to claim 18, further comprising:

a second polarity encoding unit, configured to perform a sixth type of entropy coding on the positive and negative polarities of each sample of the data block;

The code words of the positive and negative polarities of each sample carrying the data block in the encoded frame output by the second coding output unit are the sixth type of entropy coded positive of each sample of the data block Negative polarity.

20. A decoder, comprising:

a second data assembling unit, configured to: positive and negative polarity of each sample, left shift number of valid data of each sample, and shift according to a most significant bit of a sample having the largest absolute value included in the encoded frame After