CN107437976B

CN107437976B - Data processing method and equipment

Info

Publication number: CN107437976B
Application number: CN201610356216.0A
Authority: CN
Inventors: 牛凯; 杨芳僚
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2016-05-25
Filing date: 2016-05-25
Publication date: 2021-02-12
Anticipated expiration: 2036-05-25
Also published as: CN107437976A

Abstract

The embodiment of the invention discloses a data processing method and equipment, wherein the method comprises the following steps: determining coding parameters based on the partial polarization transformation; and calculating multilayer bit soft information of the source symbol sequence by using the coding parameters, and performing serial cancellation coding based on partial polarization transformation on the multilayer bit soft information to obtain a first bit sequence. The embodiment of the invention can reduce the coding complexity.

Description

Data processing method and equipment

Technical Field

The present invention relates to the field of communications technologies, and in particular, to a data processing method and device.

Background

The source coding is an important technology in the communication field, and the redundancy of a source can be removed by adopting the source coding technology, so that data can be transmitted efficiently as much as possible, and the effectiveness of a communication system is improved.

At present, a polarization coding quantization scheme for applying a polarization code to source coding exists in the field of communication scene source compression, and the polarization coding quantization scheme has excellent compression performance. However, the polar coding quantization scheme adopts a serial cancellation coding algorithm, and the polar coding quantizer comprises p polar code serial cancellation coders, and each polar code serial cancellation coder needs to perform Nlog₂N times of bit likelihood ratio calculation of coding node, so that pNlog needs to be performed in total₂Bit likelihood ratio calculation of N coding nodes, i.e. the complexity of coding is O (pNlog)₂N), where p > q, q is the coding rate, and N is the length of the source symbol sequence.

Therefore, the coding complexity is in direct proportion increase relation with the length of the information source symbol sequence and the coding code rate, and the coding complexity is high.

Disclosure of Invention

The embodiment of the invention provides a data processing method and equipment, which can reduce the encoding complexity.

The first aspect of the embodiments of the present invention discloses a data processing method, including:

and determining coding parameters based on partial polarization transformation, wherein the coding parameters comprise system parameters of a polarization coding quantizer (such as the number of the polarization code serial cancellation coders, the coding rate of the source symbol sequence and the code length of each polarization code serial cancellation coder), a mapping rule between symbols and bits and a polarization code coding structure.

And calculating multilayer bit soft information of the source symbol sequence by using the coding parameters, and performing serial cancellation coding based on partial polarization transformation on the multilayer bit soft information to obtain a first bit sequence. The method comprises the steps that bit soft information is used for representing probability information that each bit forming an information source symbol takes a value of 0 or 1, serial offset coding is a serial coding method for calculating current coding bits by using coded bits, and polarization transformation is an operation of combining and dividing a plurality of parallel independent channels to generate a plurality of subchannels with correlation and capacity difference.

Therefore, when the polarization coding quantizer performs the polarization code serial cancellation coding, only part of polarization coding processing needs to be performed on the source symbol sequence, so that the coding complexity can be reduced.

In a possible implementation, the data processing method is applied to a polar coding quantizer, the polar coding quantizer includes p partial polar transform-based polar code serial cancellation encoders, and the determining the partial polar transform-based coding parameters includes:

setting system parameters of the polar encoding quantizer;

constructing N virtual experimental channels according to the system parameters, wherein N is the length of the information source symbol sequence;

determining an input value set of the virtual experimental channel

And setting a mapping rule between symbols and bits

Wherein x is_iIn order to reconstruct the quantized symbols,

to form the p bits of the reconstructed quantized symbol,

{0,1}^pp cartesian products representing {0,1 };

and constructing a polarization code coding structure based on partial polarization transformation.

Therein, an offline phase and an online phase may be defined. The off-line stage is mainly used for setting parameters of the polarization coding quantizer, and the on-line stage is mainly used for processing the information source symbol sequence in real time. In the off-line phase, the system parameters may include the number of the polar code serial cancellation encoders, the coding rate of the source symbol sequence, and the code length of each polar code serial cancellation encoder. Setting the encoding parameters in the off-line phase can prepare for real-time processing of the source symbol sequence in the on-line phase, which can significantly reduce the processing delay of the whole process.

In a possible implementation, each of the virtual experimental channels includes p bit sub-channels, the system parameter includes a coding rate q of the source symbol sequence, and the manner of constructing the partial polarization transform-based polarization code coding structure is specifically:

calculating a reliability metric of each first bit sub-channel, wherein the first bit sub-channel is obtained after all the bit sub-channels are subjected to polarization transformation;

determining a polarization transformation stop threshold value of partial polarization coding according to the reliability metric values of a plurality of first bit sub-channels;

setting the flag bit of the coding node which needs to be subjected to polarization transformation as a first flag value and setting the flag bit of the coding node which does not need to be subjected to polarization transformation as a second flag value according to the polarization transformation stop threshold value, wherein the first flag value is different from the second flag value;

calculating the reliability metric of each second bit sub-channel according to the polarization transformation stopping threshold value, the zone bit of the coding node and a preset reliability metric algorithm, wherein the second bit sub-channel is obtained after the bit sub-channel is subjected to partial polarization transformation;

selecting qN target bit sub-channels with larger reliability metric values of the second bit sub-channels from the pN second bit sub-channels according to the sequence of the reliability metric values of the second bit sub-channels from large to small, and forming a reliable bit mark set by the subscripts of the qN target bit sub-channels;

and determining the flag bits of all the coding nodes and the reliable bit flag set as a partial polarization transformation-based polarization code coding structure.

The coding nodes which do not need to be subjected to the polarization transformation can be determined according to the polarization transformation stopping threshold value, and in the subsequent online stage, the bit likelihood ratio does not need to be calculated, and only simple iteration is needed. In addition, the selected reliable bit flag set can be used for distinguishing which bits are most important in encoding, so that the purpose of compressing the encoding length can be achieved by only outputting the part of bits.

In one possible embodiment, the virtual test channel comprises a transfer function p (y)_j| x), wherein,

said x passes through

And the above-mentioned

One-to-one correspondence, j ═ 1,2,. N };

the calculating of multi-layer bit soft information on the source symbol sequence by using the coding parameters and the serial cancellation coding based on partial polarization transformation on the multi-layer bit soft information to obtain a first bit sequence comprises:

using said p (y) in calculating the bit soft information of the i-th layer of the source symbol sequence_j| x) and the

And according to the formula

Calculate the firstBit soft information L of ith bit in j symbols_i(j) And combining said L_i(j) Bit likelihood ratio as a first stage of a coding node in an ith said polar code successive cancellation coder

Wherein, Pr { b_i1 is the probability that the ith bit in the jth symbol takes a value of 1,

is a bit sequence (b) composed of ith to pth bits in the jth symbol_i,...,b_p)，i＝{1,2,...p}；

J (th) coded bit u at i (th) layer_jIf j is an odd number, let j equal to 2z-1, according to the formula

Calculating the u_jBit likelihood ratio of

If j is an even number, let j equal 2z, according to the formula

Calculating the u_jIs/are as follows

Calculating the u_jIs/are as follows

If j is an even number, let j equal 2z, according to the formula

Calculating the u_jIs/are as follows

Wherein z is a positive integer, n is the number of stages of the coding node, and n-1 represents the previous stage of the nth stage of the coding node;

obtained by calculation for each layer

If it is

If the value is larger than the preset threshold value, determining u_jThe value of the bit is a first bit value; if it is

Less than the preset threshold value, determining u_jThe first bit value is different from the second bit value; wherein,

representing the source symbol sequence, wherein N is the length of the source symbol sequence;

all the calculated u_jIs determined as the first bit sequence.

The coding node with the flag bit as the first flag value needs to calculate the bit likelihood ratio according to a formula, the coding node with the flag bit as the second flag value does not need to calculate the bit likelihood ratio, and the coding complexity is measured by the number of the coding nodes needing to be calculated, so that the coding complexity can be reduced.

In one possible embodiment, the method further comprises:

and sending the first bit sequence and the coding parameters to receiving end equipment, wherein the coding parameters are used for decoding the first bit sequence by the receiving end equipment so as to recover the information source symbol sequence.

Wherein the encoding parameters include mapping rules between symbols and bits

Flag bits for all coding nodes, and a set of reliable bit flags.

Optionally, the bit sequence and the coding parameters may be sent to the receiving end each time, or the bit sequence and the coding parameters may be sent to the receiving end for the first time, and then the coding parameters do not need to be sent only by sending the bit sequence.

The second aspect of the embodiments of the present invention discloses a data processing method, including:

receiving a first bit sequence and a coding parameter; wherein the encoding parameters include mapping rules between symbols and bits

Flag bits for all coding nodes, and a set of reliable bit flags.

Performing serial-parallel conversion processing on the first bit sequence to obtain a multilayer conversion bit sequence;

using the coding parameters to perform partial polarization decoding processing on the transformation bit sequence of each layer to obtain a second bit sequence; the coding parameters comprise a flag bit of a coding node and a reliable bit flag set, the flag bit is a first flag value or a second flag value, the first flag value is used for the coding node to need polarization transformation, and the second flag value is used for the coding node to not need polarization transformation.

And carrying out bit-to-symbol mapping processing on the second bit sequences obtained by all the layers by using the coding parameters so as to recover the source symbol sequence. Wherein the encoding parameter is a mapping rule between symbols and bits

Therefore, the polarization decoding quantizer only performs partial polarization decoding processing on the received first bit sequence, so that the complexity of decoding can be reduced.

In a possible implementation, the coding parameters include a flag bit of a coding node and a reliable bit flag set, the flag bit includes a first flag value or a second flag value, the first flag value is used to indicate that the coding node needs to perform polarization transformation, and the second flag value is used to indicate that the coding node does not need to perform polarization transformation;

the performing, by using the encoding parameter, partial polarization decoding processing on the transform bit sequence of each layer to obtain a second bit sequence includes:

determining the zone bit of the coding node as the zone bit of a decoding node;

j (th) decoded bit at i (th) layer

If j is an odd number, let j equal to 2z-1, according to the formula

Calculating the last level of the jth decoding node

The value of (d); if j is an even number, let j equal 2z, according to the formula

Calculating the last level of the jth decoding node

The numerical values of (a), wherein,

the operation of modulo two addition is represented,

is the bit value of the first bit sequence;

j (th) decoded bit at i (th) layer

If j is an odd number, let j equal to 2z-1, according to the formula

Calculating the last level of the jth decoding node

Calculating the last level of the jth decoding node

The value of (d); wherein z is a positive integer, n is the number of stages of the decoding node, and n-1 represents the previous stage of the nth stage of the decoding node;

all the obtained N numbers are calculated

Is determined as the second bit sequence of the ith layer, where N is the length of the second bit sequence of each layer.

In a third aspect of the embodiments of the present invention, a polar coding quantizer is disclosed, which includes functional units for performing some or all of the steps of any of the methods of the first aspect of the embodiments of the present invention. Wherein the polar encoding quantizer reduces the complexity of encoding when performing part or all of the steps of any of the methods of the first aspect.

A fourth aspect of the present embodiments discloses a polar decoding quantizer comprising functional units for performing some or all of the steps of any of the methods of the second aspect of the present embodiments. Wherein the polar decoding quantizer reduces the complexity of decoding when performing part or all of the steps of any of the methods of the second aspect.

The fifth aspect of the present invention discloses a polar coding quantizer, including: a processor, a transmitter, and a memory, the memory configured to store instructions, the processor configured to execute the instructions, the processor executing the instructions to perform some or all of the steps of any of the methods of the first aspect of the embodiments of the present invention. Wherein the polar encoding quantizer reduces the complexity of encoding when performing part or all of the steps of any of the methods of the first aspect.

A sixth aspect of the present invention discloses a polarization decoding quantizer, including: a processor, a receiver, and a memory, the memory configured to store instructions, the processor configured to execute the instructions, the processor executing the instructions to perform some or all of the steps of any of the methods of the second aspect of the embodiments of the present invention. Wherein, the decoding complexity of the polar decoding quantizer can be reduced when the quantizer performs part or all of the steps of any of the methods of the second aspect.

A seventh aspect of the embodiments of the present invention discloses a computer storage medium, which stores a program, where the program specifically includes instructions for executing some or all of the steps of any of the methods of the first aspect of the embodiments of the present invention.

An eighth aspect of the embodiments of the present invention discloses a computer storage medium, which stores a program that specifically includes instructions for executing some or all of the steps of any of the methods of the second aspect of the embodiments of the present invention.

In the embodiment of the invention, the polarization coding quantizer can determine coding parameters based on partial polarization transformation, perform multi-layer bit soft information calculation on the source symbol sequence by using the coding parameters, and perform serial cancellation coding based on partial polarization transformation on the multi-layer bit soft information to obtain a first bit sequence. Therefore, when the polarization coding quantizer performs the polarization code serial cancellation coding, only part of polarization coding processing needs to be performed on the source symbol sequence, so that the coding complexity can be reduced.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a schematic diagram of a network architecture according to an embodiment of the present invention;

FIG. 2 is a flow chart of a data processing method according to an embodiment of the present invention;

FIG. 2.1 is a schematic structural diagram of a partial polarization encoding disclosed in the embodiment of the present invention;

FIG. 2.2 is a comparison diagram of coding complexity disclosed in the embodiment of the present invention;

FIG. 3 is a flow chart of another data processing method disclosed in the embodiment of the invention;

FIG. 3.1 is a schematic structural diagram of a partial polarization decoding disclosed in the embodiment of the present invention;

FIG. 4 is a schematic structural diagram of a polar encoding quantizer according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of another exemplary polarization encoded quantizer according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of a structure of a quantizer for polar decoding according to an embodiment of the present invention;

FIG. 7 is a schematic diagram of another exemplary structure of a polar encoding quantizer according to an embodiment of the present invention;

fig. 8 is a schematic structural diagram of another polar decoding quantizer according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The terms "first", "second", and "third", etc. in the description of the present invention and the above-described drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.

The embodiment of the invention discloses a data processing method and equipment, which can reduce the encoding complexity. The following are detailed below.

In order to better understand a data processing method disclosed in the embodiment of the present invention, a network architecture suitable for the embodiment of the present invention is described below.

Referring to fig. 1, fig. 1 is a schematic diagram of a Network architecture disclosed in an embodiment of the present invention, where the Network architecture shown in fig. 1 is suitable for a Cloud-Radio Access Network (C-RAN) scenario. As shown in fig. 1, the network architecture includes: a Baseband processing Unit (BBU), a Remote Radio Unit (RRU), and a terminal. The terminal may include, but is not limited to, various user terminals such as a smart phone, a notebook Computer, a Personal Computer (PC), a Personal Digital Assistant (PDA), a Mobile Internet Device (MID), and an intelligent wearable Device (e.g., a smart watch and a smart band).

In a C-RAN scene, the BBU performs centralized management and is connected with the RRUs through optical fibers, and one BBU can support a plurality of RRUs. For downlink data transmission, for example, a BBU generates an information source symbol sequence, a polarization coding quantizer in the BBU encodes the information source symbol sequence to obtain encoded data (i.e., a compressed bit sequence), and sends the encoded data to an RRU via a forward (frontaul) link.

In Long Term Evolution (LTE), a multi-antenna technology is widely used to improve the system throughput, which inevitably increases the data transmission amount between the BBU and the RRU, and thus requires more fiber resources to be consumed. In addition, with the introduction of the long term evolution advanced (LTE-advanced) technology, the bandwidth of the LTE-advanced technology will be 5 times that of the LTE, and such a high data transmission amount will bring about a larger optical fiber resource overhead. Therefore, compressing the transmission data is an urgent technical problem to be solved.

In the network architecture shown in fig. 1, on the transmitting end side, the polar coding quantizer may determine a coding parameter based on partial polar transformation, perform multi-layer bit soft information calculation on the source symbol sequence using the coding parameter, and perform serial cancellation coding based on partial polar transformation on the multi-layer bit soft information to obtain a first bit sequence. Therefore, when the polarization coding quantizer performs the polarization code serial cancellation coding, only part of polarization coding processing needs to be performed on the source symbol sequence, so that the coding complexity can be reduced.

On the receiving end side, the polarization decoding quantizer can receive the first bit sequence and the coding parameters, perform serial-to-parallel conversion processing on the first bit sequence to obtain a multi-layer conversion bit sequence, perform partial polarization decoding processing on each layer of conversion bit sequence by using the coding parameters to obtain a second bit sequence, and perform bit-to-symbol mapping processing on the second bit sequences obtained by all layers by using the coding parameters to recover the source symbol sequence. Therefore, the polarization decoding quantizer only performs partial polarization decoding processing on the received first bit sequence, so that the complexity of decoding can be reduced.

Based on the network architecture shown in fig. 1, the embodiment of the invention discloses a data processing method. Referring to fig. 2, fig. 2 is a flow chart illustrating a data processing method according to an embodiment of the present invention, wherein the data processing method is applied to a polar coding quantizer. The data processing method comprises the following steps:

201. the polar encoding quantizer determines encoding parameters based on the partial polar transform.

In the embodiment of the invention, an offline stage and an online stage can be defined. The off-line stage is mainly used for setting parameters of the polarization coding quantizer, and the on-line stage is mainly used for processing the information source symbol sequence in real time.

Specifically, the polar encoding quantizer may set the encoding parameters based on the partial polar transform at an off-line stage. As an alternative embodiment, the determining of the coding parameters based on the partial polar transform by the polar coding quantizer comprises the following steps:

11) setting system parameters of a polarization coding quantizer;

12) and constructing N virtual experimental channels according to system parameters, wherein N is the length of the information source symbol sequence.

13) Determining an input value set of the virtual experimental channel

And setting a mapping rule between symbols and bits

14) And constructing a polarization code coding structure based on partial polarization transformation.

In this alternative embodiment, in step 11), the system parameters of the polar coding quantizer may include the number of polar code serial cancellation encoders, the coding rate of the source symbol sequence, and the code length of each polar code serial cancellation encoder. Wherein the source symbol sequence is represented by (y)₁,...,y_N) Is shown in which

Is the value set of the source symbol, and N is the length of the source symbol sequence. Specifically, the polar coding quantizer may be configured to include p partial polar code serial cancellation encoders based on partial polar transformation, where the polar coding quantizer quantizes each source symbol into q bits (that is, the coding rate of the source symbol sequence is q), and the code length of each polar code serial cancellation encoder is N, where q is an integer or a decimal larger than zero, p is a positive integer larger than q, and N is a positive integer of a power of 2.

In step 12), N parallel virtual test channels can be constructed according to the coding code rate q

The virtual test channel can be understood as a transmission channel between the input of the polar encoding quantizer and the output of the polar decoding quantizer. Each virtual experiment channel includes p bit subchannels. The noise variance of the virtual test channel can be determined by a rate-distortion function, the input of the virtual test channel W

In order to reconstruct the quantized symbols,

output of the virtual test channel W

Is a sequence of source symbols.

In step 13), the whole real number interval may be quantized first, then each hierarchical level value is obtained, then the quantization level is obtained, and finally the quantization level is used as a value of a reconstructed quantization symbol, that is, an input value set of the virtual experimental channel. Wherein x is_iIn order to reconstruct the quantized symbols,

to form the p bits of the reconstructed quantized symbol,

{0,1}^prepresenting p cartesian products of 0, 1. Rules for mapping between symbols and bits

In (1), symbol x_iAnd bit

And correspond to each other. Wherein the mapping rule

May be pre-specified, such as: set partitioning mapping rules. For example, when

Then, under the set partitioning mapping rule, x₁,x₂,x₃,x₄Mapped to 00,10,01,11, respectively.

In step 14), the method for constructing the partial polarization transformation-based polarization code coding structure by the polarization coding quantizer specifically includes the following steps:

a) and calculating the reliability metric value of each first bit sub-channel, wherein the first bit sub-channel is obtained after all the bit sub-channels are subjected to polarization transformation.

b) And determining a polarization transformation stopping threshold value of the partial polarization coding according to the reliability metric values of the plurality of first bit sub-channels.

c) And setting the flag bit of the coding node which needs to be subjected to polarization transformation as a first flag value and setting the flag bit of the coding node which does not need to be subjected to polarization transformation as a second flag value according to the polarization transformation stop threshold value, wherein the first flag value is different from the second flag value.

d) And calculating the reliability metric of each second bit sub-channel according to the polarization transformation stopping threshold value, the zone bit of the coding node and a preset reliability metric algorithm, wherein the second bit sub-channel is obtained after the bit sub-channel is subjected to partial polarization transformation.

e) And selecting qN target bit sub-channels with larger reliability metric values of the second bit sub-channels from the pN second bit sub-channels according to the sequence of the reliability metric values of the second bit sub-channels from large to small, and forming a reliable bit mark set by the subscripts of the qN target bit sub-channels.

f) And determining the flag bits and the reliable bit flag sets of all the coding nodes as a polarization code coding structure based on partial polarization transformation.

In conventional polar code construction, N parallel channels need to be log₂N-level complete polarization transformation is carried out to obtain N polarized sub-channels. Wherein, the polarization code is a channel code based on polarization transformation. Specifically, the polar code is the first channel coding which is theoretically proven to be capable of achieving the capacity of a symmetric binary input discrete channel, and the core technology of the polar code is polar transformation, namely, a plurality of parallel independent channels are transformed into a plurality of sub-channels with correlation and capacity differentiation through a channel merging and channel dividing method. It is proved by theory that when the number of channels is sufficiently large, a part of the capacity of the sub-channels generated by the polarization transformation is 0, and the other part of the capacity is 1, that is, the capacity of the channels has polarization effect.

In the step a), each virtual experimental channel includes p bit sub-channels, and for each bit sub-channel, the polarization coding quantizer may first perform all polarization transformation on the bit sub-channel to obtain a first bit sub-channel, and then calculate a reliability metric of the first bit sub-channel. The "total polarization transformation" here is the pointer that needs to perform a total log for each bit subchannel₂N-order complete polarization transformation.

In particular, virtual test channels are calculated

The kth bit subchannel

Channel capacity of

Wherein k is the serial number of the bit sub-channel, and k is more than or equal to 1 and less than or equal to p; the input value set of the kth bit subchannel is {0,1}, the output is the source symbol sequence and the first (k-1) bits, {0,1}^k-1(k-1) sub-Cartesian products representing {0,1 }; further, the kth bit subchannel is equivalent to a binary input gaussian channel of equal capacity, and the channel transfer function of the gaussian channel is calculated. For N virtual test channels

Of the kth bit subchannel

Performing all polarization operations (i.e., log)₂N-level polarization transformations) to obtain N first bit subchannels of all polarizations, and performing a predetermined reliability metric algorithm (for example: a density evolution algorithm and a Gaussian approximation algorithm) for calculating the reliability metric values of the N first bit sub-channels, wherein the density evolution algorithm is a method for analyzing the progressive performance of modern high-efficiency error-correction coding and decoding, specifically, messages can be transmitted and iterated between variable nodes and check nodes in the decoding process, and the probability density of the messages can evolve, which is called as density evolution; the basic idea of the simplified calculation method of the density evolution algorithm is to simplify the probability density of the message into Gaussian distribution for analysis and convert the multidimensional operation into the one-dimensional operation of solving the mean value of the Gaussian distribution when the Gaussian approximation is the Gaussian channel model.

Further, the above calculation steps are repeated p times to obtain pN first bit sub-channels and a reliability metric value of each first bit sub-channel. In addition, the pN first bit sub-channels of all polarizations may be sorted according to the descending order of the reliability metric of the first bit sub-channels, and the sorted pN first bit sub-channels may be selected from the sorted pN first bit sub-channelsThe former qN first bit sub-channels, and the serial numbers of the selected qN first bit sub-channels are marked as a set

In step b), first, p groups of first bit sub-channels of pN total polarizations, each group having a length of N, are grouped together for a kth group of a plurality of first bit sub-channels

The first bit sub-channel of (a) is subjected to summation of error probabilities to obtain E_kAnd counting the k-th group of the plurality of first bit sub-channels belonging to the set

The number of the first bit sub-channel is recorded as l_k。

In general, of the pN total-polarization first bit subchannels, one part of the first bit subchannels is a good channel (error probability approaches 0), and the other part of the first bit subchannels is a bad channel (error probability approaches 1). The error probability threshold value T of the channel can be set_gAnd error probability threshold value T of the difference channel_b，1＞T_b＞＞T_gIs greater than 0; then, an error probability threshold value of a good channel in the k-th partial polarization code is calculated

And error probability threshold of the difference channel

Wherein h (x) -xlogx- (1-x) log (1-x), h^-1(. cndot.) represents the inverse function of h (. cndot.).

In the steps c) and d, specifically, the Flag (t, j) of all coding nodes is cleared, wherein the Flag (t, j) represents the Flag bit of the jth coding node in the tth-level polarization transformation,

1≤j≤2^tmeanwhile, Flag (1, j) of the jth coding node of the first level is set as a first Flag value (for example, Flag (1, j) ═ 0) which needs to be subjected to polarization transformation, and the reliability metric P (1, j) of the jth coding node in the first level of polarization transformation is calculated according to a preset reliability metric algorithm (for example, a gaussian approximation algorithm).

In the channel partial polarization transformation, the reliability metric of the coding node may be referred to as a reliability metric of the second bit subchannel, that is, the second bit subchannel is obtained after the original bit subchannel is subjected to partial polarization transformation, and the following is similar.

In the process of partial polarization of the channel, when the polarization conversion is carried out to a certain middle stage, the error probability of the second bit sub-channel is less than the threshold value T of the set good channel_gI.e. the second bit subchannel has performed sufficiently well, the second bit subchannel may be left unpolarized; similarly, when the partial polarization conversion is carried out to a certain middle stage, the error probability of the second bit sub-channel is larger than the threshold value T of the set difference channel_bI.e. the second bit subchannel has exhibited a sufficient difference, the second bit subchannel is also left unpolarized. Therefore, can be used to connect T_gAnd T_bA polarization transformation stop threshold value is determined as a partial polarization encoding.

Starting from the second stage, for the t-th stage, if the t-1 st stage is

Of a coding node

For the first flag value to be subjected to polarization transformation, the t-1 th level of the t-th level is calculated according to a preset reliability metric algorithm (such as a Gaussian approximation algorithm)

Reliability measures P (t, j) of other coding nodes of the spread of one coding node, if at the same time

The Flag (t, j) of the jth coding node of the tth stage can be set as a first Flag value to be subjected to polarization transformation; if it is not

Or

The Flag (t, j) of the jth coding node of the tth stage may be set to a second Flag value that does not require polarization transformation (e.g., Flag (t, j) ═ 1). Wherein, the t-1 th level represents the previous level of the t-th level, j represents the serial number of the coding node in the t-th level, t is a positive integer and t is equal to [2, log ]_2N]，

Indicating rounding up.

Starting from the second stage, for the t-th stage, if the t-1 st stage is

Of a coding node

For the second flag value not requiring polarization transformation, then order

I.e. the t-1 st stage

The reliability metric of each coding node is directly used as the reliability metric of the jth coding node of the tth level, and meanwhile, Flag (t, j) of the jth coding node and Flag bits of other coding nodes expanded by the jth coding node can be set as second Flag values which do not need polarization transformation.

Until the log of completion₂N levels are obtained, the reliability measurement values of all N partial polarization second bit sub-channels are obtained, and the p partial polarization transformation-based pN second ratios can be obtained by repeating the operation for p timesThe bit sub-channel, and the flag bits of all coding nodes.

In step e), qN target bit sub-channels with larger reliability metric values of the second bit sub-channels can be selected from the pN second bit sub-channels according to the sequence of the reliability metric values of the second bit sub-channels from large to small, and the subscripts of the qN target bit sub-channels form a reliable bit flag set.

After the steps a) to e) are completed, the preparation work of the coding quantizer in the off-line stage is completed, and the flag bits and the reliable bit flag sets of all the coding nodes can be determined as the polarization code coding structure based on the partial polarization transformation.

202. The polarization coding quantizer uses the coding parameters to calculate multi-layer bit soft information for the source symbol sequence, and carries out serial cancellation coding based on partial polarization transformation on the multi-layer bit soft information to obtain a first bit sequence.

In the embodiment of the invention, bit soft information is used for representing probability information of each bit value of 0 or 1 forming the information source symbol, and the serial offset coding is a serial coding method for calculating the current coding bit by utilizing the coded bit. Polarization transformation is an operation of combining and dividing multiple parallel, independent channels to generate multiple correlated, capacity-differentiated subchannels.

In an embodiment of the present invention, the virtual test channel includes a transfer function p (y)_j| x), wherein,

x passes the mapping rule between symbols and bits

And

one for each, j ═ 1,2,. N }.

Specifically, the calculating of the multi-layer bit soft information for the source symbol sequence by the polarization coding quantizer using the coding parameters, and the serial cancellation coding based on the partial polarization transformation for the multi-layer bit soft information to obtain the first bit sequence may include the following steps:

21) when calculating the bit soft information of the i-th layer of the source symbol sequence, p (y) is used_j| x) and

and according to the formula

Calculating bit soft information L of ith bit in jth symbol_i(j) And mixing L_i(j) Bit likelihood ratio as the first stage of coding node in ith polar code serial offset coder

is a bit sequence (b) composed of ith to pth bits in the jth symbol_i,...,b_p)，i＝{1,2,...p}。

22) J (th) coded bit u at i (th) layer_jIf j is an odd number, let j equal to 2z-1, according to the formula

Calculating u_jBit likelihood ratio of

If j is an even number, let j equal 2z, according to the formula

Calculating u_jIs/are as follows

23) J (th) coded bit u at i (th) layer_jIf j is an odd number, let j equal to 2z-1, according to the formula

)2 calculation of u_jIs/are as follows

If j is an even number, let j equal 2z, according to the formula

Calculating u_jIs/are as follows

Wherein z is a positive integer, n is the number of stages of the coding node, and n-1 represents the previous stage of the nth stage of the coding node.

24) Obtained by calculation for each layer

If it is

Less than a preset threshold, determining u_jThe first bit value is different from the second bit value; wherein,

representing the source symbol sequence and N being the length of the source symbol sequence.

25) All the calculated u_jIs determined as the first bit sequence.

In the embodiment of the invention, after the coding parameters of the polarization coding quantizer are set in the off-line stage, the information source symbol sequence can be processed in real time in the on-line stage.

In step 21), the bit likelihood ratio is used to represent probability information of a bit value of 0 or 1 on the coding node.

In step 22) and step 23), the j-th coded bit u of the polar coding quantizer performs the serial cancellation coding_jBit likelihood ratio of

It may not be necessary to compute all the way from the first level to the log as in conventional schemes₂And N levels, and the bit likelihood ratio of the subsequent coding node is replaced by the bit likelihood ratio of the middle node. Wherein the jth coded bit u_jBit likelihood ratio of

Is log of₂N-level bit likelihood ratios for the jth coding node.

Specifically, in step 22), in the case that the flag bit of the jth coding node is the first flag value, u needs to be calculated according to the formula_jBit likelihood ratio of

In step 23), in the case that the flag bit of the jth coding node is the second flag value, it is not necessary to calculate according to the formula, but only simple iteration is needed to obtain u according to the bit likelihood ratio of the middle-level coding node_jBit likelihood ratio of

Please refer to fig. 2.1, fig. 2.1 is a schematic structural diagram of a partial polarization encoding according to an embodiment of the present invention. The code length N of the polar coding quantizer shown in fig. 2.1 is 8, and it is necessary to calculate the bit likelihood ratio of the coding node (e.g., a circle or a square shown in fig. 2.1) in the coding node in fig. 2.1. As shown in fig. 2.1, only the coding nodes of the first and second levels (shown by solid lines in fig. 2.1) need to be calculated, while the coding nodes of the third level (shown by dotted lines in fig. 2.1) need not participate in the calculation, but the bit likelihood ratio of the coding node of the third level is directly replaced by the bit likelihood ratio of the coding node of the second level.

Please refer to fig. 2.2, fig. 2.2 is a schematic diagram illustrating a comparison of coding complexity according to an embodiment of the present invention. As shown in fig. 2.2, for the polarization coding quantization scheme based on all polarizations (i.e. the conventional polarization coding quantization scheme), the number of coding nodes to be calculated in coding is pNlog₂N, the coding complexity is usually measured by the number of coding nodes that need to be calculated, where

Indicating a rounding down. As shown by the line segment labeled 1 in fig. 2.2, the line segment 1 is used to represent the relationship between the coding complexity obtained by all the polarization transformations and the coding rate q, and it can be seen from the line segment 1 that the coding complexity increases linearly with the coding rate q. As shown by the line segment with the reference number 2 in fig. 2.2, the line segment 2 is used to represent the relationship between the coding complexity obtained by the partial polarization transformation and the coding rate q, and as can be seen from the line segment 2, the coding complexity does not change with the coding rate. That is, at a higher coding rate, the present invention has a greater reduction ratio of the coding complexity in the polarization coding quantization scheme based on partial polarization, that is: with the increase of the coding rate, the overall polarization code length also shows linear increase, the more the channel polarization effect is, the smaller the polarization level required by the channel to reach the set threshold in the polarization process is, and then more coding nodes do not need to participate in the operation. The complexity of the polarization coding quantization scheme based on partial polarization is 4-5 times less than that of the traditional polarization coding quantization scheme based on all polarization, so that the coding complexity can be reduced.

In step 24), log is calculated₂When N stages (i.e. the last stage) can be pairedOf each layer

And making a hard decision. Specifically, a preset threshold (e.g. 1) may be preset, if

If the value is larger than the preset threshold value, determining u_jIs a first bit value, such as u _j0; if it is

Less than a preset threshold, determining u_jTakes the value of a second bit value, e.g. u_j＝1。

In step 25), all p layers are calculated to obtain u_jIs determined as the first bit sequence.

203. The polar coding quantizer sends the first bit sequence and the coding parameters to the receiving end device.

In the embodiment of the invention, the coding parameters comprise mapping rules between symbols and bits

Flag bits for all coding nodes, and a set of reliable bit flags. After receiving the coding parameter and the first bit sequence, the receiving end device may decode the first bit sequence according to the coding parameter to recover the source symbol sequence. The receiving end device may be a polar decoding quantizer.

In the method flow described in fig. 2, when the polar coding quantizer performs the polar code serial cancellation coding, only part of the polar coding processing needs to be performed on the source symbol sequence, so that the coding complexity can be reduced.

Based on the network architecture shown in fig. 1, the embodiment of the invention discloses another data processing method. Referring to fig. 3, fig. 3 is a flow chart illustrating another data processing method according to an embodiment of the present invention, wherein the data processing method is applied to a polar decoding quantizer. The data processing method comprises the following steps:

301. a polar decoding quantizer receives the first bit sequence and the encoding parameters.

In this embodiment of the present invention, the polar decoding quantizer may receive the first bit sequence and the coding parameter sent by the polar coding quantizer. Wherein the encoding parameters include mapping rules between symbols and bits

Flag bits for all coding nodes, and a set of reliable bit flags.

302. And the polarization decoding quantizer performs serial-to-parallel conversion processing on the first bit sequence to obtain a multilayer conversion bit sequence.

In the embodiment of the present invention, the first bit sequence received by the polarization decoding quantizer is serial, and at this time, the polarization decoding quantizer needs to perform serial-to-parallel conversion on the first bit sequence to obtain a multi-layer transform bit sequence.

303. The polarization decoding quantizer uses the coding parameters to perform partial polarization decoding processing on each layer of the transformed bit sequence to obtain a second bit sequence.

The encoding parameters in step 303 include a flag bit of the encoding node and a reliable bit flag set, where the flag bit is a first flag value or a second flag value, the first flag value is used for the encoding node to perform polarization transformation, and the second flag value is used for the encoding node not to perform polarization transformation;

specifically, the performing, by the polarization decoding quantizer, partial polarization decoding processing on each layer of the transformed bit sequence using the encoding parameter to obtain the second bit sequence includes the following steps:

31) and determining the zone bit of the coding node as the zone bit of the decoding node.

32) J (th) decoded bit at i (th) layer

If j is an odd number, let j equal to 2z-1, according to the formula

Calculating the last level of the jth decoding node

Calculating the last level of the jth decoding node

The numerical values of (a), wherein,

the operation of modulo two addition is represented,

is the bit value of the first bit sequence.

33) J (th) decoded bit at i (th) layer

If j is an odd number, let j equal to 2z-1, according to the formula

Calculating the last level of the jth decoding node

Calculating the last level of the jth decoding node

The value of (d); wherein z is a positive integer, n is the number of stages of the decoding node, and n-1 represents the previous stage of the nth stage of the decoding node.

34) All the obtained N numbers are calculated

And determining the second bit sequence as the ith layer, wherein N is the length of the second bit sequence of each layer.

In step 31), the flag bit of the decoding node coincides with the flag bit of the coding node, so that the flag bit of the coding node can be determined as the flag bit of the decoding node in the polar decoding quantizer.

In step 32) and step 33), when the polarization decoding quantizer calculates the decoding bit on the decoding node, the jth decoding bit

The value of (A) may not have to be computed from the first stage all the way to the log as in conventional schemes₂And N stages, the decoding bits of the previous stage partial decoding node can be directly replaced by the input bits received by the previous stage according to the flag bit of the decoding node, and the calculation operation of the decoding bits on the decoding node is not started until a certain intermediate stage. Wherein the preceding stage can be from the first stage to the log₂Any one of the N stages. Specifically, in step 32), under the condition that the flag bit of the jth decoding node is the first flag value, the last stage of the jth decoding node needs to be calculated according to a formula

The value of (d); in step 33), when the flag bit of the jth decoding node is the second flag value, calculation according to a formula is not needed, and only the input bit received by the previous stage is directly used for substitution.

Please refer to fig. 3.1, fig. 3.1 is a schematic structural diagram of partial polarization decoding according to an embodiment of the present invention. The code length N of the polar decoding quantizer shown in fig. 3.1 is 8, and from the input side of the polar decoding quantizer, the decoded bits of the partial decoding node (shown by the dotted line in fig. 3.1) of the first stage do not need to be calculated, but the bit values of the received first bit sequence are directly transmitted to the decoding node of the second stage as decoded input bits, and the calculation of the decoded bits of the decoding node is performed only from the second stage. Since the calculation of the decoded bits of only part of the decoding nodes is performed, the complexity of decoding can be reduced.

304. The polar decoding quantizer performs a bit-to-symbol mapping process on the second bit sequences obtained by all layers using the coding parameters to recover the source symbol sequence.

In the embodiment of the present invention, the encoding parameter in step 304 is a mapping rule between symbols and bits

Polar decoding quantizer using mapping rules

And carrying out bit-to-symbol mapping processing on the second bit sequences obtained by all the layers, so that the source symbol sequence can be recovered.

In the method flow described in fig. 3, the polar decoding quantizer performs only a partial polar decoding process on the received first bit sequence, so that the decoding complexity can be reduced.

Please refer to fig. 4, and fig. 4 is a schematic structural diagram of a polar encoding quantizer according to an embodiment of the present invention, wherein the polar encoding quantizer is configured to execute the data processing method disclosed in fig. 2, and specifically refer to the related description in fig. 2, which is not repeated herein. As shown in fig. 4, the polar encoding quantizer 400 includes:

a determining unit 401 for determining coding parameters based on partial polarization transformation;

an executing unit 402, configured to perform multi-layer bit soft information calculation on the source symbol sequence by using the coding parameters, and perform serial cancellation coding based on partial polarization transformation on the multi-layer bit soft information to obtain a first bit sequence.

Please refer to fig. 5, and fig. 5 is a schematic structural diagram of another polar encoding quantizer disclosed in an embodiment of the present invention, wherein the polar encoding quantizer is configured to execute the data processing method disclosed in fig. 2, and specifically refer to the related description in fig. 2, which is not repeated herein. In the polar encoding quantizer shown in fig. 5, the polar encoding quantizer includes p polar code serial cancellation encoders based on partial polar transform, the polar encoding quantizer shown in fig. 5 is further optimized on the basis of the polar encoding quantizer shown in fig. 4, and compared with the polar encoding quantizer shown in fig. 4, the determining unit 401 of the polar encoding quantizer shown in fig. 5 includes, except for all units of the polar encoding quantizer shown in fig. 4:

a setting sub-unit 4011 configured to set system parameters of the polar encoding quantizer;

a first constructing subunit 4012, configured to construct N virtual experimental channels according to the system parameter, where N is the length of the source symbol sequence;

a first determining subunit 4013, configured to determine an input value set of the virtual experimental channel

And setting a mapping rule between symbols and bits

Wherein x is_iIn order to reconstruct the quantized symbols,

to form the p bits of the reconstructed quantized symbol,

{0,1}^pp cartesian products representing {0,1 };

a second construction subunit 4014, configured to construct a partial polarization transform-based polarization code coding structure.

Optionally, each virtual experimental channel includes p bit sub-channels, the system parameter includes a coding code rate q of the source symbol sequence, and the manner of constructing the polarization code coding structure based on the partial polarization transformation by the second constructing sub-unit 4014 is specifically:

Optionally, the virtual test channel comprises a transfer function p (y)_j| x), wherein,

said x passes through

And the above-mentioned

One-to-one correspondence, j ═ 1,2,. N };

the execution unit 402 includes:

a first calculating subunit 4021, configured to use the p (y) in calculating bit soft information of an i-th layer of a source symbol sequence_j| x) and the

And according to the formula

Calculating bit soft information L of ith bit in jth symbol_i(j) And combining said L_i(j) Bit likelihood ratio as a first stage of a coding node in an ith said polar code successive cancellation coder

A second calculating subunit 4022 for calculating a jth coded bit u at the ith layer_jIf j is an odd number, let j equal to 2z-1, according to the formula

Calculating said u_jBit likelihood ratio of

If j is an even number, let j equal 2z, according to the formula

Calculating the u_jIs/are as follows

The second calculating subunit 4022 is further configured to determine a jth coded bit u in an ith layer_jIf j is an odd number, let j equal to 2z-1, according to the formula

Calculating the u_jIs/are as follows

If j is an even number, let j equal 2z, according to the formula

Calculating the u_jIs/are as follows

a second determination subunit 4023 for calculating the obtained for each layer

If it is

Less than the preset threshold value, determining u_jIs a second bit value, the first bit value andthe second bit value is different; wherein,

the second determining subunit 4023 is further configured to determine all the calculated u_jIs determined as the first bit sequence.

Optionally, the polar coding quantizer 400 shown in fig. 5 may further include:

a sending unit 403, configured to send the first bit sequence and the coding parameter to a receiving end device, where the coding parameter is used for the receiving end device to decode the first bit sequence to recover the source symbol sequence.

In the polar coding quantizer 400 shown in fig. 4 to 5, when the polar coding quantizer 400 performs the polar code serial cancellation coding, only a partial polar coding process needs to be performed on the source symbol sequence, so that the complexity of coding can be reduced.

Referring to fig. 6, fig. 6 is a schematic structural diagram of a polarization decoding quantizer according to an embodiment of the present invention, wherein the polarization decoding quantizer is used for executing the data processing method disclosed in fig. 3, and specific reference is made to the related description in fig. 3, which is not repeated herein. As shown in fig. 6, the polar decoding quantizer 600 includes:

a receiving unit 601, configured to receive a first bit sequence and a coding parameter;

a first processing unit 602, configured to perform serial-to-parallel conversion on the first bit sequence to obtain a multi-layer conversion bit sequence;

a second processing unit 603, configured to perform partial polarization decoding processing on the transform bit sequence of each layer by using the coding parameters, so as to obtain a second bit sequence;

a third processing unit 604, configured to perform a bit-to-symbol mapping process on the second bit sequences obtained by all layers using the coding parameters, so as to recover the source symbol sequence.

Optionally, the encoding parameter includes a flag bit of the encoding node and a reliable bit flag set, where the flag bit includes a first flag value or a second flag value, the first flag value is used to indicate that the encoding node needs to perform polarization transformation, and the second flag value is used to indicate that the encoding node does not need to perform polarization transformation;

the second processing unit 603 performs partial polarization decoding on the transformed bit sequence of each layer using the coding parameters, and the manner of obtaining the second bit sequence specifically is as follows:

determining the zone bit of the coding node as the zone bit of a decoding node;

j (th) decoded bit at i (th) layer

If j is an odd number, let j equal to 2z-1, according to the formula

Calculating the last level of the jth decoding node

Calculating the last level of the jth decoding node

The numerical values of (a), wherein,

the operation of modulo two addition is represented,

is the bit value of the first bit sequence;

j (th) decoded bit at i (th) layer

If j is an odd number, let j equal to 2z-1, according to the formula

Calculating the last level of the jth decoding node

Calculating the last level of the jth decoding node

all the obtained N numbers are calculated

In the polar decoding quantizer 600 shown in fig. 6, the polar decoding quantizer performs only a partial polar decoding process on the received first bit sequence, so that the decoding complexity can be reduced.

Please refer to fig. 7, and fig. 7 is a schematic structural diagram of another polar encoding quantizer disclosed in an embodiment of the present invention, wherein the polar encoding quantizer is configured to execute the data processing method disclosed in fig. 2, and specifically refer to the related description in fig. 2, which is not repeated herein. As shown in fig. 7, the polar encoding quantizer 700 includes: a processing unit 701, a transmitting unit 702, and a storage unit 703. The processing unit 701, the transmitting unit 702, and the storage unit 703 are connected to a communication bus, respectively. The storage unit 703 may be a high-speed RAM memory or a non-volatile memory (non-volatile memory), and the transmission unit 702 may be a transmitter having a function of transmitting information to an opposite terminal, such as an antenna. It will be understood by those skilled in the art that the structure of the polar coded quantizer 700 shown in fig. 7 is not intended to limit the present invention, and may be a bus structure, a star structure, a combination of more or less than those shown in fig. 7, or a different arrangement of components.

The Processing Unit 701 is a control center of the polar coding quantizer 700, and may be a Central Processing Unit (CPU), and the Processing Unit 701 connects various parts of the entire polar coding quantizer 700 by using various interfaces and lines, and executes or executes software programs and/or modules stored in the storage Unit 703, and calls program codes stored in the storage Unit 703, so as to perform the following operations:

determining coding parameters based on the partial polarization transformation;

calculating multilayer bit soft information of the source symbol sequence by using the coding parameters, and performing serial offset coding based on partial polarization transformation on the multilayer bit soft information to obtain a first bit sequence;

the first bit sequence and the coding parameters are sent to a receiving end device through a sending unit 702, and the coding parameters are used for the receiving end device to decode the first bit sequence so as to recover the source symbol sequence.

Optionally, the above-mentioned polarization coding quantizer 700 includes p polarization code serial cancellation encoders based on partial polarization transformation, and the manner for determining the coding parameters based on partial polarization transformation by the processing unit 701 is specifically:

setting system parameters of the polar encoding quantizer;

determining a set of input values for the virtual experiment channelCombination of Chinese herbs

And setting a mapping rule between symbols and bits

Wherein x is_iIn order to reconstruct the quantized symbols,

to form the p bits of the reconstructed quantized symbol,

{0,1}^pp cartesian products representing {0,1 };

Optionally, each virtual experimental channel includes p bit sub-channels, the system parameter includes a coding rate q of the source symbol sequence, and a manner of constructing a polarization code coding structure based on partial polarization transformation by the processing unit 701 is specifically:

said x passes through

And the above-mentioned

One-to-one correspondence, j ═ 1,2,. N };

the processing unit 701 may further call a program code stored in the storage unit 703, so as to perform the following operations:

using the coding parameters to calculate multi-layer bit soft information for the source symbol sequence, and performing serial cancellation coding based on partial polarization transformation for the multi-layer bit soft information to obtain a first bit sequence, including:

And according to the formula

Calculating the u_jBit likelihood ratio of

If j is an even number, let j equal 2z, according to the formula

Calculating the u_jIs/are as follows

Calculating the u_jIs/are as follows

If j is an even number, let j equal 2z, according to the formula

Calculating the u_jIs/are as follows

obtained by calculation for each layer

If it is

all the calculated u_jIs determined as the first bit sequence.

In the polar coding quantizer 700 depicted in fig. 7, the polar coding quantizer 700 only needs to perform partial polar coding processing on the source symbol sequence when performing polar code serial cancellation coding, so that the complexity of coding can be reduced.

Please refer to fig. 8, and fig. 8 is a schematic structural diagram of another polar decoding quantizer according to an embodiment of the present invention, wherein the polar decoding quantizer is used for executing the data processing method disclosed in fig. 3, and specifically refer to the related description in fig. 3, which is not repeated herein. As shown in fig. 8, the polar decoding quantizer 800 includes: a processing unit 801, a receiving unit 802, and a storage unit 803. The processing unit 801, the receiving unit 802, and the storage unit 803 are connected to a communication bus, respectively. The storage unit 803 may be a high-speed RAM memory or a non-volatile memory (non-volatile memory), and the receiving unit 802 may be a receiver having a function of receiving peer-to-peer information, such as an antenna. It will be understood by those skilled in the art that the structure of the polar decoding quantizer 800 shown in fig. 8 is not intended to limit the present invention, and may be a bus structure, a star structure, a combination of more or fewer components than those shown in fig. 8, or a different arrangement of components.

The Processing Unit 801 is a control center of the polar decoding quantizer 800, and may be a Central Processing Unit (CPU), and the Processing Unit 801 connects various parts of the entire polar decoding quantizer 800 by using various interfaces and lines, and executes or executes software programs and/or modules stored in the storage Unit 803, and calls program codes stored in the storage Unit 803 to perform the following operations:

receiving the first bit sequence and the encoding parameter by the receiving unit 802;

using the coding parameters to perform partial polarization decoding processing on the transformation bit sequence of each layer to obtain a second bit sequence;

and carrying out bit-to-symbol mapping processing on the second bit sequences obtained by all the layers by using the coding parameters so as to recover the source symbol sequence.

the processing unit 801 performs partial polarization decoding processing on the transform bit sequence of each layer by using the encoding parameters, and obtaining a second bit sequence includes:

determining the zone bit of the coding node as the zone bit of a decoding node;

j (th) decoded bit at i (th) layer

If j is an odd number, let j equal to 2z-1, according to the formula

Calculating the last level of the jth decoding node

Calculating the last level of the jth decoding node

The numerical values of (a), wherein,

the operation of modulo two addition is represented,

is the bit value of the first bit sequence;

j (th) decoded bit at i (th) layer

If j is an odd number, let j equal to 2z-1, according to the formula

Calculating the last level of the jth decoding node

Calculating the last level of the jth decoding node

all the obtained N numbers are calculated

In the polar decoding quantizer 800 shown in fig. 8, the polar decoding quantizer 800 performs only a partial polar decoding process on the received first bit sequence, so that the decoding complexity can be reduced.

It should be noted that, for simplicity of description, the above-mentioned embodiments of the method are described as a series of acts or combinations, but those skilled in the art should understand that the present application is not limited by the order of acts described, as some steps may be performed in other orders or simultaneously according to the present application. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and elements referred to are not necessarily required in this application.

In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.

The above disclosure is only for the purpose of illustrating the preferred embodiments of the present invention, and it is therefore to be understood that the invention is not limited by the scope of the appended claims.

Claims

1. A data processing method, comprising:

determining coding parameters based on the partial polarization transformation;

wherein the data processing method is applied to a polar coding quantizer, the polar coding quantizer comprises p partial polar transform-based polar code serial cancellation encoders, and the determining of the partial polar transform-based encoding parameters comprises:

setting system parameters of the polar encoding quantizer;

determining an input value set X ═ { X ] of the virtual experimental channel_i:1≤i≤2^pAnd setting a mapping rule between symbols and bits

Wherein x is_iIn order to reconstruct the quantized symbols,

to form the p bits of the reconstructed quantized symbol,

{0,1}^pp cartesian products representing {0,1 };

2. The method according to claim 1, wherein each of the virtual experiment channels includes p bit sub-channels, the system parameter includes a coding rate q of the source symbol sequence, and the method for constructing the partial polarization transform-based polarization code coding structure specifically comprises:

3. The method of claim 2, wherein the step of removing the substrate comprises removing the substrate from the substrateThe virtual experimental channel comprises a transfer function p (y)_j| x), wherein,

said x is through said L and said

One-to-one correspondence, j ═ 1,2,. N };

using said p (y) in calculating the bit soft information of the i-th layer of the source symbol sequence_j| x) and said L, according to the formula

Computing stationU is described_jBit likelihood ratio of

If j is an even number, let j equal 2z, according to the formula

Calculating the u_jIs/are as follows

Calculating the u_jIs/are as follows

If j is an even number, let j equal 2z, according to the formula

Calculating the u_jIs/are as follows

obtained by calculation for each layer

If it is

all the calculated u_jIs determined as the first bit sequence.

4. The method according to any one of claims 1 to 3, further comprising:

5. A data processing method, comprising:

receiving a first bit sequence and a coding parameter;

using the coding parameters to perform bit-to-symbol mapping processing on the second bit sequences obtained by all the layers so as to recover the source symbol sequence;

the encoding parameters are determined as follows:

setting system parameters of a polarization coding quantizer;

Wherein x is_iIn order to reconstruct the quantized symbols,

to form the p bits of the reconstructed quantized symbol,

{0,1}^pp cartesian products representing {0,1 };

6. The method of claim 5, wherein the coding parameters comprise a flag bit of a coding node and a set of reliable bit flags, the flag bit comprises a first flag value or a second flag value, the first flag value is used for indicating that the coding node needs to perform a polarization transformation, and the second flag value is used for indicating that the coding node does not need to perform a polarization transformation;

determining the zone bit of the coding node as the zone bit of a decoding node;

j (th) decoded bit at i (th) layer

If j is an odd number, let j equal to 2z-1, according to the formula

Calculating the last level of the jth decoding node

Calculating the last level of the jth decoding node

The numerical values of (a), wherein,

the operation of modulo two addition is represented,

is the bit value of the first bit sequence;

j (th) decoded bit at i (th) layer

If j is an odd number, let j equal to 2z-1, according to the formula

Calculating the last level of the jth decoding node

Calculating the last level of the jth decoding node

The value of (d); wherein z is a positive integer, n is the number of decoding node stages, and n-1 represents decodingThe previous stage of the nth stage of the node;

all the obtained N numbers are calculated

7. A polar coded quantizer, comprising:

a processing unit for determining coding parameters based on the partial polarization transform;

the processing unit is used for calculating multilayer bit soft information of the source symbol sequence by using the coding parameters and carrying out serial cancellation coding based on partial polarization transformation on the multilayer bit soft information to obtain a first bit sequence;

a sending unit, configured to send the first bit sequence and the coding parameter to a receiving end device, where the coding parameter is used for the receiving end device to decode the first bit sequence to recover the source symbol sequence;

the polar coding quantizer comprises p polar code serial cancellation encoders based on partial polar transformation, and the mode for determining the coding parameters based on partial polar transformation by the processing unit is specifically as follows:

setting system parameters of the polar encoding quantizer;

Wherein x is_iIn order to reconstruct the quantized symbols,

to form the p bits of the reconstructed quantized symbol,

{0,1}^pp cartesian products representing {0,1 };

8. The polar-coding quantizer according to claim 7, wherein each of the virtual experimental channels comprises p bit sub-channels, the system parameter comprises a coding rate q of the source symbol sequence, and the processing unit constructs a polar-code coding structure based on partial polar transform specifically by:

9. The polar-coded quantizer according to claim 8, wherein the virtual experimental channel comprises a transfer function p (y)_j| x), wherein,

said x is through said L and said

One-to-one correspondence, j ═ 1,2,. N };

the processing unit uses the coding parameters to calculate multi-layer bit soft information for the source symbol sequence, and performs serial cancellation coding based on partial polarization transformation for the multi-layer bit soft information, and the mode of obtaining the first bit sequence specifically is as follows:

J (th) coded bit u at i (th) layer_jIs present in the reliable bit flagIf j is odd, let j be 2z-1 according to formula

Calculating the u_jBit likelihood ratio of

If j is an even number, let j equal 2z, according to the formula

Calculating the u_jIs/are as follows

Calculating the u_jIs/are as follows

If j is an even number, let j equal 2z, according to the formula

Calculating the u_jIs/are as follows

obtained by calculation for each layer

If it is

all the calculated u_jIs determined as the first bit sequence.

10. A polar decoding quantizer, comprising:

a receiving unit, configured to receive a first bit sequence and a coding parameter;

the processing unit is used for carrying out serial-parallel conversion processing on the first bit sequence to obtain a multilayer conversion bit sequence;

the processing unit is further configured to perform partial polarization decoding processing on the transform bit sequence of each layer by using the coding parameters to obtain a second bit sequence;

the processing unit is further configured to perform bit-to-symbol mapping processing on the second bit sequences obtained by all the layers by using the coding parameters, so as to recover an information source symbol sequence;

wherein the encoding parameter is determined by:

setting system parameters of the polar encoding quantizer;

Wherein x is_iIn order to reconstruct the quantized symbols,

to form the p bits of the reconstructed quantized symbol,

{0,1}^pp cartesian products representing {0,1 };

11. The quantization apparatus of claim 10, wherein the coding parameters comprise a flag bit of a coding node and a set of reliable bit flags, the flag bit comprises a first flag value or a second flag value, the first flag value is used to indicate that the coding node needs to perform the polarization transformation, and the second flag value is used to indicate that the coding node does not need to perform the polarization transformation;

the processing unit uses the coding parameters to perform partial polarization decoding processing on the transform bit sequence of each layer, and the manner of obtaining the second bit sequence specifically is as follows:

determining the zone bit of the coding node as the zone bit of a decoding node;

j (th) decoded bit at i (th) layer

If j is an odd number, let j equal to 2z-1, according to the formula

Calculating the last of the jth decoding nodeStage

Calculating the last level of the jth decoding node

The numerical values of (a), wherein,

the operation of modulo two addition is represented,

is the bit value of the first bit sequence;

j (th) decoded bit at i (th) layer

If j is an odd number, let j equal to 2z-1, according to the formula

Calculating the last level of the jth decoding node

Calculating the last level of the jth decoding node

The value of (d); wherein z is a positive integer and n is a decoding nodeThe number of stages, n-1, represents the previous stage of the nth stage of the decoding node;

all the obtained N numbers are calculated

12. A computer-readable storage medium, characterized in that it stores a computer program which, when executed by hardware, is adapted to perform the method of any one of claims 1 to 6.