CN107707333B - Method and device for stopping early iteration of polarization code based on code word estimated value - Google Patents

Method and device for stopping early iteration of polarization code based on code word estimated value Download PDF

Info

Publication number
CN107707333B
CN107707333B CN201710827228.1A CN201710827228A CN107707333B CN 107707333 B CN107707333 B CN 107707333B CN 201710827228 A CN201710827228 A CN 201710827228A CN 107707333 B CN107707333 B CN 107707333B
Authority
CN
China
Prior art keywords
iteration
output
node
decoding
estimated value
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201710827228.1A
Other languages
Chinese (zh)
Other versions
CN107707333A (en
Inventor
张小军
董雁飞
崔建明
张德学
曾庆田
隋荣全
张荣才
张作文
陈晨
李俊
烟晓凤
王庆祥
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong University of Science and Technology
Original Assignee
Shandong University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shandong University of Science and Technology filed Critical Shandong University of Science and Technology
Priority to CN201710827228.1A priority Critical patent/CN107707333B/en
Publication of CN107707333A publication Critical patent/CN107707333A/en
Application granted granted Critical
Publication of CN107707333B publication Critical patent/CN107707333B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0056—Systems characterized by the type of code used
    • H04L1/0057—Block codes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0045—Arrangements at the receiver end
    • H04L1/0047—Decoding adapted to other signal detection operation
    • H04L1/005—Iterative decoding, including iteration between signal detection and decoding operation

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Error Detection And Correction (AREA)

Abstract

The invention discloses a method and a device for stopping early iteration of a polarization code based on a code word estimated value,the method comprises the following steps: s1) presetting the maximum iteration times of BP decoding; s2) decoding the polarized coded information by using a BP decoding algorithm; s3) in each iteration, the estimated value of the code word obtained from the current iteration is used
Figure DDA0001407781510000011
Judging the change rate of the current time; if in the course of successive iterations
Figure DDA0001407781510000012
If no change occurs, stopping iteration and outputting a decoding result obtained by current iteration, otherwise, continuing iteration until reaching the preset maximum iteration times. The method and the device for stopping the early iteration of the polarization code based on the code word estimated value provided by the invention are used for judging the code word estimated value obtained by the current iteration in each iteration process
Figure DDA0001407781510000013
The change rate of the decoding is judged to realize an early iteration stop mechanism, so that the calculation complexity and the decoding delay of decoding can be greatly reduced, and the hardware resource consumption is effectively reduced.

Description

Method and device for stopping early iteration of polarization code based on code word estimated value
Technical Field
The present invention relates to a method and an apparatus for processing a polar code, and more particularly, to a method and an apparatus for stopping early iteration of a polar code based on a codeword estimation value.
Background
In the international information theory ISIT conference in 2008, Arikan first proposed the concept of channel polarization, and based on this theory, he given the first known channel coding method that can be strictly proven to reach the channel capacity and named Polar Code (Polar Code). The polarization code is designed based on the channel polarization. After a plurality of channels are subjected to channel transformation through a special recursive coding process, namely when N is large enough, the capacity of a part of polarized channels tends to 1, and the capacity of the rest of polarized channels tends to 0. Useful information is transmitted using subchannels whose channel capacity tends to 1, and certain information is transmitted using subchannels whose channel capacity tends to 0.
In practical decoding of polarization codes, a Successive Cancellation (SC) algorithm and a Belief Propagation (BP) algorithm are two common algorithms. The SC algorithm is less computationally complex. The improved SC algorithm may achieve better decoding performance than the original SC and BP algorithms by either a list decoding strategy or a stack decoding strategy. But due to the serial decoding architecture it suffers from a high decoding delay o (L), where L is the code length. High decoding delay prevents the widespread use of polar codes, especially for high speed scenarios.
Unlike SC decoding, BP decoding is an iterative algorithm and is parallel. The calculation complexity of the BP algorithm is O (I)aveNlog N), wherein IaveIs an average iterationThe number of times. It follows that for a BP decoder, the decoding delay and power consumption increase linearly with the number of iterations. The traditional BP decoder has a preset maximum iteration number Imax. BP decoding will not terminate until the maximum number of iterations is reached. This termination scheme lacks flexibility and introduces significant computational complexity. In order to reduce redundant iterations as much as possible, related researchers have proposed various early iteration stop criteria based on G Matrix (G-Matrix), Minimum log likelihood ratio (minLLR), log likelihood ratio amplitude assist (LMA), etc., but these early iteration criteria still have the problem of high computational complexity, etc.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a method and a device for stopping the early iteration of the polarization code based on a code word estimated value, which can greatly reduce the calculation complexity and decoding delay of decoding and are convenient for hardware implementation.
The technical scheme adopted by the invention for solving the technical problems is to provide a method for stopping the early iteration of the polarization code based on the estimated value of the code word, which comprises the following steps: s1) presetting the maximum iteration times of BP decoding; s2) decoding the polarized coded information by using a BP decoding algorithm; s3) in each iteration process, the estimated value of the code word x obtained by the current iteration is obtained
Figure GDA0001494256670000021
Judging the change rate of the current time; if in the course of successive iterations
Figure GDA0001494256670000022
If no change occurs, stopping iteration and outputting a decoding result obtained by current iteration, otherwise, continuing iteration until reaching the preset maximum iteration times.
In the method for stopping early iteration of a polar code based on a codeword estimation value, in step S3), the change rate of the codeword estimation value is used as a determination criterion, and if it is determined that the codeword estimation value does not change in the continuous multiple iteration processes, the iteration is stopped and a decoding result obtained by the current iteration is output.
The above-mentioned polarization code early iteration stopping method based on the codeword estimation value, wherein the codeword estimation value used in the step S3) is calculated as follows: for the polarization code with the parameter of (N, K), the corresponding factor graph is represented by N-log2N-order computing unit and N +1 column nodes, each order is composed of N/2 processing units, and the estimation value of the information bit sequence u can be obtained by hard decision of the soft information in the 1 st column node at the leftmost end of the factor graph
Figure GDA0001494256670000023
Hard decision is carried out on soft information in the (n + 1) th column node at the rightmost end to obtain an estimated value of a code word x
Figure GDA0001494256670000024
(i, j) represents a node of the ith row and the jth column from the left; the soft information of each node passing through the node (i, j) from right to left is recorded as Li，jFrom left to right, the soft information passing through the node (i, j) is denoted as Ri，j(ii) a Said step S2) propagates the L in the update node to the left firsti，jStarting to propagate R in the update node to the right after reaching the leftmost sidei，j(ii) a After the iteration is terminated, if the bit is not an information bit, the bit is decoded to 0, otherwise, the log likelihood ratio L of the leftmost end is usedi，1Determines whether the information bit is 0 or 1.
The above-mentioned polarization code early iteration stopping method based on the estimated value of the code word, wherein, the step S3) implements the early iteration stopping mechanism by using the hard decision, the equal detector and the threshold comparator in combination; updating to the n +1 th column node of the factor graph in each iteration, namely obtaining R from PE of the BP decoderi，n+1To R, to Ri，n+1Hard decision is carried out to obtain the estimated value of the code word in the current iteration
Figure GDA0001494256670000025
The equality detector obtains the codeword estimate for t-1 iterations from memory
Figure GDA0001494256670000026
Derived from a hard decision module
Figure GDA0001494256670000027
The equality detector compares
Figure GDA0001494256670000028
And
Figure GDA0001494256670000029
if the output C is the same as the output C, the output C is 1, otherwise, the output C is 0; the threshold comparator consists of an adder, a register and a comparator, wherein the register stores the same number of continuous times, if C is 1, the register is added with 1, and if C is 0, the register is cleared; the comparator compares the continuous same times stored in the memory with the size of a preset parameter X, if the continuous same times are equal to the size of the preset parameter X, the output D is 1, otherwise, the output D is 0; and if D is 1, the BP decoder stops iterating and outputting the decoding result, and if D is 0, the BP decoder continues iterating until the preset maximum iteration number is reached.
The method for stopping the early iteration of the polar code based on the estimated value of the code word comprises the steps that the preset parameter X is 1-10, and the maximum iteration number is preset to be 15-80.
The present invention further provides a device for stopping early iteration of a polar code based on a codeword estimation value, which comprises: the BP decoder decodes the polarization coding information in an iteration mode and presets the maximum iteration times; a hard decision device: the sign bit of the soft information after each iteration is taken out to obtain the estimated value of the code word in the current iteration
Figure GDA0001494256670000031
An equality detector: obtaining the estimated value of code word of last iteration from memory
Figure GDA0001494256670000032
Comparison
Figure GDA0001494256670000033
And
Figure GDA0001494256670000034
if the output C is the same as the output C, the output C is 1, otherwise, the output C is 0; a threshold comparator: the circuit comprises an adder, a register and a comparator, wherein the register stores the same number of continuous times, if C is 1, the register is added with 1, and if C is 0, the register is cleared; the comparator compares the continuous same times stored in the memory with the size of a preset parameter X, if the continuous same times are equal to the size of the preset parameter X, the output D is 1, otherwise, the output D is 0; and if D is 1, the BP decoder stops iterating and outputting the decoding result, and if D is 0, the BP decoder continues iterating until the preset maximum iteration number is reached.
The device for stopping the early iteration of the polar code based on the estimated value of the code word comprises a preset parameter X of 1-10 and a maximum iteration number preset to 15-80.
Compared with the prior art, the invention has the following beneficial effects: the method and the device for stopping the early iteration of the polarization code based on the code word estimated value provided by the invention are used for judging the code word estimated value obtained by the current iteration in each iteration process
Figure GDA0001494256670000035
The change rate of (a) is judged to realize an early iteration stop mechanism. The test result shows that compared with the traditional BP decoder with the fixed iteration times of 40 times, the iteration times can be reduced by 84.35 percent when the signal-to-noise ratio is 3.5dB, so that the calculation complexity and the decoding delay of decoding are greatly reduced. The comprehensive results show that compared with the existing three common early iteration stopping mechanisms, the logic resource consumption can be reduced by 63.25-96.88%, and the hardware resource consumption is effectively reduced. The existing early iteration stop criteria all depend on Li，1Or
Figure GDA0001494256670000036
Will cause extra decoding delay, the present invention relies on
Figure GDA0001494256670000037
No additional delay is incurred.
Drawings
FIG. 1 is a diagram of the polarization code factor with the parameter (8, 4) according to the present invention;
FIG. 2 is a diagram of the basic elements of the polarization code factor graph of the present invention;
FIG. 3 is a diagram illustrating a hardware architecture implementation of the present invention;
FIG. 4 is an equivalent detector hardware configuration;
FIG. 5 is a threshold comparator hardware configuration;
FIG. 6 is a flow chart of early iteration stop computation according to the present invention;
FIG. 7 is a diagram illustrating comparison between the decoding performance of the BP decoding method with maximum iteration number of 40 and the original BP decoder according to the polarization code with parameter (1024, 512) of the present invention;
fig. 8 is a schematic diagram showing comparison between the maximum number of iterations of the BP decoding method with the maximum number of iterations of the polarization code with the parameter (1024, 512) of the present invention and the average number of iterations of the original BP decoder under different snr channels.
Detailed Description
The invention is further described below with reference to the figures and examples.
In practice, iterative decoders may or may not converge at an early stage of the iterative process. That is, I is reached at the number of iterationsmaxPreviously, the decoding may have been successful. The early iteration stop algorithm is used for adaptively detecting whether reliable decoding output is obtained or not in the decoding iteration process, and if the condition is met, the decoding can be immediately finished. The early iteration stop algorithm can linearly reduce the decoding complexity and decoding delay.
The invention provides a polarization code early iteration stopping method based on a code word estimated value, which comprises the following steps:
s1) presetting the maximum iteration times of BP decoding;
s2) decoding the polarized coded information by using a BP decoding algorithm;
s3) in each iteration process, judging the change rate of the estimated value of the code word obtained by the current iteration; and if the code word estimated value in the continuous multi-iteration process is not changed, stopping iteration and outputting a decoding result obtained by current iteration, otherwise, continuing iteration until the preset maximum iteration times is reached.
Codeword based estimation of the invention
Figure GDA0001494256670000041
The method for stopping the early iteration of the polarization code decodes the polarization coding information by using a BP decoding algorithm, and the information updating formula is as follows:
Figure GDA0001494256670000042
f(x，y)≈α*sign(x)sign(y)min(|x|，|y|) (2)
using estimated values of codewords
Figure GDA0001494256670000043
If the rate of change is within X successive iterations
Figure GDA0001494256670000044
None of the changes have occurred, the output at this point can be considered to be sufficiently reliable, referred to as the x-assisted (x-assisted) early iteration stop criterion. The specific formula is as follows:
Figure GDA0001494256670000045
i.e. the iteration can be stopped when the above equation is satisfied. And if the formula (3) is not satisfied, continuing iteration until the preset maximum iteration number is reached. Compared with the prior art, the invention can obviously reduce the decoding iteration times under the condition of not causing the loss of decoding performance, and has more obvious effect particularly in a high signal-to-noise ratio channel; the method has strong reliability and simple hardware implementation.
Next, a test is performed using Polar code with parameters (1024, 512), where the code length is N-1024, K-512, and the code rate is 0.5. A simulation result is used as position information under the condition that the signal-to-Noise ratio is 1.5dB by using a Gaussian approximation method, the modulation mode is Binary Phase Shift Keying (BPSK), and the channel is an additive white Gaussian Noise channel (AWGN). Code word
Figure GDA0001494256670000051
From length 1024
Figure GDA0001494256670000052
And multiplying the generated matrix G. Generating a matrix
Figure GDA0001494256670000053
Representation matrix
Figure GDA0001494256670000054
Log of (a)21024 ═ 10 kronecker products. Channel received value Y1 NRepresented using Log-likelihood ratio (LLR) form. In the implementation, a 7-bit quantization scheme of 1-bit sign bit, 4-bit integer bit and 2-bit decimal bit is adopted, so that the hardware implementation is facilitated.
The factor graph of the polarization code with the parameter (1024, 512) is represented by log2Processing units (PEs) of 1024 th order 10 and 11 columns of nodes (a factor graph with a code length of 8 is shown in fig. 1), where each order consists of N/2 th order 512 PEs (fig. 2 is a schematic diagram of a single processing unit). The soft information in the 1 st node at the leftmost end of the factor graph is judged to obtain an information bit sequence
Figure GDA0001494256670000055
Is estimated value of
Figure GDA0001494256670000056
The soft information in the 11 th column node at the rightmost end is judged to obtain a code word
Figure GDA0001494256670000057
Is estimated value of
Figure GDA0001494256670000058
(i, j) represents the node of the ith row and the jth column from the left. Each node has two kinds of information, and the invention records the information passing through the node (i, j) from right to left as Li，jThe information passing through the node (i, j) from left to right is denoted as Ri，jThese information pass updates to each other in LLR form.
In the decoding process, the maximum iteration number of preset iteration is 40, and R is firstly compared with Ri，1And Li，11Initialization is performed. L isi，11Initialized to the channel acceptance value Yi，Ri，1A maximum value of 15.75 that the 0 and 7 bit quantization schemes can represent is initialized based on the position information, respectively. The specific formula is as follows:
Figure GDA0001494256670000059
Li，11＝Yi(5)
a in the formula refers to a set of information bits. Then, L in the updated node is propagated to the left according to the formula (1)i，jStarting to propagate R in the update node to the right after reaching the leftmost sidei，jα in equation (2) is set to 0.9, as specified by the equation:
f(x，y)≈0.9*sign(x)sign(y)min(|x|，|y|) (6)
the code word estimated value can be obtained according to the following two formulas in each iteration process
Figure GDA0001494256670000061
Figure GDA0001494256670000062
Figure GDA0001494256670000063
In order to reduce the computational complexity and decoding delay of decoding and facilitate hardware implementation, the idea of the invention is to use the estimated value of the codeword
Figure GDA0001494256670000064
Is taken as a criterion, the parameter X is set to 1 if in the course of 1 successive iteration
Figure GDA0001494256670000065
If no change occurs, the output of the BP decoder may be considered to be sufficiently reliable at this time. The specific formula is as follows:
Figure GDA0001494256670000066
if equation (9) is satisfied, the decoding output at this time can be regarded as a reliable decoding output, and the iteration is terminated. If the condition is not met, judging whether the decoding times reach 40 times, and if the decoding times reach the maximum iteration, terminating the iteration; if not, the information is continuously updated, and the next iteration is carried out.
After the iteration is terminated, if the bit is not an information bit, the bit is decoded to 0, otherwise, the log likelihood ratio L of the leftmost end is usedi，1(subscript 1 represents the first column of nodes, i.e., the leftmost column of nodes in the factor graph, and i represents the ith bit) to determine whether the information bit is 0 or 1, and the formula according to which the information bit is determined is as follows:
Figure GDA0001494256670000067
fig. 3 is a schematic diagram of a hardware architecture implementation of the present invention, which mainly comprises an equality detector and a threshold comparator. The R can be obtained from the PE of the BP decoder by updating to the 11 th column node of the factor graph in each iterationi，11To R, to Ri，11Hard decision is carried out to obtain the estimated value of the code word in the current iteration
Figure GDA0001494256670000068
The equality detector obtains from memory the codeword estimates for t-1 iterations
Figure GDA0001494256670000069
Derived from a hard decision module
Figure GDA00014942566700000610
Equality detector comparison
Figure GDA00014942566700000611
And
Figure GDA00014942566700000612
and if the output C is the same, the output C is 1, otherwise, the output C is 0. The threshold comparator consists of an adder, a register and a comparator, wherein the register stores the same number of continuous times, if C is equal to 1, l is added, and if C is equal to 0, zero clearing is carried out; the comparator compares the same number of consecutive times stored in the memory with the size of a preset parameter X equal to 1, if the same number is equal, the output D equal to 1, otherwise, the output D equal to 0. And if D is 1, the BP decoder stops iterating and outputting the decoding result, and if D is 0, the BP decoder continues iterating until the preset maximum iteration number is reached. Fig. 4 shows a detailed hardware structure of the equivalent detector in fig. 3, and fig. 5 shows a detailed structure of the threshold comparator.
FIG. 6 is a calculation flow of the (N, K) polarization code decoding process according to the present invention. The existing early iteration stop criteria cause extra delay and cannot be obtained
Figure GDA00014942566700000613
It was previously determined whether to proceed to the next iteration, since these criteria all depend on
Figure GDA00014942566700000614
Or to
Figure GDA00014942566700000615
Obtained after making a hard decision
Figure GDA00014942566700000616
As shown in FIG. 6, the nth clock in t iterations is followed by a pass pair
Figure GDA00014942566700000617
Making a decision to obtain
Figure GDA00014942566700000618
Then, the invention can finish the early iteration stop judgment only by two clocks, if the early iteration stop criterion calculation of the invention is satisfied
Figure GDA0001494256670000071
To pair
Figure GDA0001494256670000072
Obtaining a decoding result after hard decision, and then stopping iteration; if the early iteration stop criterion of the present invention is not met, the next iteration is performed.
Fig. 7 shows the results of testing various early iteration-stop criteria with a conventional polar-code BP decoder in a gaussian additive white noise channel. In the figure, the abscissa Eb/N0 is the signal-to-noise ratio, the FER is the frame error rate, the BER is the bit error rate, the constant iterations indicate that the iteration times are constant, and the x-aid indicates that the BP early iteration stop decoding method provided by the invention is used. It can be seen from fig. 7 that the present invention can achieve the same decoding performance as the conventional BP decoder in the case that the number of iterations is significantly less than that of the conventional BP decoder.
Fig. 8 shows the average number of iterations of various early iteration stop criteria and a conventional polar code BP decoder in different snr channels. In the figure, Eb/N0 is the signal-to-noise ratio, Average number of iterations represents the Average iteration number, and x-aid represents that the BP early iteration stop decoding method provided by the invention is used. When the maximum number of iterations is 40 and Eb/N0 is 3.5dB, the average number of iterations is reduced by 84.35% compared to the original BP decoder whose number of iterations is fixed to 40.
The following table shows the x-aid mechanism of this implementation combined with three other proposed early iteration stop mechanisms in stratxv 5SGXEA7N2F455C2, where Q is the number of LLR quantization bits.
Figure GDA0001494256670000073
As can be seen from the table, the consumption of Logic routing and Registers is reduced by 63.25% and 99.9% respectively in terms of hardware resource consumption compared to the G-Matrix mechanism. The consumption of Logic utilization was reduced by 72.44% and 96.88% compared to minLLR and LMA, respectively.
Although the present invention has been described with respect to the preferred embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (4)

1. A method for stopping early iteration of a polarization code based on a code word estimated value is characterized by comprising the following steps:
s1) presetting the maximum iteration times of BP decoding;
s2) decoding the polarized coded information by using a BP decoding algorithm;
s3) in each iteration, the estimated value of the code word obtained from the current iteration is used
Figure FDA0002443343430000011
Judging the change rate of the current time; if in the course of successive iterations
Figure FDA0002443343430000019
Stopping iteration and outputting a decoding result obtained by current iteration if the current iteration is not changed, otherwise, continuing the iteration until the preset maximum iteration times are reached;
the step S3) uses the change rate of the codeword estimation value as a determination criterion, and if it is determined that the codeword estimation value does not change in the continuous multiple iteration processes, stops the iteration and outputs a decoding result obtained by the current iteration;
the calculation of the codeword estimation value used in said step S3) is as follows: for the polarization code with parameter (N, K), N is code length, K is information bit, and the corresponding factor graph is represented by N-log2N-order computing unit and N +1 column nodes, each order is composed of N/2 processing units, and the estimation value of the information bit sequence u can be obtained by hard decision of the soft information in the 1 column node at the leftmost end of the factor graph
Figure FDA0002443343430000012
Hard decision is carried out on soft information in the (n + 1) th column node at the rightmost end to obtain an estimated value of a code word x
Figure FDA0002443343430000013
(i, j) represents a node of the ith row and the jth column from the left; the soft information of each node passing through the node (i, j) from right to left is recorded as Li,jFrom left to right, the soft information passing through the node (i, j) is denoted as Ri,j(ii) a Said step S2) propagates the L in the update node to the left firsti,jStarting to propagate R in the update node to the right after reaching the leftmost sidei,j(ii) a After the iteration is terminated, if the bit is not an information bit, the bit is decoded to 0, otherwise, the log likelihood ratio L of the leftmost end is usedi,1The sign of (1) determines whether the information bit is 0 or 1;
said step S3) implements an early iteration stop mechanism by using a hard decision, an equality detector and a threshold comparator in combination, C being the output of the equality detector and D being the output of the threshold comparator; the node of the n +1 th column of the factor graph is updated in each iteration, and R can be obtained by a processing unit of the BP decoderi,n+1To R, to Ri,n+1Hard decision is carried out to obtain a codeword estimation value in the current iteration
Figure FDA0002443343430000014
The equality detector obtains the codeword estimate for t-1 iterations from memory
Figure FDA0002443343430000015
Derived from a hard decision module
Figure FDA0002443343430000016
The equality detector compares
Figure FDA0002443343430000017
And
Figure FDA0002443343430000018
if the output C is the same as the output C, the output C is 1, otherwise, the output C is 0; the threshold comparator consists of an adder, a register and a comparator, wherein the register stores the same number of continuous times, if C is 1, the register is added with 1, and if C is 0, the register is cleared; the comparator compares successive identical times stored in the memoryThe number is equal to the size of a preset parameter X, if the number is equal to the size of the preset parameter X, the output D is equal to 1, otherwise, the output D is equal to 0; and if D is 1, the BP decoder stops iterating and outputting the decoding result, and if D is 0, the BP decoder continues iterating until the preset maximum iteration number is reached.
2. The method for stopping early iteration of polar code based on estimated value of code word as claimed in claim 1, wherein the predetermined parameter X is 1-10, and the maximum number of iterations is predetermined to be 15-80.
3. An apparatus for stopping early iteration of a polar code based on an estimated value of a codeword, comprising:
the BP decoder decodes the polarization coding information in an iteration mode and presets the maximum iteration times;
a hard decision device: the sign bit of the information after each iteration is taken out to obtain the estimated value of the code word in the current iteration
Figure FDA0002443343430000028
An equality detector: obtaining the estimated value of code word of last iteration from memory
Figure FDA0002443343430000029
Comparison
Figure FDA00024433434300000210
And
Figure FDA00024433434300000211
if the output C is the same as the output C, the output C is 1, otherwise, the output C is 0;
a threshold comparator: the circuit comprises an adder, a register and a comparator, wherein the register stores the same number of continuous times, if C is 1, the register is added with 1, and if C is 0, the register is cleared; the comparator compares the continuous same times stored in the memory with the size of a preset parameter X, if the continuous same times are equal to the size of the preset parameter X, the output D is 1, otherwise, the output D is 0; if D is 1, the BP decoder stops iterating to output a decoding result, and if D is 0, the BP decoder continues iterating until a preset maximum iteration time is reached;
the codeword estimate is calculated as follows: for the polarization code with parameter (N, K), N is code length, K is information bit, and the corresponding factor graph is represented by N-log2N-order computing unit and N +1 column nodes, each order is composed of N/2 processing units, and the estimation value of the information bit sequence u can be obtained by hard decision of the soft information in the 1 column node at the leftmost end of the factor graph
Figure FDA0002443343430000021
Hard decision is carried out on soft information in the (n + 1) th column node at the rightmost end to obtain an estimated value of a code word x
Figure FDA0002443343430000022
(i, j) represents a node of the ith row and the jth column from the left; the soft information of each node passing through the node (i, j) from right to left is recorded as Li,jFrom left to right, the soft information passing through the node (i, j) is denoted as Ri,j(ii) a Said step S2) propagates the L in the update node to the left firsti,jStarting to propagate R in the update node to the right after reaching the leftmost sidei,j(ii) a After the iteration is terminated, if the bit is not an information bit, the bit is decoded to 0, otherwise, the log likelihood ratio L of the leftmost end is usedi,1The sign of (1) determines whether the information bit is 0 or 1;
during iteration, an early iteration stopping mechanism is realized by using a hard decision, an equality detector and a threshold comparator in a combined mode, wherein C is the output of the equality detector, and D is the output of the threshold comparator; the node of the n +1 th column of the factor graph is updated in each iteration, and R can be obtained by a processing unit of the BP decoderi,n+1To R, to Ri,n+1Hard decision is carried out to obtain a codeword estimation value in the current iteration
Figure FDA0002443343430000023
The equality detector obtains the codeword estimate for t-1 iterations from memory
Figure FDA0002443343430000024
Derived from a hard decision module
Figure FDA0002443343430000025
The equality detector compares
Figure FDA0002443343430000026
And
Figure FDA0002443343430000027
if the output C is the same as the output C, the output C is 1, otherwise, the output C is 0; the threshold comparator consists of an adder, a register and a comparator, wherein the register stores the same number of continuous times, if C is 1, the register is added with 1, and if C is 0, the register is cleared; the comparator compares the continuous same times stored in the memory with the size of a preset parameter X, if the continuous same times are equal to the size of the preset parameter X, the output D is 1, otherwise, the output D is 0; and if D is 1, the BP decoder stops iterating and outputting the decoding result, and if D is 0, the BP decoder continues iterating until the preset maximum iteration number is reached.
4. The apparatus for stopping early iteration of polar code based on estimated value of code word as claimed in claim 3, wherein the predetermined parameter X is 1-10, and the maximum number of iterations is predetermined to be 15-80.
CN201710827228.1A 2017-09-14 2017-09-14 Method and device for stopping early iteration of polarization code based on code word estimated value Active CN107707333B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710827228.1A CN107707333B (en) 2017-09-14 2017-09-14 Method and device for stopping early iteration of polarization code based on code word estimated value

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710827228.1A CN107707333B (en) 2017-09-14 2017-09-14 Method and device for stopping early iteration of polarization code based on code word estimated value

Publications (2)

Publication Number Publication Date
CN107707333A CN107707333A (en) 2018-02-16
CN107707333B true CN107707333B (en) 2020-06-09

Family

ID=61172465

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710827228.1A Active CN107707333B (en) 2017-09-14 2017-09-14 Method and device for stopping early iteration of polarization code based on code word estimated value

Country Status (1)

Country Link
CN (1) CN107707333B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109495116B (en) * 2018-10-19 2022-07-12 东南大学 SC-BP mixed decoding method of polarization code and adjustable hardware architecture thereof
CN109547035B (en) * 2018-12-30 2023-02-10 山东科技大学 Method for establishing hardware architecture of pipelined BP polarized decoder and hardware architecture of decoder

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105187073A (en) * 2015-10-13 2015-12-23 东南大学 BP decoding method and device for polarization code

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150333775A1 (en) * 2014-05-15 2015-11-19 Broadcom Corporation Frozen-Bit Selection for a Polar Code Decoder

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105187073A (en) * 2015-10-13 2015-12-23 东南大学 BP decoding method and device for polarization code

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
《极化码的改进BP译码算法研究》;刘军旗;《中国优秀硕士学位论文全文库》;20170415;全文 *

Also Published As

Publication number Publication date
CN107707333A (en) 2018-02-16

Similar Documents

Publication Publication Date Title
CN107612560B (en) Polarization code early iteration stopping method based on partial information bit likelihood ratio
CN108847848B (en) BP decoding algorithm of polarization code based on information post-processing
CN108039891B (en) Polarization code BP decoding method and device based on multistage updating process
CN107241106B (en) Deep learning-based polar code decoding algorithm
CN108462558B (en) Method and device for decoding polarization code SCL and electronic equipment
JP3674111B2 (en) Data transmission device
US7539920B2 (en) LDPC decoding apparatus and method with low computational complexity algorithm
CN109586730B (en) Polarization code BP decoding algorithm based on intelligent post-processing
CN108702290A (en) Cascade Polarization Coding and sliding window Polarization Coding
KR20070039353A (en) Method for updating check node in low density parity check decoder
CN111726202B (en) Early termination iteration method for polarization code belief propagation decoding
Yuan et al. Construction and decoding algorithms for polar codes based on 2× 2 non-binary kernels
CN109547035B (en) Method for establishing hardware architecture of pipelined BP polarized decoder and hardware architecture of decoder
CN112332864A (en) Polar code decoding method and system for self-adaptive ordered mobile pruning list
CN110995279B (en) Polarization code combined SCF spherical list overturning decoding method
CN107707333B (en) Method and device for stopping early iteration of polarization code based on code word estimated value
CN111313913B (en) Low-delay cross-scheduling polarization code BP decoding method and device
GB2403106A (en) a turbo type decoder which performs decoding iterations on sub-blocks to improve convergence
KR20090012189A (en) Apparatus and method for decoding using performance enhancement algorithm for ldpc codes with scaling based min-sum iterative decoding
Albayrak et al. Low-complexity early termination method for rateless soft decoder
CN110855298B (en) Low iteration number polarization code BP decoding method based on subchannel freezing condition
Lv et al. On the performance of generalized SCL-flip decoding for polar codes
Ahmed et al. Viterbi algorithm performance analysis for different constraint length
Cui et al. Simplified early stopping criterion for polar codes
CN118054797B (en) Coding and decoding method, device and equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant