CN110212922B - Polarization code self-adaptive decoding method and system - Google Patents
Polarization code self-adaptive decoding method and system Download PDFInfo
- Publication number
- CN110212922B CN110212922B CN201910476954.2A CN201910476954A CN110212922B CN 110212922 B CN110212922 B CN 110212922B CN 201910476954 A CN201910476954 A CN 201910476954A CN 110212922 B CN110212922 B CN 110212922B
- Authority
- CN
- China
- Prior art keywords
- path
- decoding
- layer
- neural network
- unit
- 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
Links
Images
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M13/00—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
- H03M13/03—Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words
- H03M13/05—Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words using block codes, i.e. a predetermined number of check bits joined to a predetermined number of information bits
- H03M13/13—Linear 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
- H04L1/0052—Realisations of complexity reduction techniques, e.g. pipelining or use of look-up tables
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Theoretical Computer Science (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
- Error Detection And Correction (AREA)
Abstract
The invention discloses a polarization code self-adaptive decoding method and a polarization code self-adaptive decoding system, and belongs to the field of wireless communication. The method comprises the steps of firstly constructing a neural network and training, and then inputting path information to be decoded into the neural network to obtain the maximum path number of each layer; and then inputting the maximum path number of each layer into a decoder, and decoding the path information to be decoded by using the decoder to obtain a decoding result. The system comprises a neural network unit and a decoding unit, wherein the neural network unit is electrically connected with the decoding unit; the decoding unit comprises a path expansion unit and a sequencing unit, the path expansion unit is electrically connected with the sequencing unit, and the path expansion unit is electrically connected with the neural network unit. The invention aims to overcome the defects that the decoding algorithm based on deep learning has higher training complexity and is not suitable for long code application in the prior art, and can reduce the decoding complexity and meet different channel environments and configuration requirements of a communication system.
Description
Technical Field
The present invention relates to the field of wireless communication, and in particular, to a method and a system for adaptive decoding of a polarization code.
Background
To date, channel coding has been developed for over 70 years,
the proposed polar code has been proven to be the first code that can achieve the channel capacity of a symmetric binary input discrete memoryless channel (B-DMC). In 5G field experiments, the polar code achieves a great performance improvement effect, and is selected as a Forward Error Correction (FEC) code of an enhanced mobile broadband (eMBB) control channel. In order to meet the low delay and high speed requirements of 5G, researchers have made many efforts to design polar code decoders with high hardware efficiency. At present, it is urgent and necessary to develop an efficient polar code decoder that can well balance complexity and performance.
As one of the most widely used decoders for polar codes, a Successive Cancellation (SC) decoder may show advantages in terms of complexity, but is limited by a bit-by-bit decoding strategy and cannot achieve satisfactory finite length error correction performance. By extending more paths in each stage, sequential Cancellation List (SCL) and Sequential Cancellation Stack (SCS) decoders are proposed as improved SC decoders. The SCL decoder may be considered an application of a binary tree breadth-first search algorithm, and the SCS decoder may be approximated as a depth-first search algorithm. Simulation results show that both the SCL and SCS decoders can achieve near Maximum Likelihood (ML) decoder performance and have acceptable time complexity O (LNlogN), where L represents the search width and N represents the code length.
To avoid performance degradation, the search width L of the SCL needs to be of moderate size. In this case, the temporal complexity of the SCL decoder is still slightly higher. Therefore, a search width adaptation strategy and a pruning strategy are proposed to reduce the computational complexity. However, most of these strategies are controlled by a threshold value, and the reasonability of path deletion cannot be guaranteed, so that performance gains are few.
In recent years, deep Learning (DL) has attracted attention worldwide due to its powerful ability to solve complex tasks. With deep learning, significant performance improvements are achieved in many fields, such as computer vision, gaming and bioinformatics, and deep learning can also be applied to the field of polar code decoding. Conventional deep learning based decoders learn a large number of codewords to achieve near optimal Bit Error Rate (BER) performance. However, the exponentially increasing complexity in training and running neural networks has hindered their practical application in long code decoding. Therefore, the huge training amount and high running complexity are two main obstacles preventing the application of deep learning as an auxiliary scheme for polar code decoding.
In summary, it is easy to find out by observing the decoding algorithm that the existing SCL algorithm has the following problems: 1) The searching width L of the traditional SCL/SCS decoding algorithm is fixed and cannot be changed according to the decoding scene, and the complexity is higher; 2) The SCL/SCS decoding algorithm based on path deletion can bring certain performance loss; 3) The SC decoding algorithm based on deep learning has higher training complexity and is not suitable for long code application. Based on the above analysis, the existing decoding method of polarization code is not enough to meet the requirement of practical application.
Disclosure of Invention
1. Problems to be solved
The invention aims to overcome the defects that the SC decoding algorithm based on deep learning in the prior art has higher training complexity and is not suitable for long code application, and provides a polarization code self-adaptive decoding method which can reduce the decoding complexity and meet different channel environments and configuration requirements of a communication system.
2. Technical scheme
In order to solve the problems, the technical scheme adopted by the invention is as follows:
the invention relates to a polarization code self-adaptive decoding method, which comprises the steps of firstly constructing a neural network and training, and then inputting path information to be decoded into the neural network to obtain the maximum path number of each layer of an information code binary tree; then, the maximum path number of each layer is input into a decoder, paths to be decoded are parallelly expanded according to the maximum path number of each layer to obtain candidate paths, then the transition probabilities of the candidate paths are calculated and sequenced, and a decoding result is obtained according to the path length and the maximum transition probability.
Further, the method comprises the following steps: the method comprises the following steps of firstly, constructing a neural network, and constructing and training the neural network, wherein the neural network comprises an input layer, a hidden layer and an output layer; step two, predicting, namely inputting path information y to be decoded into a neural network to obtain the maximum path number Z of each layer i Then, the maximum number of paths Z of each layer is determined i Inputting to a decoder; wherein, i belongs to (1, \8230;, K-2), and K is the number of information bits; step three, initializing a decoder, and enabling the width of Q in the decoder to be L max Q is a list storing paths of each layer; initializing decoding paths and memory matrixes in Q and Q to be null; wherein the decoding path in Q is
Step four, expanding, if the ith level of the binary tree of the information code is a frozen bit, all decoding paths in Q are expanded to be the frozen bit
Storing the expanded decoding path into Q and executing step six; if the ith level of the binary tree of the information code is the information bit, reading the maximum path number Z corresponding to the layer i And Q is added to the front Z i The candidate path obtained by expanding the path is
And
and the transition probabilities of the candidate paths are different; step five, sorting, calculating the transition probability of the candidate paths, sorting, and selecting L from big to small max Storing the candidate paths into Q; and step six, deciding, if the path length in Q is the same as the code length N, outputting the path with the maximum transfer probability in Q as a decoding result, and if not, returning to the step four.
Further, the maximum number of paths per layer is calculated according to the following formula:
wherein, W ij For each layer corresponding weight, b j For each layer corresponding bias, x j For each layer corresponding input value.
Further, the transition probability is calculated as follows:
wherein the content of the first and second substances,
representing an input of u 1 The transition probability of a path of length i,
represents the odd terms of the decoding path of length i,
representing even terms of a coding path of length i,
in order to receive the vector of code words,
for the previous decoding path, u i Is the previous channel input.
Furthermore, the node number of the input layer is equal to the code length N, the input vector is the path information y to be decoded, the node number of the hidden layer is 8 × N, and the node number of the output layer is K-2.
Further, the neural network is trained using a TensorFlow platform, wherein the maximum number of iterations T epoch Is 50.
The invention relates to a decoding system adopting the self-adaptive decoding method of the polarization code, which comprises a neural network unit and a decoding unit, wherein the neural network unit is electrically connected with the decoding unit; the neural network unit is used for calculating the maximum path number of each layer in the information code binary tree, and the decoding unit is used for decoding the path information to be decoded.
Furthermore, the decoding unit comprises a path expansion unit and a sorting unit, the path expansion unit is electrically connected with the sorting unit, and the path expansion unit is electrically connected with the neural network unit.
3. Advantageous effects
Compared with the prior art, the invention has the beneficial effects that:
(1) According to the self-adaptive decoding method for the polarization code, the maximum path number of each layer is obtained through the neural network, so that the search width of each layer can be selected and determined, the method can be further suitable for various signal-to-noise ratio scenes, and the applicability of the method is improved; the candidate paths are obtained by parallelly expanding the paths to be decoded according to the maximum path number of each layer, so that the path expansion number of each layer is reduced, and the time and space complexity of the decoder is further reduced;
(2) The self-adaptive decoding system of the polar code combines deep learning with the traditional polar code decoder by arranging the neural network unit and the decoding unit, and limits path expansion by predicting the search width of each stage, thereby reducing the calculation complexity and simultaneously improving the decoding performance by early interruption of an error path.
Drawings
Fig. 1 is a flowchart illustrating a method for adaptive decoding of a polar code according to the present invention;
FIG. 2 is a schematic diagram of a adaptive decoding system for a polar code according to the present invention;
FIG. 3 is a schematic block diagram of a decoding system according to embodiment 2;
fig. 4 is a comparison between the performance of the decoding system of embodiment 2 and the conventional SC and SCL.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, 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 some embodiments of the present invention, but not all embodiments; moreover, the embodiments are not independent, and can be combined with each other according to requirements, so that a better effect is achieved. Thus, the following detailed description of the embodiments of the present invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.
For a further understanding of the invention, reference should be made to the following detailed description taken in conjunction with the accompanying drawings and examples.
Example 1
With reference to fig. 1, in the adaptive decoding method for the polarization code of the present invention, a neural network is first constructed and trained, and then path information to be decoded is input to the neural network to obtain the maximum path number of each layer of the binary tree of the information code; the maximum path number of each layer in the binary tree of the information code of the decoder is obtained through the neural network, so that the search width of each layer can be selected and determined, the method is further suitable for various signal-to-noise ratio scenes, and the applicability of the method is improved. Furthermore, the maximum path number of each layer is input into the decoder, and paths to be decoded are parallelly expanded according to the maximum path number of each layer to obtain candidate paths, so that the path expansion number of each layer is reduced, and the time and space complexity of the decoder is further reduced; then, the transition probability of the candidate paths is calculated and sequenced, and then a decoding result is obtained according to the path length and the maximum transition probability.
It should be noted that, in the present invention, the path information to be decoded is y, the code length (length of decoding) is N, the number of information bits is K, and the maximum search width of each layer in the decoding process is L max The list of paths in each layer is stored as Q, and the path with length i in Q is stored as
(i∈(1,…,N),j∈(1,…,L max ) A transition probability corresponding to the path is
The invention discloses a polarization code self-adaptive decoding method, which comprises the following specific steps:
step one, constructing a neural network
Constructing a neural network and training, wherein the neural network comprises an input layer, a hidden layer and an output layer; it is worth to be noted that the node number of the input layer is equal to the code length N, the input vector is the path information y to be decoded, the node number of the hidden layer is 8 × N, and the node of the output layerThe number of points is K-2; in this embodiment, a tensrflow platform is used to train a neural network, and when training neural network units, a training set is composed of 240000 groups of code words generated under different signal-to-noise ratios, each 120 code words is a batch, a learning rate is set to 0.001, and the maximum iteration number T is set to epoch Is 50.
Step two, forecasting
Inputting the path information y to be decoded into the neural network to obtain the maximum path number Z of each layer i Then, the maximum number of paths Z of each layer is determined i Inputting to a decoder; wherein, i belongs to (1, \8230;, K-2), and K is the number of information bits; it is worth noting that the maximum number of paths per layer is calculated according to the following formula:
wherein, W ij Is a weight, b j Offset, x j For each layer corresponding input value, σ is an activation function sigmoid,
wherein a = ∑ Σ j W ij x j +b j 。
Step three, decoder initialization
Let Q in decoder be L in width max Q is a list storing paths of each layer; initializing decoding paths and storage matrixes in Q and Q to be null; wherein, the decoding path in Q is
(i∈(1,…,N)；
Step four, expanding
In the binary information code tree of the decoder, if the ith level of the binary information code tree is a frozen bit, all decoding paths in Q are expanded into
The transition probability of the path is not changed, and the expanded decoding path is stored in Q andexecuting the step six; if the ith level of the binary tree of the information code is the information bit, reading the maximum path number Z corresponding to the layer i And Q is added to the front Z i The candidate path obtained by expanding the path is
And
and the transition probabilities of the candidate paths are different;
step five, sorting
Calculating the transition probability of the candidate paths, sequencing, and selecting L from big to small max Storing the candidate paths into Q; the calculation formula of the transition probability is as follows:
wherein the content of the first and second substances,
representing an input of u 1 The transition probability of a path of length i,
an odd term representing a decoding path of length i,
representing even terms of a coding path of length i,
is connected toThe received vector of code words is then transmitted to the receiver,
for the previous decoding path, u i Is the previous channel input.
The present embodiment ranks transition probabilities of candidate paths using a quick ranking policy. It is worth to state that L is selected from large to small max And the candidate paths are stored in the Q, so that the error paths are deleted in advance, and the performance superior to that of the traditional list decoding algorithm is realized.
Step six, deciding
And if the path length in the Q is the same as the code length, outputting the path with the maximum transfer probability in the Q as a decoding result, and otherwise, returning to the step four to continue decoding.
According to the polarization code self-adaptive decoding method, the search widths under different code lengths and code rates can be predicted by modifying parameters, scales and input vectors of a neural network, different channel environments and configuration requirements of a communication system can be met, and the polarization code self-adaptive decoding method has rich flexibility and configurability; the invention can effectively improve the communication performance and the data processing capacity of the system and can reduce the decoding complexity.
With reference to fig. 2, the adaptive decoding system for a polarization code of the present invention employs the above adaptive decoding method for a polarization code, and the system includes a decoding unit and a neural network unit, where the decoding unit is electrically connected to the neural network unit, the neural network unit is configured to calculate the maximum path number in each layer of the binary tree of information codes, and the decoding unit is configured to decode path information to be decoded. Specifically, the neural network unit calculates the maximum path number of each layer in the information code binary tree according to the path information to be decoded; it should be noted that the decoding unit is the above decoder, and the neural network unit of the present invention includes an input layer, a hidden layer, and an output layer; the node number of the input layer is equal to the code length N, the input vector is path information y to be decoded, the node number of the hidden layer is 8 × N, and the node number of the output layer is K-2.
Further, the decoding unit comprises a path expanding unit and a sequencing unitThe path extension unit is electrically connected with the neural network unit, and the sequencing unit is electrically connected with the path extension unit; the path expansion unit is used for reading the paths to be decoded stored in the Q, expanding the paths to be decoded in parallel and calculating corresponding transition probabilities, and then inputting the calculation results to the sorting unit; the sequencing unit sequences the transition probabilities from large to small and then selects L from large to small max Storing the candidate paths into Q; and deleting the rest paths and the corresponding transition probabilities thereof, and returning the rest paths and the corresponding transition probabilities to the path expansion unit. The self-adaptive decoding system for the polar code limits the path expansion by predicting the search width of each stage, thereby reducing the calculation complexity, and simultaneously improving the decoding performance by early interruption of an error path.
Example 2
With the decoding system shown in FIG. 3, the code length N is 8, the number of information bits K is 4, and the search width L is max And 4, the black nodes are the nodes accessed by the system, and the gray nodes are the nodes accessed by the traditional list decoding algorithm. The input vector of the neural network unit of this embodiment is a received codeword with a length of 8, and the output is a search width sequence with a length of 2, and is transmitted to the decoding unit. In the coding unit, when i =7, the corresponding Z 1 =1, so only one path is extended, for the same reason when i = 8. Compared with the traditional list decoding algorithm, the access node number of the system is obviously reduced, and the system has obvious advantages in the aspect of computational complexity.
As shown in fig. 4, the code length N =64 and the initial maximum search width L in fig. 4 max The decoding system performance of the invention is superior to the traditional SC and SCL decoders under various signal-to-noise ratios, and the performance of the decoding system of the invention is close to the theoretically optimal SCL decoder under high signal-to-noise ratio, and the decoding system of the invention has excellent performance.
Table 1 presents the search width comparison of the decoding system of the present invention with a conventional SCL decoder, where the code length N =64, the initial maximum search width L max =4, signal-to-noise ratio EbN0=1.0-4.0dB, routable because the computational complexity of the list polar decoder is positively correlated to the number of path extensionsThe path expansion number reflects the computational complexity of the design, and as can be seen from table 1, the decoding system complexity of the present invention is significantly reduced at various signal-to-noise ratios. The calculation shows that when the signal-to-noise ratio is 4.0dB, the calculation complexity of the decoding system is reduced by 67.51% compared with the traditional SCL decoding algorithm, so that the decoding system obviously reduces the calculation complexity of the algorithm and is particularly suitable for the condition of higher signal-to-noise ratio.
TABLE 1 comparison of path expansion numbers at different signal-to-noise ratios
| EbN0(dB) | 1.0 | 2.0 | 3.0 | 4.0 |
| |
4 | 4 | 4 | 4 |
| The invention relates to a decoding system | 1.6723 | 1.4987 | 1.3420 | 1.2998 |
In summary, the adaptive decoding method and system for the polar code of the present invention combine deep learning with the conventional polar code decoder, and limit the path expansion by predicting the search width of each stage, thereby reducing the computational complexity, and improving the decoding performance by early interruption of the wrong path. The embodiment presents its advantages in both complexity and performance, and its high hardware adaptability also shows great potential for practical applications.
The invention has been described in detail hereinabove with reference to specific exemplary embodiments thereof. It will, however, be understood that various modifications and changes may be made without departing from the scope of the invention as defined in the appended claims. The detailed description and drawings are to be regarded as illustrative rather than restrictive, and any such modifications and variations are intended to be included within the scope of the present invention as described herein. Furthermore, the background is intended to be illustrative of the state of the art as developed and the meaning of the present technology and is not intended to limit the scope of the invention or the application and field of application of the invention.
Claims (8)
1. A self-adaptive decoding method for a polarization code is characterized in that a neural network is constructed and trained, and path information to be decoded is input into the neural network to obtain the maximum path number of each layer of a binary tree of the information code; then, the maximum path number of each layer is input into a decoder, paths to be decoded are parallelly expanded according to the maximum path number of each layer to obtain candidate paths, then the transition probabilities of the candidate paths are calculated and sequenced, and a decoding result is obtained according to the path length and the maximum transition probability.
2. The adaptive decoding method of claim 1, comprising the steps of:
step one, constructing a neural network
Constructing a neural network and training, wherein the neural network comprises an input layer, a hidden layer and an output layer;
step two, forecasting
Inputting the path information y to be decoded into the neural network to obtain the maximum path of each layerRadial number Z i Then, the maximum number of paths Z of each layer is determined i Inputting to a decoder; wherein, i belongs to (1, \8230;, K-2), and K is the number of information bits;
step three, decoder initialization
Let Q in decoder be L max Q is a list for storing paths of each layer; initializing decoding paths and storage matrixes in Q and Q to be null; wherein the decoding path in Q is
Step four, expanding
In the binary information code tree of the decoder, if the ith level of the binary information code tree is a frozen bit, all decoding paths in Q are expanded into
Storing the expanded decoding path into Q and executing step six; if the ith level of the binary tree of the information code is the information bit, reading the maximum path number Z corresponding to the layer i And Q is added to the front Z i The candidate path obtained by expanding the path is
And
and the transition probabilities of the candidate paths are different;
step five, sorting
Calculating the transition probability of the candidate paths, sequencing the candidate paths, and selecting L from big to small max Storing the candidate paths into Q;
step six, deciding
And if the path length in Q is the same as the code length N, outputting the path with the maximum transfer probability in Q as a decoding result, and otherwise, returning to the step four.
3. The adaptive polar code decoding method according to claim 2, wherein the maximum number of paths per layer is calculated according to the following formula:
wherein, W ij Is a weight, b j To be offset, x j For each layer corresponding input value.
4. The adaptive decoding method of claim 2, wherein the transition probability is calculated as follows:
wherein the content of the first and second substances,
representing an input of u 1 The transition probability of a path of length i,
represents the odd terms of the decoding path of length i,
representing even terms of a coding path of length i,
in order to receive the vector of code words,
for the previous decoding path, u i For the previous channel transmissionAnd (3) adding.
5. The adaptive polar-code decoding method according to claim 2, wherein the node number of the input layer is equal to a code length N, the input vector is path information y to be decoded, the node number of the hidden layer is 8 × N, and the node number of the output layer is K-2.
6. A polarization code self-adaptive decoding method according to any one of claims 1 to 5, characterized in that the neural network is trained by using TensorFlow platform, wherein the maximum number of iterations T epoch Is 50.
7. A decoding system using the adaptive decoding method for the polarization code according to any one of claims 1 to 6, comprising a neural network unit and a decoding unit, wherein the neural network unit and the decoding unit are electrically connected; the neural network unit is used for calculating the maximum path number of each layer in the information code binary tree, and the decoding unit is used for decoding the path information to be decoded.
8. The adaptive polar code decoding system according to claim 7, wherein the decoding unit comprises a path expansion unit and an ordering unit, the path expansion unit and the ordering unit are electrically connected, and the path expansion unit is electrically connected to the neural network unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910476954.2A CN110212922B (en) | 2019-06-03 | 2019-06-03 | Polarization code self-adaptive decoding method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910476954.2A CN110212922B (en) | 2019-06-03 | 2019-06-03 | Polarization code self-adaptive decoding method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110212922A CN110212922A (en) | 2019-09-06 |
| CN110212922B true CN110212922B (en) | 2022-11-11 |
Family
ID=67790444
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910476954.2A Active CN110212922B (en) | 2019-06-03 | 2019-06-03 | Polarization code self-adaptive decoding method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110212922B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110798228A (en) * | 2019-10-29 | 2020-02-14 | 南京宁麒智能计算芯片研究院有限公司 | Polarization code turning decoding method and system based on deep learning |
| CN112332863B (en) * | 2020-10-27 | 2023-09-05 | 东方红卫星移动通信有限公司 | Polar code decoding algorithm, receiving end and system under low signal-to-noise ratio scene of low orbit satellite |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016168962A1 (en) * | 2015-04-20 | 2016-10-27 | 华为技术有限公司 | Decoding method and decoding apparatus for polar code |
| CN108777584A (en) * | 2018-07-06 | 2018-11-09 | 中国石油大学(华东) | A kind of fast Optimization of polarization code decoding parameter |
| CN109450456A (en) * | 2018-10-30 | 2019-03-08 | 南京大学 | A kind of adaptive storehouse interpretation method and system based on polarization code |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170076199A1 (en) * | 2015-09-14 | 2017-03-16 | National Institute Of Information And Communications Technology | Neural network system, and computer-implemented method of generating training data for the neural network |
-
2019
- 2019-06-03 CN CN201910476954.2A patent/CN110212922B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016168962A1 (en) * | 2015-04-20 | 2016-10-27 | 华为技术有限公司 | Decoding method and decoding apparatus for polar code |
| CN108777584A (en) * | 2018-07-06 | 2018-11-09 | 中国石油大学(华东) | A kind of fast Optimization of polarization code decoding parameter |
| CN109450456A (en) * | 2018-10-30 | 2019-03-08 | 南京大学 | A kind of adaptive storehouse interpretation method and system based on polarization code |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110212922A (en) | 2019-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108462558B (en) | Method and device for decoding polarization code SCL and electronic equipment | |
| Doan et al. | Neural successive cancellation decoding of polar codes | |
| CN109660264B (en) | High performance polar code decoding algorithm | |
| CN111294058B (en) | Channel coding and error correction decoding method, equipment and storage medium | |
| US20050229087A1 (en) | Decoding apparatus for low-density parity-check codes using sequential decoding, and method thereof | |
| CN107612560B (en) | Polarization code early iteration stopping method based on partial information bit likelihood ratio | |
| JP2004533766A (en) | Method for decoding a variable length codeword sequence | |
| CN107565978B (en) | BP decoding method based on Tanner graph edge scheduling strategy | |
| CN110233628B (en) | Self-adaptive belief propagation list decoding method for polarization code | |
| CN110212922B (en) | Polarization code self-adaptive decoding method and system | |
| CN111988045B (en) | Improved polarization code SCF decoder based on genetic algorithm | |
| CN108833052B (en) | Channel polarization decoding path metric value sorting method | |
| CN114070331B (en) | Self-adaptive serial offset list flip decoding method and system | |
| Ullah et al. | Low complexity bit reliability and predication based symbol value selection decoding algorithms for non-binary LDPC codes | |
| Lu et al. | Deep learning aided SCL decoding of polar codes with shifted-pruning | |
| CN112332864A (en) | Polar code decoding method and system for self-adaptive ordered mobile pruning list | |
| Hashemi et al. | A tree search approach for maximum-likelihood decoding of Reed-Muller codes | |
| Teng et al. | Convolutional neural network-aided bit-flipping for belief propagation decoding of polar codes | |
| Zhang et al. | A hybrid sphere decoding for short polar codes using variable step size | |
| CN114598334A (en) | Segmented CRC (cyclic redundancy check) assisted convolutional polarization code coding and decoding scheme | |
| Rao et al. | CNN-SC decoder for polar codes under correlated noise channels | |
| Azouaoui et al. | An efficient soft decoder of block codes based on compact genetic algorithm | |
| CN110855298A (en) | Low iteration number polarization code BP decoding method based on subchannel freezing condition | |
| Dhok et al. | ATRNN: Using seq2seq approach for decoding polar codes | |
| CN111835363A (en) | LDPC code decoding method based on alternative direction multiplier method |
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 |