Coding and decoding method, device, system and medium for polar code

Abstract

The present disclosure provides a coding and decoding method, device, information transmission system, and storage medium for polar codes, and relates to the field of communication technology, wherein the coding method comprises: sequentially obtaining a group of coded codewords generated after polar code encoding of each group of coded codewords of length N; wherein N is an integer power of 2; obtaining path information of a currently generated group of coded codewords in a decoding code tree, and using the path information as a check code for the next group of coded codewords; polar code encoding of the next group of information codewords to be encoded and a frozen codeword sequence containing the check code, and obtaining the next group of coded codewords corresponding to the next group of codewords to be encoded. The coding and decoding method, device, information transmission system, and storage medium disclosed in the present disclosure can improve the SCL decoding performance by using the feedforward path information as a check code to assist SCL decoding.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technology, and in particular to a coding and decoding method and device for polar codes, an information transmission system, and a storage medium. Background Art

[0002] In a symmetric binary channel, polarization codes (Polar codes) have been proven to have a capacity that can reach the Shannon limit. The channel polarization method varies depending on the channel type. The BEC (Binary Erasure Channel) channel has the simplest polarization method, the BSC (Binary Symmetric Channel) channel has a more complex polarization method, and the Gaussian channel uses a self-heuristic method, a density evolution polarization method, and a Gaussian approximation method. The decoding schemes for Polar codes mainly include two types: the serial cancellation (SC) decoding algorithm and the serial cancellation list (SCL) decoding algorithm. The SCL algorithm uses CRC and other auxiliary decoding methods to improve the decoding performance. Summary of the invention

[0003] In view of this, a technical problem to be solved by the present invention is to provide a coding and decoding method, device, information transmission system and storage medium for improving SCL decoding performance.

[0004] According to a first aspect of the present disclosure, a polar code encoding method is provided, comprising: sequentially obtaining a group of encoding codewords generated by performing polar code encoding on each group of to-be-encoded codewords of length N, wherein N is an integral power of 2; obtaining path information of a group of encoding codewords currently generated in a decoding code tree, and using the path information as a check code of a next group of encoding codewords; and performing polar code encoding on a next group of information codewords to be encoded and a frozen codeword sequence including the check code, to obtain a next group of encoding codewords corresponding to the next group of to-be-encoded codewords.

[0005] Optionally, a preset binary symbol sequence is obtained as an initial check code corresponding to a first group of information symbols to be encoded, wherein a length of the initial check code is greater than a length of the check code; and polar code encoding is performed on the first group of information symbols to be encoded including the initial check code and the frozen symbol sequence to obtain a first group of encoding codewords corresponding to the first group of symbols to be encoded.

[0006] Optionally, when encoding the last group of information symbols to be encoded, the check code is not generated; polar code encoding is performed on the last group of information symbols to be encoded and the frozen code symbol sequence containing the path information of the penultimate group of encoded codewords in the decoding code tree to obtain the last group of encoded codewords, and the encoding ends.

[0007] Optionally, the obtaining of path information of a currently generated set of encoding codewords in a decoding code tree includes: obtaining a path corresponding to the currently generated set of encoding codewords in the decoding code tree; obtaining a likelihood ratio, a transition probability ratio or an occurrence probability ratio corresponding to the last n information codeword nodes in the path, and calculating the path information based on the likelihood ratio, the transition probability ratio or the occurrence probability ratio.

[0008] Optionally, calculating the path information according to the likelihood ratio includes: among the last n information codeword nodes, starting from the information codeword node of the lowest layer, obtaining n likelihood ratios of the n-bit information codeword nodes; wherein, the likelihood ratio is processed based on a preset reserved bit number processing rule; setting a likelihood ratio interval, and if the likelihood ratio is within the likelihood ratio interval, retaining the likelihood ratio; if the likelihood ratio is less than or equal to the lower limit value of the likelihood ratio interval, setting the likelihood ratio to the lower limit value of the likelihood ratio, and if the likelihood ratio is greater than or equal to the upper limit value of the likelihood ratio interval, setting the likelihood ratio to the upper limit value of the interval; encoding the n likelihood ratios respectively to obtain n encoding information; and splicing the n encoding information to generate the path information.

[0009] Optionally, the calculation of the path information based on the transfer probability ratio includes: among the last n information codeword nodes, starting from the nodes in the lowest layer, calculating the ratios of the transfer probabilities with the information codeword nodes in the upper layer in turn according to the transfer probability, and obtaining n-1 ratios; wherein, the ratios are processed based on a preset reserved bit number processing rule; a ratio interval is set, and if the ratio is within the ratio interval, the ratio is retained; if the ratio is less than or equal to the lower limit value of the ratio interval, the ratio is set to the lower limit value of the interval, and if the ratio is greater than or equal to the upper limit value of the ratio interval, the ratio is set to the upper limit value of the interval; encoding each ratio separately to obtain n-1 encoding information; and concatenating the n-1 encoding information to generate the path information.

[0010] Optionally, the calculation of the path information based on the occurrence probability ratio includes: among the last n information codeword nodes, starting from the nodes in the lowest layer, calculating the ratios of the occurrence probabilities of the information codeword nodes in the upper layer in turn according to the occurrence probabilities, and obtaining n-1 ratios; wherein, the ratios are processed based on a preset reserved bit number processing rule; a ratio interval is set, and if the ratio is within the ratio interval, the ratio is retained; if the ratio is less than or equal to the lower limit value of the ratio interval, the ratio is set to the lower limit value of the interval, and if the ratio is greater than or equal to the upper limit value of the ratio interval, the ratio is set to the upper limit value of the interval; encoding each ratio separately to obtain n-1 encoding information; and concatenating the n-1 encoding information to generate the path information.

[0011] According to a second aspect of the present disclosure, a decoding method for polar codes is provided, including: performing SCL decoding on groups of coded codewords sent by a coding end in sequence, and obtaining multiple candidate paths corresponding to a current group of coded codewords in a decoding code tree; obtaining a check code corresponding to the current group of coded codewords in a decoding result of a previous group of coded codewords; wherein the check code is path information of the previous group of coded codewords in the decoding code tree; obtaining multiple candidate decoding results corresponding to the current group of coded codewords based on the multiple candidate paths corresponding to the current group of coded codewords; determining an alternative decoding result including the check code from the multiple candidate decoding results, and using the alternative path corresponding to the alternative decoding result as the decoding path of the current group of coded codewords.

[0012] Optionally, SCL decoding is performed on the first group of coded words sent by the encoding end, and multiple alternative paths corresponding to the first group of coded words are obtained in the decoding code tree; based on the multiple alternative paths corresponding to the first group of coded words, multiple alternative decoding results corresponding to the first group of coded words are obtained; an alternative decoding result including a preset initial check code is determined among the multiple alternative decoding results, and the alternative path corresponding to the alternative decoding result is used as the decoding path of the first group of coded code words; and path information of the decoding path of the first group of coded code words is generated as the check code when decoding the second group of coded code words.

[0013] Optionally, when the current group of encoding codewords is the last group of encoding codewords, a check code corresponding to the last group of encoding codewords is obtained in the decoding result of the penultimate group of encoding codewords; wherein, this check code is the path information of the penultimate group of encoding codewords in the decoding code tree; based on multiple alternative paths corresponding to the last group of encoding codewords, multiple alternative decoding results corresponding to the last group of encoding codewords are obtained; an alternative decoding result including this check code is determined among the multiple alternative decoding results, and the alternative path corresponding to this alternative decoding result is used as the decoding path of the last group of encoding codewords, and decoding is completed.

[0014] Optionally, the generation of path information of the decoding path of the first group of encoded codewords includes: obtaining the likelihood ratio, transition probability ratio or occurrence probability ratio corresponding to the last n information codeword nodes in the decoding path of the first group of encoded codewords, and calculating the path information based on the likelihood ratio, the transition probability ratio or the occurrence probability ratio.

[0015] Optionally, calculating the path information according to the likelihood ratio includes: among the last n information codeword nodes, starting from the information codeword node of the lowest layer, obtaining n likelihood ratios of the n-bit information codeword nodes; wherein, the likelihood ratio is processed based on a preset reserved bit number processing rule; setting a likelihood ratio interval, and if the likelihood ratio is within the likelihood ratio interval, retaining the likelihood ratio; if the likelihood ratio is less than or equal to the lower limit value of the likelihood ratio interval, setting the likelihood ratio to the lower limit value of the likelihood ratio, and if the likelihood ratio is greater than or equal to the upper limit value of the likelihood ratio interval, setting the likelihood ratio to the upper limit value of the interval; encoding the n likelihood ratios respectively to obtain n encoding information; and splicing the n encoding information to generate the path information.

[0016] Optionally, the calculation of the path information based on the transfer probability ratio includes: among the last n information codeword nodes, starting from the nodes in the lowest layer, calculating the ratios of the transfer probabilities with the information codeword nodes in the upper layer in turn according to the transfer probability, and obtaining n-1 ratios; wherein, the ratios are processed based on a preset reserved bit number processing rule; a ratio interval is set, and if the ratio is within the ratio interval, the ratio is retained; if the ratio is less than or equal to the lower limit value of the ratio interval, the ratio is set to the lower limit value of the interval, and if the ratio is greater than or equal to the upper limit value of the ratio interval, the ratio is set to the upper limit value of the interval; encoding each ratio separately to obtain n-1 encoding information; and concatenating the n-1 encoding information to generate the path information.

[0017] Optionally, the calculation of the path information based on the occurrence probability ratio includes: among the last n information codeword nodes, starting from the nodes in the lowest layer, calculating the ratios of the occurrence probabilities of the information codeword nodes in the upper layer in turn according to the occurrence probabilities, and obtaining n-1 ratios; wherein, the ratios are processed based on a preset reserved bit number processing rule; a ratio interval is set, and if the ratio is within the ratio interval, the ratio is retained; if the ratio is less than or equal to the lower limit value of the ratio interval, the ratio is set to the lower limit value of the interval, and if the ratio is greater than or equal to the upper limit value of the ratio interval, the ratio is set to the upper limit value of the interval; encoding each ratio separately to obtain n-1 encoding information; and concatenating the n-1 encoding information to generate the path information.

[0018] Optionally, if the candidate decoding result including the check code is not included in the multiple candidate decoding results, an alternative path with the highest probability of occurrence is selected as the decoding path, or decoding is stopped.

[0019] Optionally, a preset verification method is used to verify the verification code; wherein the verification code is verified using a parity or cyclic redundancy check code.

[0020] According to a third aspect of the present disclosure, a polar code encoding device is provided, including: a memory; and a processor coupled to the memory, wherein the processor is configured to execute the method described above based on instructions stored in the memory.

[0021] According to a fourth aspect of the present disclosure, a decoding device for polar codes is provided, including: a memory; and a processor coupled to the memory, wherein the processor is configured to execute the method described above based on instructions stored in the memory.

[0022] According to a fifth aspect of the present disclosure, an information transmission system is provided, including: an information transmitting end, including the encoding device for polar code as described above; an information receiving end, including the decoding device for polar code as described above.

[0023] Optionally, channels for sending and receiving information include: a BEC channel, a BSC channel and a Gaussian channel.

[0024] According to a sixth aspect of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the instructions are executed by a processor as described above.

[0025] The polar code encoding and decoding method, device, information transmission system, and storage medium disclosed in the present invention use the path information of the encoding codeword in the decoding code tree as the check code, perform polar code encoding together with the check code and the information codeword, and compare the check code obtained by decoding with the path information of the candidate codeword obtained during decoding to select the correct decoding path, thereby improving the SCL decoding performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0027] Figure 1 is a flow chart of an embodiment of a polar code encoding method according to the present disclosure;

[0028] Figure 2 is a schematic diagram of a flow chart of calculating path information in one embodiment of a polar code encoding method according to the present disclosure;

[0029] Figure 3 Schematic diagram of the decoding structure of polar code;

[0030] Figure 4 is a coding schematic diagram of an embodiment of a coding method for polar codes according to the present disclosure;

[0031] Figure 5 is a flow chart of an embodiment of a decoding method for polar codes according to the present disclosure;

[0032] Figure 6 is a decoding schematic diagram of an embodiment of a decoding method for polar codes according to the present disclosure;

[0033] Figure 7 1 is a schematic diagram of modules of an embodiment of a polar code encoding device according to the present disclosure;

[0034] Figure 8 1 is a schematic diagram of a module of an embodiment of a decoding device for polar codes according to the present disclosure;

[0035] Fig. 9 It is a module diagram of an embodiment of an information transmission system according to the present disclosure. DETAILED DESCRIPTION

[0036] The present disclosure is described more fully below with reference to the accompanying drawings, in which exemplary embodiments of the present disclosure are described. The technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure. The technical solutions of the present disclosure are described in many aspects below in conjunction with various figures and embodiments.

[0037] The polarization code encoding formula of Polar code is:

[0038] x1 N =u1 N G N (1-1);

[0039] Among them, u1 N is the original codeword sequence, x1 N is the coded codeword sequence, G N is the generator matrix, the code length is N = 2 n , n≥0.

[0040] The polar code encoding formula can also be:

[0041]

[0042] Among them, G N (λ) is the matrix G N The submatrix consisting of the rows corresponding to the indices in the set λ selected from G corresponds to the channel coding selected for transmitting the information symbol. N (λc) from the matrix G N The submatrix consisting of the rows corresponding to the indices in the set λc is selected from the set λc; when encoding, λ and u λ c is fixed, λ c Generally, all are set to 0. Information code element sequence u λ As a free variable, with u λ The change can be obtained from u λ to x1 N The encoding.

[0043] The polar code encoding formula assisted by the check code is:

[0044]

[0045] Among them, u λ cc It is a check code symbol sequence, occupying non-frozen code symbol positions.

[0046] Polar code decoding algorithms include SC decoding algorithms and SCL decoding algorithms. When decoding, the SC decoding algorithm calculates the probability of occurrence of the first codeword based on the physical channel parameters and the likelihood ratio (LR) value. The likelihood ratio value is the ratio of the transfer probabilities calculated according to the existing SC decoding algorithm, and the ratio is the likelihood number. The path with a high probability of occurrence is used as the decoding result of the first codeword, and then the probability of occurrence of the second codeword is calculated based on the probability of occurrence of the first codeword and the likelihood ratio. The path with a high probability of occurrence is used as the decoding result of the second codeword, and so on, until the decoding of the codeword is completed. The SCL decoding algorithm uses the L paths with the highest probability of occurrence as alternative paths, and selects the path with the highest probability of occurrence as the decoding result after the decoding of the codeword is completed.

[0047] Figure 1 FIG. 1 is a flow chart of an embodiment of a method for encoding polar codes according to the present disclosure, as shown in FIG. Figure 1 As shown:

[0048] Step 101, sequentially obtaining a group of coded codewords generated by performing polar code encoding on each group of to-be-coded codewords of length N, wherein N is an integral power of 2.

[0049] Step 102, obtaining the path information of a group of currently generated encoding code words in the decoding code tree, and using the path information as the check code of the next group of encoding code words.

[0050] Step 103: Perform polar code encoding on the next group of information symbols to be encoded and the frozen symbol sequence including the check code to obtain a next group of encoding codewords corresponding to the next group of information symbols to be encoded.

[0051] A preset binary symbol sequence is obtained as an initial check code corresponding to a first group of information symbols to be encoded, wherein the length of the initial check code is greater than the length of the check code. Polar code encoding is performed on the first group of information symbols to be encoded including the initial check code and the frozen symbol sequence to obtain a first group of encoding codewords corresponding to the first group of symbols to be encoded.

[0052] Figure 2 FIG. 1 is a schematic diagram of a flow chart of calculating path information in an embodiment of a polar code encoding method according to the present disclosure, as shown in FIG. Figure 2 As shown:

[0053] Step 201, obtaining a path corresponding to a set of currently generated encoding codewords in a decoding code tree.

[0054] Step 202, obtain the likelihood ratio, transition probability ratio or occurrence probability ratio corresponding to the last n information symbol nodes in the path, and calculate the path information according to the likelihood ratio, transition probability ratio or occurrence probability ratio.

[0055] A variety of methods can be used to calculate path information based on transition probabilities. In the last n information symbol nodes, starting from the lowest level of information symbol nodes, n likelihood ratios of n information symbol nodes are obtained, and the likelihood ratios can be calculated using existing methods. n is a natural number greater than 2, such as 3, 4, 5, etc.

[0056] The ratio is processed based on a preset number of digits to be retained processing rule. There may be multiple number of digits to be retained processing rules, for example, when the likelihood ratio is greater than or equal to 0.1 and less than 1, two decimal places are retained; when the likelihood ratio is greater than or equal to 1 and less than or equal to 10, one decimal place is retained.

[0057] In one embodiment, Figure 3 As shown, the likelihood ratio can be numerically expressed as LR 8,1 =0.03 / 0.02=1.5,LR 8,2 =0.10 / 0.05=2,LR 8,3 =0.07 / 0.03=2.333333,LR 8,4 =0.20 / 0.05=4,LR 8,5 =0.10 / 0.02=5,LR 8,6 =0.01 / 0.02=0.5,LR 8,7 =0.18 / 0.02=9,LR 8,8 =0.04 / 0.01=5.

[0058] Set the likelihood ratio interval. If the likelihood ratio is within the likelihood ratio interval, keep this likelihood ratio. If the likelihood ratio is less than or equal to the interval lower limit of the likelihood ratio interval, set this likelihood ratio to the likelihood ratio lower limit. If the likelihood ratio is greater than or equal to the interval upper limit of the likelihood ratio interval, set this likelihood ratio to the interval upper limit.

[0059] The probability of occurrence is Figure 3 For example, the probability of the two nodes on the leftmost side of the lowest layer in the decoding code tree is 0.03 and 0.02. Obtain the path corresponding to the currently generated set of encoding codewords 00001010 in the decoding code tree ( Figure 3 In the last 3 information symbol nodes of this path, starting from the lowest level node, the likelihood ratios of the last 3 information symbol nodes are: LR 8,4 =4,LR 7,2 =0.4,LR 5,1 =0.714286.

[0060] The likelihood ratio interval can be 0.10-9.9. If the likelihood ratio is within 0.10-9.9, the likelihood ratio is retained. If the likelihood ratio is less than or equal to the lower limit of the likelihood ratio interval, 0.10, the likelihood ratio is set to 0.10. If the likelihood ratio is greater than or equal to the upper limit of the likelihood ratio interval, the likelihood ratio is set to the upper limit of the interval, 9.9. Encode the n likelihood ratios separately to obtain n encoding information; concatenate the n encoding information to generate path information.

[0061] The coded information may be binary coded information, etc. A variety of binary coding methods may be used. For example, when the likelihood ratio is 0.10<=ratio<0.99, the two digits after the decimal point of the likelihood ratio are taken as the true value, and binary coding is performed with 1 as the prefix. For example, if the likelihood ratio is 0.21, the two digits after the decimal point of 0.21 are taken as 21, and the binary coding of 21 is 0010101, and the binary coding of 0.21 is 10010101.

[0062] When the likelihood ratio is 1.0 <= ratio <= 9.9, the two digits after the decimal point are taken as the true value for binary encoding, with a prefix of 0. For example, if the likelihood ratio is 2.1, the two digits after the decimal point of 2.1 are 21, the binary encoding of 21 is 0010101, and the binary encoding of 0.21 is 00010101.

[0063] In one embodiment, the likelihood ratio is Where LR is the likelihood ratio, W(y|0) is the transition probability of decoding to 0, and W(y|1) is the transition probability of decoding to 1.

[0064] In the last n information symbol nodes, starting from the nodes in the lowest layer, the ratios of the transition probabilities with the information symbol nodes in the upper layer are calculated in turn according to the transition probabilities, and n-1 ratios are obtained, and the ratios are processed based on the preset reserved digit processing rules. For example, when the ratio is greater than or equal to 0.1 and less than 1, two decimal places are retained; when the ratio is greater than or equal to 1 and less than or equal to 10, one decimal place is retained.

[0065] Set the ratio interval. If the ratio is within the ratio interval, keep the ratio; if the ratio is less than or equal to the lower limit of the ratio interval, set the ratio to the lower limit; if the ratio is greater than or equal to the upper limit of the ratio interval, set the ratio to the upper limit. Encode each ratio separately to obtain n-1 encoding information; concatenate the n-1 encoding information to generate path information.

[0066] In one embodiment, the existing method can be used to calculate the transition probability of each information symbol node. Obtain the path corresponding to the currently generated set of encoding codewords 00001010 in the decoding code tree ( Figure 3 In the last 4 information symbol nodes of this path, starting from the node of the second-to-last layer, the ratio of the transition probability with the node of the previous layer is calculated in sequence according to the transition probability to obtain 2 ratios. The method of encoding the ratio of the two transition probabilities and generating path information is the same as the method of encoding the likelihood ratio and generating path information as described above.

[0067] In one embodiment, among the last n information symbol nodes, starting from the lowest layer of nodes, the ratios of the probability of occurrence of the information symbol nodes in the upper layer are calculated in turn according to the probability of occurrence, and n-1 ratios are obtained, and the ratios are processed based on the preset reserved digit processing rules. For example, when the ratio is greater than or equal to 0.1 and less than 1, two decimal places are retained; when the ratio is greater than or equal to 1 and less than or equal to 10, one decimal place is retained.

[0068] Set the ratio interval. If the ratio is within the ratio interval, keep the ratio. If the ratio is less than or equal to the lower limit of the ratio interval, set the ratio to the lower limit. If the ratio is greater than or equal to the upper limit of the ratio interval, set the ratio to the upper limit. Encode each ratio separately to obtain n-1 encoding information. Concatenate the n-1 encoding information to generate path information.

[0069] In one embodiment, the existing method can be used to calculate the occurrence probability of each information symbol node. Obtain the path corresponding to the currently generated set of encoding codewords 00001010 in the decoding code tree ( Figure 3 ). In the last 3 information symbol nodes of this path, starting from the node of the last layer, the ratio of the occurrence probability of the node in the previous layer is calculated in sequence according to the occurrence probability to obtain 2 ratios. The method of encoding the ratio of the two occurrence probabilities and generating path information is the same as the method of encoding the likelihood ratio and generating path information as described above.

[0070] In one embodiment, Figure 4 As shown, when encoding the polar code, the information symbol u to be encoded is λ,i The binary code u of the likelihood ratio of the last n bits of the decoding code tree λ,i cc Arrange in u λ,i+1The initial check code element u output by the likelihood ratio check code generator is used to encode the polar code together with the frozen code element sequence, 1<=i<=N. λ,0 cc Arranged in the first group of information codeword sequence u to be encoded λ,1 In the preset position, and freeze the code element sequence u λ c After polarization coding, the first group of coded codewords x is output 1,1 N The initial check code element can be any binary code, usually a code pattern of alternating 01, such as 010101…01. λ,i , the likelihood ratio check code generator does not generate a check code.

[0071] Figure 5 FIG. 1 is a flow chart of an embodiment of a decoding method for polar codes according to the present disclosure, as shown in FIG. Figure 5 As shown:

[0072] Step 501 , sequentially perform SCL decoding on each group of coded codewords sent by the coding end, and obtain multiple candidate paths corresponding to the current group of coded codewords in the decoding code tree.

[0073] Step 502: Obtain a check code corresponding to the current group of encoded code words from the decoding result of the previous group of encoded code words; wherein the check code is the path information of the previous group of encoded code words in the decoding code tree.

[0074] Step 503 , based on multiple candidate paths corresponding to the current group of encoded codewords, obtain multiple candidate decoding results corresponding to the current group of encoded codewords.

[0075] Step 504: determine an alternative decoding result including a check code from among the multiple alternative decoding results, and use an alternative path corresponding to the alternative decoding result as a decoding path for the current group of coded codewords.

[0076] For example, each group of coded codewords sent by the encoder is sequentially SCL decoded, and multiple candidate paths corresponding to the current group of coded codewords are obtained in the decoding code tree. In the decoding result of the previous group of coded codewords, a check code corresponding to the current group of coded codewords is obtained; wherein the check code is the path information of the previous group of coded codewords in the decoding code tree, and the path information can be a set of binary codes, for example, the path information is 01011101.

[0077] Based on multiple candidate paths corresponding to the current group of encoding codewords, multiple candidate decoding results corresponding to the current group of encoding codewords are obtained. Based on the structure of the candidate decoding results, it can be determined whether the check code position of the candidate decoding results contains an initial check code or a check code in the multiple candidate decoding results. If it does, the candidate decoding result is selected.

[0078] The multiple candidate decoding results can be multiple binary codes, for example, one candidate decoding result is 10101111 ... 0101011101. It is determined that the check code position of the candidate decoding result "10101111 ... 0101011101" among the multiple candidate decoding results contains the check code "01011101", and the candidate path corresponding to the candidate decoding result "10101111 ... 0101011101" is used as the decoding path of the current group of coded codewords for decoding.

[0079] In one embodiment, the first group of code words sent by the encoder are subjected to SCL decoding, and multiple candidate paths corresponding to the first group of code words are obtained in the decoding code tree. Based on the multiple candidate paths corresponding to the first group of code words, multiple candidate decoding results corresponding to the first group of code words are obtained.

[0080] The initial check code of the encoding end and the decoding end is the same. The alternative decoding result containing the preset initial check code is determined in the check code position of multiple alternative decoding results, and the alternative path corresponding to the alternative decoding result is used as the decoding path of the first group of encoded code words. The path information of the decoding path of the first group of encoded code words is generated as the check code when the second group of encoded code words are decoded. The path information generation method of the encoding end and the decoding end is the same.

[0081] When the current group of encoding codewords is the last group of encoding codewords, a check code corresponding to the last group of encoding codewords is obtained in the decoding result of the penultimate group of encoding codewords; wherein the check code is the path information of the penultimate group of encoding codewords in the decoding code tree. Based on multiple alternative paths corresponding to the last group of encoding codewords, multiple alternative decoding results corresponding to the last group of encoding codewords are obtained. An alternative decoding result including the check code is determined among the multiple alternative decoding results, and the alternative path corresponding to the alternative decoding result is used as the decoding path of the last group of encoding codewords, and the decoding ends.

[0082] In one embodiment, the decoding code tree of the encoding end and the decoding end is the same, and the method for obtaining the path information of the encoding end and the decoding end is the same. Generate the path information of the decoding path of the first group of encoding codewords: obtain the likelihood ratio, transition probability ratio or occurrence probability ratio corresponding to the last n information codeword nodes in the decoding path of the first group of encoding codewords, and calculate the path information according to the likelihood ratio, transition probability ratio or occurrence probability ratio.

[0083] Calculate path information based on likelihood ratio: among the last n information symbol nodes, starting from the lowest level of information symbol nodes, obtain n likelihood ratios of n information symbol nodes, and process the likelihood ratios based on a preset reserved bit number processing rule. The specific processing method is the same as that during encoding.

[0084] Set the likelihood ratio interval. If the likelihood ratio is within the likelihood ratio interval, keep the likelihood ratio. If the likelihood ratio is less than or equal to the lower limit of the likelihood ratio interval, set the likelihood ratio to the lower limit. If the likelihood ratio is greater than or equal to the upper limit of the likelihood ratio interval, set the likelihood ratio to the upper limit. Encode the n likelihood ratios separately to obtain n encoding information. Concatenate the n encoding information to generate path information.

[0085] Calculate the path information based on the transition probability ratio: among the last n information codeword nodes, starting from the nodes in the lowest layer, calculate the ratio of the transition probability with the information codeword nodes in the upper layer in turn according to the transition probability, and obtain n-1 ratios. Process this ratio based on the preset reserved bit number processing rule. The specific processing method is the same as the processing method during encoding.

[0086] Set the ratio interval. If the ratio is within the ratio interval, keep the ratio. If the ratio is less than or equal to the lower limit of the ratio interval, set the ratio to the lower limit. If the ratio is greater than or equal to the upper limit of the ratio interval, set the ratio to the upper limit. Encode each ratio separately to obtain n-1 encoding information. Concatenate the n-1 encoding information to generate path information.

[0087] Calculate the path information based on the probability of occurrence ratio: among the last n information codeword nodes, starting from the nodes in the lowest layer, calculate the ratio of the probability of occurrence of the information codeword nodes in the upper layer in turn according to the probability of occurrence, and obtain n-1 ratios. Process this ratio based on the preset reserved bit number processing rule. The specific processing method is the same as the processing method during encoding.

[0088] Set the ratio interval. If the ratio is within the ratio interval, keep the ratio. If the ratio is less than or equal to the lower limit of the ratio interval, set the ratio to the lower limit. If the ratio is greater than or equal to the upper limit of the ratio interval, set the ratio to the upper limit. Encode each ratio separately to obtain n-1 encoding information. Concatenate the n-1 encoding information to generate path information.

[0089] If there is no candidate decoding result containing the check code among the multiple candidate decoding results, the candidate path with the highest probability of occurrence is selected as the decoding path, or the decoding is stopped. To ensure the accuracy of the verification, the check code can be re-verified using a preset verification method, for example, the check code is re-verified using a parity or cyclic redundancy check code.

[0090] In one embodiment, Figure 6 As shown, the SCL decoder receives the input coded codeword y 1,1 N ,y 1,2 N ,y 1,3 N ,…,y 1,w N , and perform SCL decoding. The likelihood ratio check generator outputs the check code based on the decoding result Among them, the initial check code and the initial check code u at the encoding end λ,0 cc same.

[0091] Figure 7 FIG. 1 is a schematic diagram of a module of another embodiment of a polar code encoding device according to the present disclosure. Figure 7 As shown, the device may include a memory 71, a processor 72, a communication interface 73, and a bus 74. The memory 71 is used to store instructions, the processor 72 is coupled to the memory 71, and the processor 72 is configured to implement the above-mentioned encoding method for polar code based on the instructions stored in the memory 71.

[0092] The memory 71 may be a high-speed RAM memory, a non-volatile memory, etc. The memory 71 may also be a memory array. The memory 71 may also be divided into blocks, and the blocks may be combined into virtual volumes according to certain rules. The processor 72 may be a central processing unit CPU, or an application specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the encoding method for polar codes disclosed in the present invention.

[0093] Figure 8 FIG. 1 is a schematic diagram of a module of an embodiment of a decoding device for polar codes according to the present disclosure. Figure 8 As shown, the device may include a memory 81, a processor 82, a communication interface 83 and a bus 84. The memory 81 is used to store instructions, the processor 82 is coupled to the memory 81, and the processor 82 is configured to execute the above-mentioned decoding method for polar code based on the instructions stored in the memory 81.

[0094] The memory 81 may be a high-speed RAM memory, a non-volatile memory, etc. The memory 81 may also be a memory array. The memory 81 may also be divided into blocks, and the blocks may be combined into virtual volumes according to certain rules. The processor 82 may be a central processing unit CPU, or an application specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the decoding method for polar codes disclosed in the present invention.

[0095] In one embodiment, Fig. 9 As shown, the present disclosure provides an information transmission system, including: an information transmitting end 91 and an information receiving end 92, the information transmitting end 91 includes an encoding device for polarization code in any of the above embodiments; the information receiving end 92 includes a decoding device for polarization code in any of the above embodiments. Channels for information transmission and information reception include BEC channels, BSC channels, etc.

[0096] In one embodiment, the present disclosure provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and when the instructions are executed by a processor, the encoding method for polar codes and the decoding method for polar codes in any of the above embodiments are implemented.

[0097] In the polar code encoding and decoding method, device, information transmission system, and storage medium in the above-mentioned embodiments, path information of the encoding codeword in the decoding code tree is used as a check code, the check code and the information codeword are encoded together as polar codes, and the check code obtained by decoding is compared with the path information of the candidate codeword obtained during decoding to select the correct decoding path, thereby improving the SCL decoding performance.

[0098] The method and system of the present disclosure may be implemented in many ways. For example, the method and system of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present disclosure may also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present disclosure. Therefore, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.

[0099] The description of the present disclosure is given for the purpose of illustration and description, and is not intended to be exhaustive or to limit the present disclosure to the disclosed form. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present disclosure, and to enable those of ordinary skill in the art to understand the present disclosure and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. A coding method for polar codes, comprising: Obtaining in sequence a group of coded codewords generated by performing polar code encoding on each group of to-be-coded codewords of length N, wherein N is an integral power of 2; Obtaining path information of a currently generated set of encoding codewords in the decoding code tree, and using the path information as a check code for the next set of encoding codewords; The step of obtaining path information of a set of currently generated encoding codewords in a decoding code tree includes: Obtaining a path corresponding to the currently generated set of encoding codewords in the decoding code tree; obtaining a likelihood ratio, a transition probability ratio, or an occurrence probability ratio corresponding to the last n information symbol nodes in the path, and calculating the path information according to the likelihood ratio, the transition probability ratio, or the occurrence probability ratio; Polar code encoding is performed on a next group of information symbols to be encoded and a frozen symbol sequence including the check code to obtain a next group of encoding codewords corresponding to the next group of information symbols to be encoded.

2. The method of claim 1, further comprising: Obtaining a preset binary codeword sequence as an initial check code corresponding to the first group of information codewords to be encoded; wherein the length of the initial check code is greater than the length of the check code; Polar code encoding is performed on the first group of information symbols to be encoded including the initial check code and the frozen symbol sequence to obtain a first group of encoded codewords corresponding to the first group of information symbols to be encoded.

3. The method of claim 1, comprising: When encoding the last group of information symbols to be encoded, the check code is not generated; Polar code encoding is performed on the last group of information symbols to be encoded and the frozen symbol sequence containing the path information of the penultimate group of encoded codewords in the decoding code tree to obtain the last group of encoded codewords, and the encoding is completed.

4. The method according to claim 3, wherein calculating the path information according to the likelihood ratio comprises: Among the last n information symbol nodes, starting from the lowest layer of information symbol nodes, n likelihood ratios of the n information symbol nodes are obtained; wherein the likelihood ratios are processed based on a preset reserved bit number processing rule; Setting a likelihood ratio interval, if the likelihood ratio is within the likelihood ratio interval, retaining the likelihood ratio; if the likelihood ratio is less than or equal to the interval lower limit value of the likelihood ratio interval, setting the likelihood ratio to the likelihood ratio lower limit value; if the likelihood ratio is greater than or equal to the interval upper limit value of the likelihood ratio interval, setting the likelihood ratio to the interval upper limit value; Encoding the n likelihood ratios respectively to obtain n coded information; The n pieces of encoding information are concatenated to generate the path information.

5. The method according to claim 3, wherein the step of calculating the path information according to the transition probability ratio comprises: Among the last n information symbol nodes, starting from the nodes at the lowest level, the ratios of the transition probabilities with the information symbol nodes at the upper level are calculated in sequence according to the transition probabilities to obtain n-1 ratios; wherein the ratios are processed based on a preset reserved bit number processing rule; Setting a ratio interval, if the ratio is within the ratio interval, retaining the ratio; if the ratio is less than or equal to the lower limit of the ratio interval, setting the ratio to the lower limit; if the ratio is greater than or equal to the upper limit of the ratio interval, setting the ratio to the upper limit; Encode each ratio respectively to obtain n-1 encoding information; The n-1 pieces of coded information are concatenated to generate the path information.

6. The method according to claim 1, wherein the step of calculating the path information according to the occurrence probability ratio comprises: Among the last n information symbol nodes, starting from the nodes at the lowest level, the ratios of the occurrence probabilities of the information symbol nodes at the upper level are calculated in sequence according to the occurrence probabilities to obtain n-1 ratios; wherein the ratios are processed based on a preset reserved bit number processing rule; Setting a ratio interval, if the ratio is within the ratio interval, retaining the ratio; if the ratio is less than or equal to the lower limit of the ratio interval, setting the ratio to the lower limit; if the ratio is greater than or equal to the upper limit of the ratio interval, setting the ratio to the upper limit; Encode each ratio respectively to obtain n-1 encoding information; The n-1 pieces of coded information are concatenated to generate the path information.

7. A decoding method for polar codes, comprising: Perform SCL decoding on each group of coded codewords sent by the coding end in turn, and obtain multiple candidate paths corresponding to the current group of coded codewords in the decoding code tree; In the decoding result of the previous group of encoded codewords, a check code corresponding to the current group of encoded codewords is obtained; wherein the check code is the path information of the previous group of encoded codewords in the decoding code tree; Based on the multiple candidate paths corresponding to the current group of encoded codewords, obtaining multiple candidate decoding results corresponding to the current group of encoded codewords; Determine an alternative decoding result including the check code from among multiple alternative decoding results, and use an alternative path corresponding to the alternative decoding result as a decoding path for the current group of coded codewords; Wherein, the path information of the decoding path of the first group of coded codewords sent by the encoding end is generated as a check code when the second group of coded codewords are decoded; wherein the path information of the decoding path of the first group of coded codewords is generated includes: Obtain the likelihood ratio, transition probability ratio or occurrence probability ratio corresponding to the last n information codeword nodes in the decoding path of the first group of encoded codewords, and calculate the path information based on the likelihood ratio, the transition probability ratio or the occurrence probability ratio.

8. The method of claim 7, wherein: performing SCL decoding on the first group of encoded codewords to obtain a plurality of candidate paths corresponding to the first group of encoded codewords in a decoding code tree; Based on the multiple candidate paths corresponding to the first group of encoded codewords, obtaining multiple candidate decoding results corresponding to the first group of encoded codewords; An alternative decoding result including a preset initial check code is determined from a plurality of alternative decoding results, and an alternative path corresponding to the alternative decoding result is used as a decoding path of the first group of encoded codewords.

9. The method according to claim 8, wherein: When the current group of encoding codewords is the last group of encoding codewords, in the decoding result of the penultimate group of encoding codewords, a check code corresponding to the last group of encoding codewords is obtained; wherein the check code is the path information of the penultimate group of encoding codewords in the decoding code tree; Based on the multiple candidate paths corresponding to the last set of encoded codewords, obtaining multiple candidate decoding results corresponding to the last set of encoded codewords; An alternative decoding result including the check code is determined from among multiple alternative decoding results, and an alternative path corresponding to the alternative decoding result is used as a decoding path for the last group of encoded codewords, and decoding is terminated.

10. The method according to claim 8, wherein calculating the path information according to the likelihood ratio comprises: Among the last n information symbol nodes, starting from the lowest layer of information symbol nodes, n likelihood ratios of the n information symbol nodes are obtained; wherein the likelihood ratios are processed based on a preset reserved bit number processing rule; Setting a likelihood ratio interval, if the likelihood ratio is within the likelihood ratio interval, retaining the likelihood ratio; if the likelihood ratio is less than or equal to the interval lower limit value of the likelihood ratio interval, setting the likelihood ratio to the likelihood ratio lower limit value; if the likelihood ratio is greater than or equal to the interval upper limit value of the likelihood ratio interval, setting the likelihood ratio to the interval upper limit value; Encoding the n likelihood ratios respectively to obtain n coded information; The n pieces of encoding information are concatenated to generate the path information.

11. The method according to claim 8, wherein calculating the path information according to the transition probability ratio comprises: Among the last n information symbol nodes, starting from the nodes at the lowest level, the ratios of the transition probabilities with the information symbol nodes at the upper level are calculated in sequence according to the transition probabilities to obtain n-1 ratios; wherein the ratios are processed based on a preset reserved bit number processing rule; Setting a ratio interval, if the ratio is within the ratio interval, retaining the ratio; if the ratio is less than or equal to the lower limit of the ratio interval, setting the ratio to the lower limit; if the ratio is greater than or equal to the upper limit of the ratio interval, setting the ratio to the upper limit; Encode each ratio respectively to obtain n-1 encoding information; The n-1 pieces of coded information are concatenated to generate the path information.

12. The method according to claim 8, wherein the step of calculating the path information according to the occurrence probability ratio comprises: Among the last n information symbol nodes, starting from the nodes at the lowest level, the ratios of the occurrence probabilities of the information symbol nodes at the upper level are calculated in sequence according to the occurrence probabilities to obtain n-1 ratios; wherein the ratios are processed based on a preset reserved bit number processing rule; Setting a ratio interval, if the ratio is within the ratio interval, retaining the ratio; if the ratio is less than or equal to the lower limit of the ratio interval, setting the ratio to the lower limit; if the ratio is greater than or equal to the upper limit of the ratio interval, setting the ratio to the upper limit; Encode each ratio respectively to obtain n-1 encoding information; The n-1 pieces of coded information are concatenated to generate the path information.

13. The method of claim 8, further comprising: If the candidate decoding result of the check code is not included in the multiple candidate decoding results, the candidate path with the highest probability of occurrence is selected as the decoding path, or the decoding is stopped.

14. The method of claim 8, comprising: The check code is checked using a preset check method; wherein the check code is checked using a parity or cyclic redundancy check code.

15. A coding device for polar codes, comprising: Memory; and a processor coupled to the memory, wherein the processor is configured to execute the method according to any one of claims 1 to 6 based on instructions stored in the memory.

16. A decoding device for polar codes, comprising: Memory; and a processor coupled to the memory, the processor being configured to execute the method according to any one of claims 7 to 14 based on instructions stored in the memory.

17. An information transmission system, comprising: An information transmitting end, comprising the encoding device for polarization code as claimed in claim 15; An information receiving end comprises the decoding device for polar codes as claimed in claim 16.

18. The system of claim 17, wherein: The channels for sending and receiving information include: BEC channel, BSC channel and Gaussian channel.

19. A computer-readable storage medium storing computer instructions, wherein the instructions are executed by a processor to execute the method according to any one of claims 1 to 7 and / or 8 to 14.

Images