A method for constructing and decoding spatially coupled polar codes in AWGN channels

Abstract

This invention belongs to the field of communication channel coding technology, specifically relating to a method for constructing and decoding spatially coupled polar codes under AWGN channels, comprising the following steps executed sequentially: S1: Obtaining a reliability distribution sequence adapted to the AWGN channel using a Gaussian approximation algorithm; S2: Using the bit error rate and frame error rate of the first iteration as a baseline, calculating the performance improvement score under different coupling ratios, and selecting the ratio with the highest performance improvement score as the optimal coupling ratio; S3: Generating a spatially coupled structure based on the optimal coupling ratio and the reliability distribution sequence; S4: Obtaining system operating parameters based on time aggregation weight optimization of normalized kurtosis score; S5: Performing enhanced iterative decoding and outputting the decoding result; introducing quality indicators to optimize weight coefficients fully exploits the temporal correlation between different iteration stages.

Description

Technical Field

[0001] This invention belongs to the field of communication channel coding technology, specifically relating to a method for constructing and decoding spatially coupled polar codes in AWGN channels. Background Technology

[0002] Polar codes, as the first channel coding scheme theoretically proven to achieve Shannon capacity, have been selected as the control channel coding scheme for 5G eMBB scenarios. To further improve the performance of polar codes with finite code lengths, spatially coupled polar codes (SC-PC) have been proposed. SC-PC couples multiple independent polar code blocks in space, utilizing the bridging effect of this coupling to allow reliable information to propagate between code blocks, thereby improving overall decoding performance.

[0003] However, despite the theoretical performance advantages of SC-PC, existing construction schemes still face many challenges in practical applications for AWGN channels:

[0004] Early spatial coupling schemes were primarily designed for binary erased channels (BEC), mainly employing an information-to-frozen coupling strategy, mapping the information bits of the previous code block to the frozen bits of the current code block. While research indicates an empirically optimal coupling ratio of approximately 0.33 for this specific structure, this conclusion is based on specific channels (such as BEC) and specific mapping rules, making it difficult to directly generalize to AWGN channels. Furthermore, the frozen bits themselves are typically fixed at 0, limiting information carrying capacity. In addition, some schemes utilize the parity bits of the system polar code for coupling, which, while providing error detection capabilities, offers limited direct improvement to error correction performance, and the coupling object is singular, failing to fully utilize the channel polarization characteristics of the polar code.

[0005] Subsequent solutions, while focusing on performance optimization in AWGN channels, employed coupling mechanisms based on information bit sharing and puncturing (such as Partial Information Coupling (PIC) and Partial Information Coupling Bit Interleaved Polar Coded Modulation (PIC-BIPCM)). These methods achieve association by copying and puncturing partial information bits between adjacent code blocks. However, this copying and sharing mechanism relies on complex puncturing operations and employs empirically fixed patterns (simply selecting the first N information bits) or preset positions based on the system code structure, lacking a quantitative filtering mechanism based on performance gain evaluation. This "structure-oriented" rather than "channel-oriented" design logic often ignores the reliability differences of different bit positions in actual noise environments, leading to the potential waste of coupling resources on inherently reliable strong bits, while truly weak links are not effectively protected, thus limiting the system's resource utilization efficiency.

[0006] In addition, some schemes attempt to fuse the information blocks of adjacent code blocks using modulo-2 addition. The problem is that full coupling introduces too much correlation. If the previous code block is decoded incorrectly, it can easily affect the current code block through the strong coupling link, causing serious error propagation. At the same time, this coupling strategy lacks specificity and cannot guarantee that the coupling gain will be applied precisely to the bit positions that need the most protection, resulting in low resource utilization efficiency.

[0007] Other schemes, while also utilizing algorithms such as Gaussian approximation to calculate reliability and couple data, merely use GA as a simple sorting tool. Essentially, they still divide information bits into fixed sets (such as Q2, Q3, etc., medium reliability sets) based on experience and employ a crude "copy-paste" strategy to directly map the high-reliability set from the previous frame to the current frame. This strategy lacks an effective mechanism for precisely identifying "high-yield" bottleneck bits through quantitative indicators, resulting in coupling gain failing to accurately target the truly weak links. Furthermore, its core function is only to provide a path for "inter-frame bit hard replacement" after decoding failure. This "post-hoc remedy" strategy relies on hard decision feedback, introducing additional retransmission delays and failing to utilize the temporal correlation of soft information during iteration to fundamentally suppress oscillations. Although existing technologies have publicly used Gaussian approximation algorithms to calculate bit reliability in AWGN channels and apply it to the selection of punched patterns or frozen bits, these existing schemes only use the GA calculation results for static construction within a single frame, failing to utilize this reliability distribution information to guide the optimization of spatial coupling structures across multiple frames. In particular, there is a lack of mechanisms to utilize GA results to construct coupled scoring indicators, thereby achieving direct coupling between information bits.

[0008] In decoding, SC-PC typically employs sliding window decoding or iterative decoding based on soft cancellation (SCAN) or belief propagation (BP). Existing iterative decoding optimization schemes passively monitor error rate fluctuations and preserve the optimal state during iteration. While this avoids performance degradation, it fails to fundamentally utilize historical iteration information to suppress oscillations themselves. This approach ignores the temporal correlation of soft information between different iteration stages and lacks effective information aggregation and smoothing mechanisms, resulting in unnecessary fluctuations in the decoding process and a tendency to get trapped in local optima.

[0009] To address the problems existing in the current technology, there is an urgent need for an SC-PC construction and decoding method optimized for AWGN channels. This method aims to overcome the limitations of traditional coupling schemes, such as the single object and lack of adaptability in construction, and explore a mechanism to directly establish information-to-information (Info-to-Info) coupling relationships between low-reliability information bits. By flexibly adjusting the coupling strategy according to channel characteristics, precise allocation and efficient utilization of coupling resources can be achieved, thereby improving error correction performance while avoiding the error propagation risk caused by full coupling. Furthermore, in the decoding stage, there is a need to introduce a mechanism that can effectively utilize historical iteration information to fully explore the temporal correlation between different iteration stages, overcome the oscillation problem caused by neglecting process information in traditional iterative decoding, and further improve decoding convergence performance. Summary of the Invention

[0010] The purpose of this invention is to provide a method for constructing and decoding spatially coupled polar codes in AWGN channels, which helps to improve the error correction performance of polar codes in complex channels.

[0011] To achieve the above objectives, the present invention adopts the following technical solution:

[0012] A method for constructing and decoding spatially coupled polar codes under an AWGN channel includes the following steps performed sequentially:

[0013] S1: The reliability distribution sequence adapted to the AWGN channel is obtained by using the Gaussian approximation algorithm. Here, AWGN represents additive white Gaussian noise.

[0014] S2: Using the bit error rate and frame error rate of the first iteration as a baseline, calculate the performance improvement score under different coupling ratios, and select the ratio with the highest performance improvement score as the optimal coupling ratio;

[0015] S3: Based on optimal coupling ratio Based on the reliability distribution sequence, calculate the number of information bits to be coupled in each frame. , Given the information bit length of each frame, select the code block with the lowest reliability. Each information bit is added modulo-2 with the information bit at the corresponding position of the previous code block to generate a spatially coupled structure.

[0016] S4: Based on the time-aggregated weights of normalized kurtosis scores, the system operating parameters are selected. The specific steps are as follows:

[0017] S4-1: Define the quality metrics used to quantify the reliability of the current decoding state. :

[0018] ;

[0019] in, The frame error rate is the corresponding weight. This represents the time-aggregated scoring parameters;

[0020] S4-2: Setting a baseline based on the performance limitations at both ends of the system: when At that time, the system continued to oscillate due to a lack of smoothness. At that time, the system stopped converging due to a lack of information updates;

[0021] Use the average performance of these two endpoints as a baseline :

[0022] ;

[0023] Define normalized kurtosis score The deviation of the current quality from the baseline:

[0024] ;

[0025] S4-3: Determining the optimal parameters: Traversal Interval, select one that makes Reaching the maximum value As a time decay weight;

[0026] S5: Perform enhanced iterative decoding and output the decoding result. The specific steps are as follows:

[0027] S5-1: Initial SCAN and Baseline Acquisition: The receiver performs a standard SCAN decoding round to acquire the raw soft information of the current iteration. ;

[0028] Using this raw soft information Perform hard decision and calculate the current bit error rate and frame error rate, which will be used as the baseline performance record for this iteration;

[0029] S5-2: Time-domain weighted aggregation: A time-domain weighted aggregation mechanism is introduced during the iterative decoding process to transform the original soft information... The soft information that iterated over the same bit in the previous round Perform weighted fusion to obtain the weighted fused soft information. ;

[0030] The following formula is used for weighted fusion. :

[0031] ;

[0032] in, The time decay factor is taken; the optimal score value established in S4-3 is used, i.e. , Indicates the iteration time index;

[0033] S5-3: Spatial Domain Interaction and Closed-Loop Iteration: Integrating Weighted Soft Information As prior knowledge, combined with the spatial coupling structure constructed in step 3, forward-backward soft information is transmitted between adjacent code blocks, and it is iterated repeatedly until the preset maximum number of iterations 6 is reached.

[0034] S5-4: Final Decision: After completing the iteration, perform a hard decision on the final soft information and output the decoding result.

[0035] Preferably, the specific steps of step S2 are as follows:

[0036] The improvement score in bit error rate (BER) relative to the baseline is calculated using the following formula:

[0037] ;

[0038] in, This represents the final bit error rate after completing the iteration at the target coupling ratio. This represents the bit error rate with the first iteration as the baseline.

[0039] The improvement score in frame error rate (FER) relative to the baseline is calculated using the following formula:

[0040] ;

[0041] in, This represents the final frame error rate after iteration at the target coupling ratio. This represents the frame error rate with the first iteration as the baseline.

[0042] The performance improvement score is calculated using the following formula. : .

[0043] Preferably, the specific steps of step S3 are as follows:

[0044] S3-1: Generate bit selection mask: based on optimal coupling ratio Calculate the number of information bits to be coupled in each frame. Generate a length of binary mask vector ;

[0045] For each information bit index If this location has the lowest reliability... One bit, set This position is identified as participating in spatial coupling operations, i.e., the bit to be coupled; otherwise, it is set. This position is designated to be transmitted independently, i.e., an uncoupled bit; this process is represented by the following formula:

[0046] ;

[0047] in, The one with the lowest reliability A set of information bits;

[0048] S3-2: Chained Information Injection: For the first The data block and the first Conditional coupling of data blocks;

[0049] The number of couplings is the number of information bits to be coupled. ;

[0050] Filtering low-reliability bits: Based on this reliability distribution sequence, in the current bit... From the set of information bits in the frame, select the one with the lowest reliability. One location;

[0051] Perform modulo-2 addition coupling: This The information bits at each low-reliability location are added modulo-2 with the corresponding information bits from the previous frame. This process is represented by the following formula:

[0052] ;

[0053] in, This represents modulo 2 addition. This represents the logical AND operation;

[0054] S3-3: Encoded Transmission: Definition of the... The input information vector of each polar code encoder is ,in, The length of the information bits. , The length of the coupling chain is the input information vector. Mapping to the unfrozen bits of the polar code, with the frozen bits fixed at 0, and performing standard polar coding operations yields the codeword. And it is transmitted via the AWGN channel.

[0055] By adopting the aforementioned design scheme, the beneficial effects of the present invention are as follows: This application breaks through the limitation of the traditional method of using only frozen bits or check bits for coupling, and directly establishes coupling association between low reliability information bits. This coupling can directly introduce external prior information to reduce entropy value. Moreover, unlike the crude set copying mode in the prior art, this application adopts precise bit-level screening based on reliability distribution, thereby making full use of the channel polarization characteristics of polar codes and significantly improving error correction performance.

[0056] To address the lack of theoretical support for existing technologies that employ empirical fixed ratios or full coupling strategies, this application calculates performance improvement scores under different coupling ratios and selects the ratio with the highest performance improvement score as the optimal coupling ratio. This pre-screens the optimal coupling ratio for the system under the AWGN channel. This "offline optimization, online configuration" strategy ensures that the system is at its optimal operating point when it is put into use, avoiding the complexity of real-time calculations and being more targeted and efficient than blindly setting fixed ratios.

[0057] To address the resource waste and error propagation problems caused by the lack of specificity in the full coupling strategy, this application selects the ratio with the highest performance improvement score as the optimal coupling ratio and only couples the selected low reliability bits. This mechanism ensures that the coupling gain is applied precisely to the bit position that needs the most protection, thereby significantly improving resource utilization efficiency while effectively cutting off the error propagation link caused by full coupling.

[0058] Unlike existing technologies that employ remedial mechanisms such as "post-event hard decision replacement" or "passive state selection," this application proactively introduces a time aggregation weighting mechanism during the iteration process. This mechanism utilizes historical iteration soft information to smooth current fluctuations. By introducing quality indicators to optimize weight coefficients, this strategy fully explores the temporal correlation between different iteration stages, fundamentally suppressing decoding oscillations and bringing additional performance improvements. Attached Figure Description

[0059] Figure 1 This is a schematic diagram of the selective spatial coupling structure based on bit reliability of the present invention;

[0060] Figure 2 This is a comparison of the BER performance of the novel spatially coupled polar code of this invention and the traditional spatially coupled polar code at a 30% coupling rate.

[0061] Figure 3 This is a comparison chart of the FER performance of the novel spatially coupled polar code of the present invention and the traditional spatially coupled polar code at a coupling rate of 30%.

[0062] Figure 4 This is a comparison chart of the BER performance of the traditional spatially coupled polar code using conventional decoding algorithms and the new spatially coupled polar code using a time aggregation weighting mechanism.

[0063] Figure 5 This is a comparison chart of the FER performance of the traditional spatially coupled polar code using conventional decoding algorithms and the new spatially coupled polar code using a time aggregation weighting mechanism. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0065] The terms "first," "second," "third," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0066] A method for constructing and decoding spatially coupled polar codes under an AWGN channel includes the following steps performed sequentially:

[0067] S1: Use the Gaussian approximation algorithm to obtain the reliability distribution sequence adapted to the AWGN channel;

[0068] In this embodiment, for the AWGN channel, the traditional BEC channel Bartholomew parameter calculation method is abandoned, and the Gaussian approximation (GA) algorithm, which is commonly used in this field, is selected as the reliability metric tool to construct an accurate reliability distribution sequence.

[0069] Specifically, based on a given channel noise variance The Gaussian approximation algorithm is called to obtain the reliability value of each bit position, thereby generating the physical layer input data required for subsequent steps.

[0070] It should be noted that the core of this step lies in obtaining the accurate reliability distribution adapted to the AWGN channel, rather than the Gaussian approximation algorithm itself. Therefore, any evaluation algorithm that can accurately reflect the channel quality of polar codesons under the AWGN channel (such as the density evolution DE algorithm) can be used instead without affecting the implementation of this invention. This provides the necessary quantitative basis for subsequent precise coupling.

[0071] S2: Using the bit error rate and frame error rate of the first iteration as a baseline, calculate the performance improvement score under different coupling ratios, and select the ratio with the highest performance improvement score as the optimal coupling ratio;

[0072] The specific steps of step S2 are as follows:

[0073] The improvement score in bit error rate (BER) relative to the baseline is calculated using the following formula:

[0074] ;

[0075] in, This represents the final bit error rate after completing the iteration at the target coupling ratio. This represents the bit error rate with the first iteration as the baseline. Used to characterize the improvement in bit-level error correction capability;

[0076] The improvement score in frame error rate (FER) relative to the baseline is calculated using the following formula:

[0077] ;

[0078] in, This represents the final frame error rate after iteration at the target coupling ratio. This represents the frame error rate with the first iteration as the baseline. It represents an improvement in code block-level integrity;

[0079] The performance improvement score is calculated using the following formula. : .

[0080] This metric comprehensively considers both bit-level local error correction capability and block-level global integrity, achieving dual optimization. The total score is selected. The highest ratio is the optimal coupling ratio. Through traversal calculations and verification, the optimal value in this embodiment is 0.3, thus theoretically guaranteeing the robustness of the coupled structure.

[0081] Using the bit error rate (BER) and frame error rate (FER) of the first iteration as baselines, performance improvement scores are calculated for different coupling ratios, and the total score is selected. The highest ratio is the optimal coupling ratio. This avoids the limitations of optimizing a single indicator.

[0082] To overcome the inaccuracy of single metrics, such as bit error rate alone, in short code length assessments, this application constructs a multidimensional evaluation model based on a "coupling score index" S. This model is not a simple parameter scan, but rather introduces a mechanism to balance the inherent trade-off between bit-level error correction capability and code block-level integrity.

[0083] Using the decoding performance of the "first iteration" as a baseline reference, the coupling ratio for each candidate... (e.g., 0.1, 0.2, ..., 1):

[0084] S3: As Figure 1As shown, SC-PC construction based on selective coupling is performed. This is based on the determined optimal coupling ratio. Calculate the number of bits to be coupled , Given the information bit length of each frame, select the code block with the lowest reliability. Each information bit is combined with the corresponding information bit in the previous code block using a modulo-2 addition (XOR) operation to achieve direct Info-to-Info coupling;

[0085] Let the number to be transmitted be... The original information vector of each data block is ,in The length of the information bits. , This represents the length of the coupling chain.

[0086] Definition of the first The input information vector of each polar code encoder is .

[0087] Based on the determined optimal coupling ratio The coupled structure is generated using the reliability distribution sequence obtained in step 1. The specific steps are as follows:

[0088] S3-1: Generate bit selection mask: based on optimal coupling ratio Calculate the number of information bits that need to be coupled in each frame. Generate a length of binary mask vector .

[0089] For each information bit index If this location has the lowest reliability... One bit, set This position is identified as participating in spatial coupling operations, i.e., the bit to be coupled; otherwise, it is set. This position is designated to be transmitted independently, i.e., an uncoupled bit; this process is represented by the following formula:

[0090] ;

[0091] in, The one with the lowest reliability A set of information bits.

[0092] S3-2: Chained Information Injection: For the first The data block and the first Conditional coupling of data blocks.

[0093] The specific implementation of conditional coupling is as follows:

[0094] Determine the number of couplings: based on the optimal coupling ratio. Calculate the number of information bits that need to be coupled in each frame. ;in In this embodiment, the total number of bits of information per frame is [number]. .

[0095] Filtering low-reliability bits: Based on the reliability distribution sequence obtained in step 1, in the current bit... From the set of information bits in the frame, select the one with the lowest reliability. One position.

[0096] Perform modulo-2 addition coupling: This The information bits at the low-reliability location are the same as those in the previous frame, i.e., the first... The information bits at the corresponding positions in the frame are subjected to modulo-2 addition (XOR) operation.

[0097] This process is represented by the following formula:

[0098] ;

[0099] in, This represents modulo 2 addition. Represents logical AND operations.

[0100] In this embodiment, when At that time, the information bits at the corresponding positions of the previous time step will be... By introducing the current code block and establishing a deterministic dependency between specific low-reliability locations of adjacent code blocks through an XOR operation, information can be transferred across blocks.

[0101] This direct Info-to-Info coupling mode directly enhances the error correction capability of weak information bits, unlike the traditional indirect Info-to-Frozen coupling. Simultaneously, this selection mechanism is based on channel statistical characteristics, calculated using the GA algorithm, enabling bit reliability tracking and effectively avoiding the inefficient and resource-wasting methods of traditional schemes that rely on empirical set partitioning, such as mapping the entire high-reliability area to the low-reliability area.

[0102] S3-3: Encoded Transmission: Transmitting the constructed input vector Mapping to the unfrozen bits of the polar code, with the frozen bits fixed at 0, and performing standard polar coding operations yields the codeword. And it is transmitted via the AWGN channel.

[0103] S4: Time-aggregated weight optimization based on normalized kurtosis score; specific steps are as follows:

[0104] To address the oscillation problem caused by ignoring historical states in traditional iterative decoding, this application does not employ a general parameter traversal but instead establishes an evaluation system based on normalized kurtosis scores. This system utilizes the unimodal physical characteristic of system performance varying with the degree of aggregation over time.

[0105] S4-1: Define the quality metrics used to quantify the reliability of the current decoding state. :

[0106] ;

[0107] in, The frame error rate is the corresponding weight. This represents the time-aggregated score parameter, used to normalize the kurtosis score. Offline optimization, online runtime command As a weighting coefficient.

[0108] S4-2: Setting a baseline based on the performance limitations at both ends of the system: when When there is no aggregation, the system continues to oscillate due to a lack of smoothing; when In the "full hold" scenario, the system stops converging due to a lack of information updates. Neither approach achieves optimal performance.

[0109] Therefore, the average performance of these two endpoints is used as a baseline. :

[0110] ;

[0111] Define normalized kurtosis score The deviation of the current quality from the baseline:

[0112] ;

[0113] The physical meaning of this indicator lies in: utilizing the performance curve in By leveraging the unimodal characteristics within the interval, we can find the optimal balance point that can both suppress oscillations and ensure effective information updates, thereby making the system performance significantly better than the benchmarks at both ends.

[0114] S4-3: Determining the optimal parameters: Traversal Interval, select one that makes Reaching the maximum value As a time decay weight.

[0115] In this embodiment, the calculated The maximum value is determined, therefore the optimal weight is determined. During the online operation phase, the time aggregation weight is taken as follows: This process ensures that parameter selection is based on clear performance gain criteria.

[0116] S5: Enhanced iterative decoding process, the specific steps are as follows:

[0117] S5-1: Initial SCAN and Baseline Acquisition: The receiver first performs a standard SCAN decoding to acquire the raw soft information of the current iteration. .

[0118] Using this A hard decision is made, and the current bit error rate and frame error rate are recorded as the baseline performance for this iteration. This step provides a reference benchmark for subsequent quality assessments.

[0119] S5-2: Temporal Aggregation: Introducing a temporal aggregation weighting mechanism during iterative decoding. This involves calculating the soft information... The soft information that iterated over the same bit in the previous round Perform weighted fusion to obtain the weighted fused soft information. The previous round here refers to the previous moment in the iteration time of the same bit segment, that is, the aggregated soft information of the previous iteration.

[0120] The following formula is used for weighted fusion. :

[0121] ;

[0122] in, The time decay factor (Time Decay Tau) is the time decay weight calculated in S4-3, i.e. , Represents the iteration time index, indicating the time-domain weighted aggregation. Round-by-round iteration. This step utilizes the "inertia" of historical information to smooth out fluctuations in the current iteration, effectively preventing oscillations during the decoding process.

[0123] S5-3: Spatial Domain Interaction and Closed-Loop Iteration: Integrating Weighted Soft Information As prior knowledge, combined with the spatial coupling structure constructed in step 3, forward-backward soft information transmission is performed between adjacent code blocks. This process is repeated iteratively until the preset maximum number of iterations is reached; in this embodiment, the maximum number of iterations is 6.

[0124] S5-4: Final Decision: After all iterations are completed, a hard decision is made on the final soft information, and the decoding result is output.

[0125] The effects of this invention can be further illustrated by the following simulations:

[0126] The SC-PC construction method based on selective coupling proposed in this invention was compared with the traditional fully coupled method under the same conventional decoding algorithm. Figure 2 , Figure 3 As shown. Experimental results show that, under the preferred coupling ratio (30%), relying solely on the improvement of the construction method, the method of the present invention achieves a bit error rate gain of approximately 0.35 dB and a frame error rate gain of approximately 0.45 dB compared to the traditional method in the high signal-to-noise ratio region.

[0127] Furthermore, to verify the superiority of the overall scheme, the performance of traditional spatially coupled polar codes using conventional decoding algorithms was compared with that of new spatially coupled polar codes using a time-aggregation weighting mechanism, such as... Figure 4 , Figure 5 As shown. The results show that at the optimal weighting coefficient ( At 0.6, the new construction method, combined with the time aggregation mechanism, can effectively reduce error leveling, resulting in an additional performance improvement of about 0.4dB.

[0128] In summary, the simulation results fully demonstrate the effectiveness of the proposed method for constructing spatially coupled polar codes and optimizing iterative decoding in AWGN channels.

[0129] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for constructing and decoding spatially coupled polar codes in an AWGN channel, characterized in that: The steps are as follows, performed sequentially: S1: Use the Gaussian approximation algorithm to obtain the reliability distribution sequence adapted to the AWGN channel; S2: Using the bit error rate and frame error rate of the first iteration as a baseline, calculate the performance improvement score under different coupling ratios, and select the ratio with the highest performance improvement score as the optimal coupling ratio; S3: Based on optimal coupling ratio Based on the reliability distribution sequence, calculate the number of information bits to be coupled in each frame. , Given the information bit length of each frame, select the code block with the lowest reliability. Each information bit is added modulo-2 with the information bit at the corresponding position of the previous code block to generate a spatially coupled structure. S4: Based on the time-aggregated weights of normalized kurtosis scores, the system operating parameters are selected. The specific steps are as follows: S4-1: Define the quality metrics used to quantify the reliability of the current decoding state. : ； in, The frame error rate is the corresponding weight. This represents the time-aggregated scoring parameters; S4-2: Setting a baseline based on the performance limitations at both ends of the system: when At that time, the system continued to oscillate due to a lack of smoothness. At that time, the system stopped converging due to a lack of information updates; Use the average performance of these two endpoints as a baseline : ； Define normalized kurtosis score The deviation of the current quality from the baseline: ； S4-3: Determining the optimal parameters: Traversal Interval, select one that makes Reaching the maximum value As a time decay weight; S5: Perform enhanced iterative decoding and output the decoding result. The specific steps are as follows: S5-1: Initial SCAN and Baseline Acquisition: The receiver performs a standard SCAN decoding round to acquire the raw soft information of the current iteration. ; Using this raw soft information Perform hard decision and calculate the current bit error rate and frame error rate, which will be used as the baseline performance record for this iteration; S5-2: Time-domain weighted aggregation: A time-domain weighted aggregation mechanism is introduced during the iterative decoding process to transform the original soft information... The soft information that iterated over the same bit in the previous round Perform weighted fusion to obtain the weighted fused soft information. ; The following formula is used for weighted fusion. : ； in, As the time decay factor, the optimal score value established in S4-3 is taken, i.e. , Indicates the iteration time index; S5-3: Spatial Domain Interaction and Closed-Loop Iteration: Integrating Weighted Soft Information As prior knowledge, combined with the spatial coupling structure constructed in step 3, forward-backward soft information is transmitted between adjacent code blocks, and it is iterated repeatedly until the preset maximum number of iterations 6 is reached. S5-4: Final Decision: After completing the iteration, perform a hard decision on the final soft information and output the decoding result.

2. The method for constructing and decoding spatially coupled polar codes in an AWGN channel as described in claim 1, characterized in that: The specific steps of step S2 are as follows: The improvement score in bit error rate (BER) relative to the baseline is calculated using the following formula: ； in, This represents the final bit error rate after completing the iteration at the target coupling ratio. This represents the bit error rate with the first iteration as the baseline. The improvement score in frame error rate (FER) relative to the baseline is calculated using the following formula: ； in, This represents the final frame error rate after iteration at the target coupling ratio. This represents the frame error rate with the first iteration as the baseline. The performance improvement score is calculated using the following formula. : .

3. The method for constructing and decoding spatially coupled polar codes in an AWGN channel as described in claim 2, characterized in that: The specific steps of step S3 are as follows: S3-1: Generate bit selection mask: based on optimal coupling ratio Calculate the number of information bits to be coupled in each frame. Generate a length of binary mask vector ; For each information bit index If this location has the lowest reliability... One bit, set This position is identified as participating in spatial coupling operations, i.e., the bit to be coupled; otherwise, it is set. This position is designated to be transmitted independently, i.e., an uncoupled bit; this process is represented by the following formula: ； in, The one with the lowest reliability A set of information bits; S3-2: Chained Information Injection: For the first The data block and the first Conditional coupling of data blocks; The number of couplings is the number of information bits to be coupled. ; Filtering low-reliability bits: Based on this reliability distribution sequence, in the current bit... From the set of information bits in the frame, select the one with the lowest reliability. One location; Perform modulo-2 addition coupling: This The information bits at each low-reliability location are added modulo-2 with the corresponding information bits from the previous frame. This process is represented by the following formula: ； in, This represents modulo 2 addition. This represents the logical AND operation; S3-3: Encoded Transmission: Definition of the... The input information vector of each polar code encoder is ,in, The length of the information bits. , The length of the coupling chain is the input information vector. Mapping to the unfrozen bits of the polar code, with the frozen bits fixed at 0, and performing standard polar coding operations yields the codeword. And it is transmitted via the AWGN channel.

Images