Polarization code construction method and system for underground non-stationary channel
By employing the Monte Carlo algorithm and bidirectional mapping relationship to optimize the construction of polar codes in an underground magnetic induction communication system, the problems of channel non-stationarity and high computational complexity were solved, achieving efficient utilization of coding resources and improved bit error rate performance.
Patent Information
- Application Number
- CN202511456887.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-13
- Publication Date
- 2026-01-09
AI Technical Summary
Underground magnetic induction communication systems face problems such as channel non-stationarity, insufficient coding adaptability, and high computational complexity. Existing polar code construction methods are difficult to adapt to the asymmetric channel structure of underground wireless communication scenarios, resulting in inaccurate bit allocation, degraded bit error rate performance, and low efficiency in utilizing computational resources.
The Monte Carlo algorithm is used to set a preset number of iterations. A bidirectional mapping relationship is established through degenerate mapping and evolution mapping. The upper and lower bounds of the Bach parameters of the degenerate channel and the evolution channel are calculated. By combining weight optimization and minimizing difference loss, the polar code encoding process is optimized, reducing computational complexity and improving encoding reliability.
While ensuring communication reliability, the computational complexity of constructing channel polar codes is reduced, the efficiency of coding resource utilization and the robustness and adaptability of the system are improved, and the bit error rate performance of the underground magnetic induction communication system is enhanced.
Smart Images

Figure CN121308915A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of wireless magnetic induction communication technology, and more specifically, relates to a method and system for constructing polar codes for underground non-stationary channels. Background Technology
[0002] With the rapid development of applications such as underground resource exploration, disaster early warning, and smart agriculture, the demand for underground wireless communication is becoming increasingly urgent. Traditional electromagnetic wave communication technologies face problems such as high path loss, poor penetration, and significant multipath effects in underground environments. Magnetic induction communication technology, due to its advantages such as low path loss, high medium penetration, and channel stability, has become an important solution for underground communication.
[0003] However, underground magnetic induction communication systems face the following technical challenges in practical applications: 1) Channel nonstationarity: Factors such as the conductivity distribution of the geological medium, conductor disturbance, and humidity changes cause the magnetic induction channel to exhibit significant spatial nonstationarity and statistical inequality, with significant differences in reliability between sub-channels.
[0004] 2) Insufficient coding adaptability: Existing coding schemes are based on the assumption of independent and identical distribution, which makes it difficult to adapt to the asymmetric channel structure of underground wireless communication scenarios, resulting in inaccurate bit allocation and decreased bit error rate performance.
[0005] 3) High computational complexity: Traditional polar code construction relies on Monte Carlo simulation, which requires a large number of iterative calculations and cannot meet the real-time requirements of resource-constrained scenarios.
[0006] In existing technologies, although polar codes have been introduced into underground communication due to their symmetric capacity reachability and channel dependence characteristics, traditional construction methods such as Monte Carlo construction, density evolution construction, and Gaussian approximation construction are mainly for channels with a single statistical characteristic distribution and do not fully consider the characteristics of underground magnetic induction channels. For example, the Monte Carlo construction method obtains a large number of sub-channel reliability samples through numerous iterations to provide a sub-channel reliability ranking, but it requires repeated selection of coded codewords to dynamically adapt to the dynamic characteristics of non-stationary channels, resulting in high computational overhead and low efficiency in the utilization of coding resources, thus limiting the actual deployment efficiency.
[0007] Therefore, there is an urgent need for a polar code construction scheme that can integrate channel statistical characteristics, flexibly adjust coding parameters, and achieve a flexible trade-off between computational efficiency and communication performance, so as to improve the robustness and adaptability of underground magnetic induction communication systems. Summary of the Invention
[0008] In view of the above-mentioned defects or improvement needs of the existing technology, the present invention provides a method and system for constructing polar codes for underground non-stationary channels. Its purpose is to reduce the computational complexity of constructing polar codes for the channel while ensuring the reliability of magnetic induction communication.
[0009] To achieve the above objectives, in a first aspect, the present invention provides a polar code construction method for underground non-stationary channels, for use in an underground magnetic induction communication system; the underground magnetic induction communication system includes: a transmitting coil and a receiving coil with parallel cross sections and placed coaxially; The above-mentioned polar code construction methods include: The number of iterations in the Monte Carlo algorithm is set to a preset number. The Monte Carlo algorithm is then used to calculate the Barcol parameters of each sub-channel in the original channel, which serve as the reference values for the Barcol parameters of each sub-channel. The original channel of the underground magnetic induction communication system is degraded and mapped to obtain a degraded channel, thereby mapping the continuous channel output of the original channel to a finite symbol set; the degraded channel is then transformed into an evolved channel through an evolutionary mapping, thereby establishing an inverse mapping relationship from the output of the evolved channel back to the output of the original channel. Calculate the Barton parameters of each subchannel in the degraded channel, and use them as the upper bound of the Barton parameters of each subchannel; Calculate the Barton parameters of each subchannel in the evolutionary channel separately, and use them as the lower bound of the Barton parameters of each subchannel; The upper and lower bounds of the Barton parameters for each sub-channel are weighted and summed to obtain the calibration value of the Barton parameters for that sub-channel. The weights are optimized by minimizing the difference loss between the calibrated value of the Barthelium parameter of each sub-channel and the corresponding reference value, and the calibrated value of the Barthelium parameter of each sub-channel obtained based on the optimal weight is taken as its final value. The final values of the Parvian parameters for each subchannel are sorted, and the subchannel indices corresponding to the top RN Parvian parameter final values with the largest values are used as frozen bit indices, while the subchannel indices corresponding to the remaining Parvian parameter final values are used as information bit indices; R is the code rate of the polar code; N is the length of the original channel; Will transmit signal X The transmit bit under the middle freeze bit index is set to 0, and the transmit bit under the information bit index is set to 1, in order to... X Encode the vector to obtain the encoded vector, and then combine the encoded vector with the generator matrix G of the polar code of length N. N Multiplying these together yields the polar code encoding of the underground magnetic induction communication system channel.
[0010] More preferably, the above-mentioned degradation mapping of the original channel of the underground magnetic induction communication system includes: A1. Calculation Each received bit in The log-likelihood ratio of each transmitted bit is calculated, and the resulting log-likelihood ratio is divided into L equal-width intervals. These L equal-width intervals are then compared with L distinct primary symbols. One-to-one correspondence; will The corresponding partition is divided into L subsets of transmitted bits, each corresponding to one of L equal-width intervals; The corresponding data is divided into L subsets of received bits, each corresponding to one of L equally wide intervals; L ≥ 2. ; For a pre-defined set of discrete numerical values; initialize the set. Initialize the degraded channel Q; A2. Determine if the iteration deadline condition is met. If yes, proceed to A4; otherwise, calculate the merging cost of every two adjacent primary symbols in the current set S; where, and Merger costs ; Indicates a merge operation; ; The number of primary symbols in the current set S; Primary symbols The specific parameters of the Bartholomew's Law are as follows: ; and They are respectively The number of transmitted bits with a value of 0 and the number of transmitted bits with a value of 1 in the corresponding subset of transmitted bits; The new symbol after merging, its Bartholin's parameter ; and They are respectively The corresponding subset of transmitted bits and The number of transmitted bits with values of 0 and 1 in the merged set of the corresponding subset of transmitted bits; and The output bits are respectively the output bits when the input bit is 0 and the output bit is 1 under the current degraded channel. The transition probability; ; The deadline conditions include: The number of polar code information bits is less than or equal to the number of bits, and the capacity loss between the current degraded channel and the original channel is greater than or equal to the corresponding upper bound; The upper bound of the capacity loss between the current degraded channel and the original channel is: Where K is the number of the currently received subset of bits; This represents the current k-th received bit subset; for Received bits in The marginal density function; Let KL divergence be a metric. For the original channel, the input is The output for each transmitted bit is the received bit. The vector formed by the transition probabilities; For the current degraded channel, the input is The output is when each bit is transmitted. The vector formed by the transition probabilities; for The corresponding primary symbols; A3. Merge the two adjacent primary symbols with the minimum merging cost in set S, merge the intervals corresponding to these two adjacent primary symbols, merge the transmit bit subsets corresponding to these two adjacent primary symbols, and merge the receive bit subsets corresponding to these two adjacent primary symbols; update the degraded channel. Proceed to A2; To represent the i-th subchannel, this includes the probability of the transmitted bit in the i-th time slot mapping to the primary symbol; specifically, the transmitted bit in the i-th time slot... Mapped to the corresponding primary symbols The probability of: ; The input under the original channel is The output is The probability value at that time; To receive bits as preset Mapped to primary symbols The probability of; A4. Treat the current degraded channel as the degraded channel after degrading the original channel, and use the current set S as the primary symbol output set.
[0011] More preferably, the above-mentioned evolutionary mapping of the degraded channel includes: performing the following on each sub-channel C in the degraded channel: Select M primary symbols from the primary symbol output set to form the output primary symbol subset of subchannel C. M≥1; Perform an evolutionary mapping on sub-channel C such that the evolved sub-channel C satisfies: When the input symbol of subchannel C is 0, the preset probability is used respectively. The first M-1 symbols should be output. With probability Output the Mth symbol ; When the input symbol of subchannel C is 1, the preset probability is used respectively. The first M-1 symbols should be output. With probability Output the Mth symbol ; in, ; ; ; ; satisfy: ; This is the reliability metric for the m-th symbol.
[0012] More preferably, M is 2 or 3.
[0013] More preferably, the reliability metric of the m-th symbol is represented by its likelihood ratio.
[0014] In a second aspect, the present invention provides a polar code construction system for underground non-stationary channels, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor executes the polar code construction method provided in the first aspect of the present invention when executing the computer program.
[0015] Thirdly, the present invention also provides a computer-readable storage medium comprising a stored computer program, wherein the computer program, when executed by a processor, controls the device where the storage medium is located to execute the polar code construction method provided in the first aspect of the present invention.
[0016] Fourthly, the invention also provides a computer program product, including a computer program / instruction that, when executed by a processor, implements the polar code construction method provided in the first aspect of the invention.
[0017] Fifthly, the invention also provides an information processing terminal equipped with the polar code construction system provided in the second aspect of the invention.
[0018] Sixthly, the invention also provides a network information processing platform equipped with the polar code construction system provided in the second aspect of the invention.
[0019] In summary, the above-described technical solutions conceived in this invention can achieve the following beneficial effects: This invention provides a method for constructing polar codes for underground non-stationary channels. It establishes a bidirectional mapping relationship to perform a degenerate mapping on the original channel of an underground magnetic induction communication system, mapping the continuous channel output of the original channel to a finite symbol set, thus obtaining a degenerate channel and reducing the complexity of channel modeling. An evolutionary mapping is then performed on the degenerate channel, establishing a reverse mapping relationship from the evolved output back to the original output, resulting in an evolved channel that accurately reflects the statistical characteristics of the original channel. The Bartholomew's parameters of each sub-channel in both the degenerate and evolved channels are calculated separately, serving as upper and lower bounds for the Bartholomew's parameters of each sub-channel. Using the Bartholomew's parameters of each sub-channel in the original channel calculated by a Monte Carlo algorithm based on a preset number of iterations as benchmark values, and through the fusion of the upper and lower bounds of the Bartholomew's parameters and flexible weight calibration, high-precision estimation of sub-channel reliability is achieved. This avoids the high computational overhead of traditional Monte Carlo methods and reduces the computational complexity of polar code construction while ensuring the reliability of magnetic induction communication. Attached Figure Description
[0020] Figure 1 A schematic diagram of an underground magnetic induction communication system provided in an embodiment of the present invention; Figure 2 The scatter plot of the upper bound of the Barcol parameters, the lower bound of the Barcol parameters, and the Barcol parameters obtained by the Monte Carlo algorithm is provided in the embodiments of the present invention. Figure 3 Scatter plots of Bartholomew's parameters obtained by the Monte Carlo algorithm and Bartholomew's parameters after MMSE correction, provided in embodiments of the present invention; Figure 4 This is a comparative scatter plot of the Parseltongue parameters after correction using MMSE at different code lengths, provided in an embodiment of the present invention. Figure 5 This is a schematic diagram illustrating the simulation comparison of communication performance at different code rates and code lengths provided in an embodiment of the present invention. Figure 6 This is a schematic diagram illustrating the simulation comparison of reliable communication distance and communication performance under different code rates and code lengths, provided in an embodiment of the present invention. Detailed Implementation
[0021] 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 and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.
[0022] To achieve the above objectives, in a first aspect, the present invention provides a polar code construction method for underground non-stationary channels, for use in an underground magnetic induction communication system; the underground magnetic induction communication system includes: a transmitting coil and a receiving coil with parallel cross sections and placed coaxially; The above-mentioned polar code construction methods include: The number of iterations in the Monte Carlo algorithm is set to a preset number. Using the Monte Carlo algorithm, the Barcol parameters of each sub-channel in the original channel are calculated separately, serving as the baseline values for the Barcol parameters of each sub-channel. In one optional implementation, the preset number is 500~10. 6 The optimal value is 500.
[0023] The original channel of the underground magnetic induction communication system is degraded and mapped to obtain a degraded channel, thereby mapping the continuous channel output of the original channel to a finite symbol set; the degraded channel is then transformed into an evolved channel through an evolutionary mapping, thereby establishing an inverse mapping relationship from the output of the evolved channel back to the output of the original channel. Calculate the Barton parameters of each subchannel in the degraded channel, and use them as the upper bound of the Barton parameters of each subchannel; Calculate the Barton parameters of each subchannel in the evolutionary channel separately, and use them as the lower bound of the Barton parameters of each subchannel; The upper and lower bounds of the Barton parameters for each sub-channel are weighted and summed to obtain the calibration value of the Barton parameters for that sub-channel. The weights are optimized by minimizing the difference loss between the calibrated value of the Barthelium parameter of each sub-channel and the corresponding reference value, and the calibrated value of the Barthelium parameter of each sub-channel obtained based on the optimal weight is taken as its final value. The final values of the Parvian parameters for each subchannel are sorted, and the subchannel indices corresponding to the top RN Parvian parameter final values with the largest values are used as frozen bit indices, while the subchannel indices corresponding to the remaining Parvian parameter final values are used as information bit indices; R is the code rate of the polar code; N is the length of the original channel; Will transmit signal X The transmit bit under the middle freeze bit index is set to 0, and the transmit bit under the information bit index is set to 1, in order to... X Encode the vector to obtain the encoded vector, and then combine the encoded vector with the generator matrix G of the polar code of length N. N Multiplying these together yields the polar code encoding of the underground magnetic induction communication system channel.
[0024] It should be noted that there are multiple methods for calculating the Bach parameters of each subchannel in the degraded and evolved channels, such as the direct calculation method (calculated according to the definition) and the Monte Carlo algorithm, etc., which are not limited here.
[0025] In one alternative implementation, the degradation mapping of the original channel of the aforementioned underground magnetic induction communication system includes: A1. Calculation Each received bit in The log-likelihood ratio of each transmitted bit is calculated, and the resulting log-likelihood ratio is divided into L equal-width intervals. These L equal-width intervals are then compared with L distinct primary symbols. One-to-one correspondence; will The corresponding partition is divided into L subsets of transmitted bits, each corresponding to one of L equal-width intervals; The corresponding data is divided into L subsets of received bits, each corresponding to one of L equally wide intervals; L ≥ 2. ; For a pre-defined set of discrete numerical values; initialize the set. Initialize the degraded channel Q; A2. Determine if the iteration deadline condition is met. If yes, proceed to A4; otherwise, calculate the merging cost of every two adjacent primary symbols in the current set S; where, and Merger costs ; Indicates a merge operation; ; The number of primary symbols in the current set S; Primary symbols The specific parameters of the Bartholomew's Law are as follows: ; and They are respectively The number of transmitted bits with a value of 0 and the number of transmitted bits with a value of 1 in the corresponding subset of transmitted bits; The new symbol after merging, its Bartholin's parameter ; and They are respectively The corresponding subset of transmitted bits and The number of transmitted bits with values of 0 and 1 in the merged set of the corresponding subset of transmitted bits; and The output bits are respectively the output bits when the input bit is 0 and the output bit is 1 under the current degraded channel. The transition probability; ; The deadline conditions include: The number of polar code information bits is less than or equal to the number of bits, and the capacity loss between the current degraded channel and the original channel is greater than or equal to the corresponding upper bound; The upper bound of the capacity loss between the current degraded channel and the original channel is: Where K is the number of the currently received subset of bits; This represents the current k-th received bit subset; for Received bits in The marginal density function; Let KL divergence be a metric. For the original channel, the input is The output for each transmitted bit is the received bit. The vector formed by the transition probabilities; For the current degraded channel, the input is The output is when each bit is transmitted. The vector formed by the transition probabilities; for The corresponding primary symbols; A3. Merge the two adjacent primary symbols with the minimum merging cost in set S, merge the intervals corresponding to these two adjacent primary symbols, merge the transmit bit subsets corresponding to these two adjacent primary symbols, and merge the receive bit subsets corresponding to these two adjacent primary symbols; update the degraded channel. Proceed to A2; To represent the i-th subchannel, this includes the probability of the transmitted bit in the i-th time slot mapping to the primary symbol; specifically, the transmitted bit in the i-th time slot... Mapped to the corresponding primary symbols The probability of: ; The input under the original channel is The output is The probability value at that time; To receive bits as preset Mapped to primary symbols The probability of; A4. Treat the current degraded channel as the degraded channel after degrading the original channel, and use the current set S as the primary symbol output set.
[0026] In one alternative implementation, the above-described evolutionary mapping of the degraded channel includes: performing the following on each sub-channel C in the degraded channel: Select M primary symbols from the primary symbol output set to form the output primary symbol subset of subchannel C. M≥1; Perform an evolutionary mapping on sub-channel C such that the evolved sub-channel C satisfies: When the input symbol of subchannel C is 0, the preset probability is used respectively. The first M-1 symbols should be output. With probability Output the Mth symbol ; When the input symbol of subchannel C is 1, the preset probability is used respectively. The first M-1 symbols should be output. With probability Output the Mth symbol ; in, ; ; ; ; satisfy: ; This is the reliability metric for the m-th symbol.
[0027] Preferably, in one alternative implementation, M is 2 or 3.
[0028] It should be noted that there are various methods for representing the reliability metric of a symbol, such as mutual information, likelihood ratio, expectation / quantile of the likelihood ratio, Barthel parameter, upper or lower bound of the error probability, etc., which are not limited here. Preferably, in an optional implementation, the reliability metric of the m-th symbol is represented by its likelihood ratio.
[0029] To further illustrate the polar code construction method for underground non-stationary channels provided by this invention, the following detailed description is provided in conjunction with specific embodiments: To achieve efficient and reliable transmission of underground magnetic induction wireless communication systems in complex geological environments, this embodiment proposes a non-stationary channel modeling and flexible coding construction framework based on polar code optimization. This method addresses the spatial non-stationarity, statistical inequality, and significant differences in sub-channel reliability existing in underground magnetic induction communication channels. Through refined channel modeling, degradation and evolution mapping mechanism design, and dynamic reliability fusion, it achieves efficient allocation of coding resources and a significant improvement in system performance.
[0030] Specifically, this embodiment first establishes a magnetic induction channel model that integrates path loss and non-stationary characteristics. By introducing a variational path loss function and a polarization coding framework, the dynamic relationship between channel gain and symbol signal-to-noise ratio is quantified. Based on this, a degradation modeling method for unequally distributed sub-channels is proposed. By utilizing intermediate degradation channels and an output alphabet discretization mechanism, continuous channel outputs are mapped to a finite set of symbols. Combined with a likelihood ratio-driven symbol merging strategy, the complexity of channel modeling is reduced. Furthermore, a channel evolution modeling and enhanced symbol mapping strategy is designed. By introducing reversible mapping intermediate channels, the resolution and capacity expression accuracy of sub-channels are improved. Finally, a dynamic reliability fusion method and sub-channel scheduling mechanism are constructed. By combining the upper and lower bounds of the Bach parameter calibration and code rate adjustment, the allocation of frozen information bits is optimized and the bit error rate is reduced.
[0031] Through the above technical solution, this embodiment effectively solves the problem of inaccurate reliability assessment of traditional polar codes in non-uniform channels. While ensuring controllable computational complexity, it significantly improves the robustness and communication distance of the system in highly dynamic magnetic induction environments. The specific implementation plan is as follows: 1) Establish an underground magnetic induction communication model: To adapt to the practical application scenarios of polar codes in magnetic induction underground communication environments, this embodiment first performs fine modeling of the magnetic induction channel, establishes the statistical structure of unequally distributed sub-channels, and introduces variational path loss expression under non-stationary conditions to support the subsequent coding structure and channel approximation mechanism.
[0032] like Figure 1 As shown, in an underground magnetic induction communication system, a transmitting coil generates a primary magnetic field through low-frequency alternating current. This magnetic field propagates through the geological medium and couples to the receiving coil, thus achieving an energy conversion process from electricity to magnetism and back to electricity. This transmission method differs significantly from traditional far-field electromagnetic waves, with its main influencing factors concentrated on coupling strength, coil orientation, and the spatial distribution of the intermediate conductive medium. In real-world environments, due to factors such as soil conductivity, doped metal objects, and humidity variations, the magnetic induction channel exhibits significant spatial non-stationarity and statistical inequality.
[0033] The spatial distribution of the conductor in the i-th time slot is represented as follows: The corresponding channel gain can be modeled as:
[0034] Combining the polar coding framework, the system transmitter will transmit the original bit sequence The codeword bits are generated after encoding by a polar encoder. The transmitted signal is obtained through modulation. Each transmit bit within it .
[0035] During channel propagation, the amplitude attenuation and phase rotation caused by the distribution of geological materials can be described by the following receiving model:
[0036] in, The received signal in the i-th time slot ; Let j be the channel gain in the i-th time slot; j is the imaginary sign. This refers to random phase shifts induced by the medium. The transmitted signal in the i-th time slot ; This refers to the nth transmitted bit in the i-th time slot; N is the polar code length; It is additive white Gaussian noise. ; This represents the one-sided power spectral density.
[0037] To further quantify the communication reliability of the system, bit signal-to-noise ratio (Bit SNR) is introduced. Based on the polar code coding structure, the code rate is... With modulation efficiency Under given conditions, the relationship between symbol signal-to-noise ratio and bit signal-to-noise ratio is as follows: .
[0038] Considering the path loss characteristics of magnetic propagation in magnetic induction communication systems, the above expression is further generalized to construct a bit signal-to-noise ratio (B-SNR) model based on the path loss model:
[0039] in, Let f be the symbolic energy, f be the excitation frequency of the magnetic field propagation, and d be the transmission distance. Let f be the path loss function determined by the frequency f, the transmission distance d, and the distribution of conductors in the environment D. It is worth noting that in the near-field region of magnetic induction communication, magnetic field propagation is mainly affected by the reflection / transmission effect of the conductor, and the dielectric constant has a relatively weak effect. Therefore, the traditional free-space loss expression is not applicable.
[0040] Expanding on this further, the bit signal-to-noise ratio model described above is as follows:
[0041] in, This refers to the transmission power. For the bandwidth of the underground magnetic induction communication system; The code rate of the polar code; The modulation order; This is the distance between the transmitting coil and the receiving coil; and This refers to the number of turns and radius of the transmitting coil; and The number of turns and radius of the receiving coil; Let be the fading coefficient in the i-th time slot.
[0042] In the construction of polar codes, the above modeling process not only effectively captures the non-stationarity of the underground magnetic induction channel and the asymmetry between sub-channels, but also provides theoretical support and modeling structure for the degradation mechanism, channel evolution strategy, and reliability ranking algorithm introduced in subsequent implementations. In particular, by introducing the channel gain in the i-th time slot... Spatial distribution of conductors in the i-th time slot By employing joint modeling techniques, polar code construction can adapt to the complex characteristics of underground magnetic induction communication at the underlying sub-channel level, thus possessing greater practical value and wider applicability.
[0043] The above process establishes a magnetic induction channel model that integrates path loss and non-stationary characteristics. By introducing a variational path loss function and a polarization coding framework, the dynamic relationship between channel gain and symbol signal-to-noise ratio is quantified.
[0044] 2) Sub-channel degradation mapping mechanism and output alphabet construction: In underground magnetic induction wireless communication systems, the statistical characteristics of each sub-channel exhibit significant unequal distribution due to the substantial influence of geological conductivity, coupling path disturbances, and transmission distance variations in the communication channel environment. Traditional polar codes, built under the assumption of independent and identically distributed polarization, struggle to adapt to such complex channel structures, leading to increased reliability assessment errors, inaccurate bit allocation, and consequently affecting overall bit error rate performance. Therefore, this embodiment proposes a degradation modeling method suitable for unequally distributed sub-channels. By introducing an intermediate modeling mechanism and an output alphabet structure optimization method, it achieves fidelity-preserving dimensionality reduction of channel statistical characteristics, providing stable input for subsequent polar code construction.
[0045] Specifically, this embodiment proposes a flexible degradation mapping mechanism, which involves continuously outputting channels (original channels). (in , This is converted into a discrete output channel (degraded signal). (in (using a finite alphabet) significantly reduces the computational complexity of subsequent polar code construction while maintaining channel capacity.
[0046] This embodiment first introduces an intermediate degradation channel. Used to preserve the original channel Under the premise of statistical properties, achieve a finite alphabet representation of the output symbol. Among them, This refers to the nth transmitted bit in the i-th time slot (i.e., the nth bit of the input to the original channel in the i-th time slot). This refers to the nth received bit in the i-th time slot (i.e., the nth bit of the original channel output in the i-th time slot). for The corresponding discretized output symbol; This degradation process is caused by intermediate channels. The decision, its mathematical structure, satisfies the following form:
[0047] In the above formula, This indicates that the original channel has the following input: The output probability function at that time, To make continuous symbols Mapping to discrete symbols The mapping probability.
[0048] To more accurately represent the above continuous symbols Mapping to discrete symbols The mapping relationship, the present invention will The domain is defined according to the channel capacity function Divided into multiple sub-intervals And select a representative symbol for each interval. , its in The probability of this is given by the following formula:
[0049] in, ; It is a normalization factor used to correct the consistency of output density of different sub-channels during the discretization process, thereby ensuring that the degraded channel is as close as possible to the statistical characteristics of the original channel in a statistical sense.
[0050] The output symbol set of a degraded channel still has a high order of magnitude, which is detrimental to bit decision path management in subsequent polarization transformation operations. Therefore, this embodiment further designs an output symbol merging mechanism based on likelihood ratio sorting, specifically including: firstly defining the original channel output symbols... The conditional probabilities for inputs of 0 and 1, and their likelihood ratios calculated accordingly:
[0051] However, within each sub-channel, the set of output symbols is... according to Sort the alphabet in ascending order to ensure numerical continuity in the output alphabet during subsequent merging processes. After sorting, the alphabet is then merged according to the preset target alphabet size. The system selects adjacent symbol pairs from the sorted set for merging. To minimize the disturbance to capacity estimation in the polarization structure caused by merging, the merging rule aims to minimize the change in the Bartholin's parameter before and after merging, selects the optimal merging path, and updates the output set.
[0052] After completing multiple rounds of symbol merging, a degraded channel output alphabet and a degraded transfer function are constructed.
[0053] The specific process is as follows: A1. Calculation Each received bit in The log-likelihood ratio of each transmitted bit is calculated, and the resulting log-likelihood ratio is divided into L equal-width intervals. These L equal-width intervals are then compared with L distinct primary symbols. One-to-one correspondence; will The corresponding partition is divided into L subsets of transmitted bits, each corresponding to one of L equal-width intervals; The corresponding data is divided into L subsets of received bits, each corresponding to one of L equally wide intervals; L ≥ 2. ; For a pre-defined set of discrete numerical values; initialize the set. Initialize the degraded channel Q; A2. Determine if the iteration deadline condition is met. If yes, proceed to A4; otherwise, calculate the merging cost of every two adjacent primary symbols in the current set S; where, and Merger costs ; Indicates a merge operation; ; The number of primary symbols in the current set S; Primary symbols The specific parameters of the Bartholomew's Law are as follows: ; and They are respectively The number of transmitted bits with a value of 0 and the number of transmitted bits with a value of 1 in the corresponding subset of transmitted bits; The new symbol after merging, its Bartholin's parameter ; and They are respectively The corresponding subset of transmitted bits and The number of transmitted bits with values of 0 and 1 in the merged set of the corresponding subset of transmitted bits; and The output bits are respectively the output bits when the input bit is 0 and the output bit is 1 under the current degraded channel. The transition probability; ; The deadline conditions include: The number of polar code information bits is less than or equal to the number of bits, and the capacity loss between the current degraded channel and the original channel (i.e., the difference in Shannon capacity between the degraded channel and the original channel) is greater than or equal to the corresponding upper bound. The upper bound of the capacity loss between the current degraded channel and the original channel is: Where K is the number of the currently received subset of bits; This represents the current k-th received bit subset; for Received bits in The marginal density function; Let KL divergence be a metric. For the original channel, the input is The output for each transmitted bit is the received bit. The vector formed by the transition probabilities; For the current degraded channel, the input is The output is when each bit is transmitted. The vector formed by the transition probabilities; for The corresponding primary symbols; A3. Merge the two adjacent primary symbols with the minimum merging cost in set S, merge the intervals corresponding to these two adjacent primary symbols, merge the transmit bit subsets corresponding to these two adjacent primary symbols, and merge the receive bit subsets corresponding to these two adjacent primary symbols; update the degraded channel. Proceed to A2; To represent the i-th subchannel, this includes the probability of the transmitted bit in the i-th time slot mapping to the primary symbol; specifically, the transmitted bit in the i-th time slot... Mapped to the corresponding primary symbols The probability of degradation transfer function ; The input under the original channel is The output is The probability value at that time; To receive bits as preset Mapped to primary symbols The probability of; A4. Treat the current degraded channel as the degraded channel after degrading the original channel, and use the current set S as the primary symbol output set.
[0054] Using the modeling methods described above, this embodiment achieves the degradation modeling process of unequally distributed sub-channels in underground magnetic induction wireless communication by introducing an intermediate modeling function, an output alphabet discretization mechanism, and a likelihood ratio-driven symbol merging strategy. Compared to existing uniform channel modeling methods, this scheme can significantly improve the accuracy of polarization structure construction and coding resource utilization efficiency, while enhancing the robustness and adaptability of the overall communication system while ensuring controllable computational complexity.
[0055] This embodiment proposes a degradation modeling method for unequally distributed sub-channels. By utilizing the intermediate degradation channel and the output alphabet discretization mechanism, the continuous channel output is mapped to a finite set of symbols. Combined with a likelihood ratio-driven symbol merging strategy, the complexity of channel modeling is reduced.
[0056] 3) Sub-channel evolutionary mapping mechanism and enhanced symbol mapping strategy After completing the degradation modeling of unequally distributed sub-channels, and considering the complex statistical characteristics and significant reliability differences of sub-channels in underground magnetic induction wireless communication systems, this embodiment further proposes a channel evolution mapping mechanism to improve the resolution and system reliability of sub-channels during polar code construction. This mechanism aims to enhance the resolution and capacity representation accuracy of sub-channels. The method constructs an upgraded channel... Combined with intermediate channels The introduced reversible mapping mechanism achieves enhanced approximation of the original degraded channel, expanding the output alphabet while preserving statistical consistency, so that each sub-channel has better reliability differentiation capability during polarization.
[0057] In the channel evolution mapping designed in this embodiment, for each degraded subchannel C, M primary symbols are selected from the primary symbol output set to form a subset of the output primary symbols of subchannel C. M≥1; Compute the primary symbol subset of the output of subchannel C The reliability metric for each symbol in the code.
[0058] Perform an evolutionary mapping on sub-channel C such that the evolved sub-channel C satisfies: When the input symbol of subchannel C is 0, the preset probability is used respectively. The first M-1 symbols should be output. With probability Output the Mth symbol ; When the input symbol of subchannel C is 1, the preset probability is used respectively. The first M-1 symbols should be output. With probability Output the Mth symbol ; in, ; ; ; ; satisfy: ; This is the reliability metric for the m-th symbol.
[0059] The above design ensures that the evolutionary channel does not disrupt the original reliability order. Only the output probabilities of the first M-1 symbols are defined, and the last term is supplemented through normalization to reduce redundancy and minimize parameters.
[0060] Through the aforementioned evolutionary mapping process, each sub-channel can more precisely distinguish different input signals, improving the accuracy of signal transmission. This improvement enables more accurate allocation of information bits and freeze bits during polar code encoding, thereby enhancing the reliability and transmission distance of the entire communication system.
[0061] In this embodiment, taking M=2 or 3 as an example, the likelihood ratio is used to represent the reliability metric. First, the likelihood ratios of the symbols in the discrete output alphabet are identified. Based on this, the required channel evolution upgrade mapping structure is determined.
[0062] When M=2, the output alphabet contains two symbols. , , respectively, are the first symbols of the channel output. and the second symbol The corresponding primary symbols are denoted as follows, and their likelihood ratios are respectively and ,in, At this point, the following two sets of parameters are defined for constructing channel evolution:
[0063] in, and These represent the channel in the i-th time slot at the input symbol level. and When, output the first symbol. The conditional probability. Specifically, , Represents the first output symbol of the channel in the i-th time slot. The probability values under different input conditions.
[0064] By introducing the above parameters and The transfer function of the constructed evolutionary channel The expression is as follows:
[0065] Among them, the selection function Defined as: ; for Conjugate; For the degraded channel input is Output primary symbols The conditional probability.
[0066] When M=3, the output alphabet contains three symbols. , , , which are the first symbol, the second symbol and the third symbol of the channel output, respectively. Denote their likelihood ratios as and , and , respectively. Among them, . Further preferably, the following four groups of parameters are defined as follows:
[0067] Based on the above parameters, the constructed evolutionary channel transfer function is:
[0068] where the selection function is defined as: ; is the conditional probability that the output primary symbol is when the input of the degraded channel is
[0069] To further ensure the reversible mapping between the evolutionary channel and the original channel, in this embodiment, when constructing the evolutionary channel transfer function (i.e., determining the probability set ), a two-stage intermediate channel mechanism is also introduced. The first stage is the mapping function , which is defined as:
[0070] where f(y|0) is the conditional probability density that the original channel outputs y when the input bit is 0; f(y|1) is the conditional probability density that the original channel outputs y when the input bit is 1; π_i is a normalization factor used to ensure that the sum of probabilities in each interval is 1; A_i = {y: (i - 1) / ν < C[λ(y)] ≤ i / ν}, where ν is the preset total number of quantization intervals; λ(y) is the likelihood ratio of the received signal y, defined as λ(y) = f(y|0) / f(y|1); the role of this mapping function is to divide and map the original continuous output symbol y according to its channel capacity to the discrete symbol set {z1, z2, ⋯, z M} processed by the subsequent evolutionary mapping. This stage provides discrete input symbols for the evolutionary mapping. After completing the preparation from the continuous signal to the discrete symbol set through the mapping function , the transfer function of the evolutionary channel can be specifically constructed based on this symbol set, that is: assign the transition probabilities when the input is 0 and 1 to each symbol and , respectively, so as to completely define the evolutionary channel .
[0071] The second stage is the probability inverse mapping channel. Defined as:
[0072] in, This is used to ensure that the statistical properties of the channel are better recovered during the process of regressing from the evolved channel to the original channel. It ensures that the statistical properties of the channel are better recovered based on discrete symbol sets and probability sets. The defined evolutionary channel. Let α be the likelihood ratio of the received signal; Let be the threshold parameter for the i-th interval; This represents the discretized output symbol.
[0073] Therefore, the two-stage mechanism and the evolutionary mapping process form a coherent whole: the mapping function This is a preliminary step in constructing the evolutionary channel transfer function, better preparing structured discrete inputs for the evolutionary mapping; while the probabilistic inverse mapping channel... These are subsequent verification and safeguard steps that further ensure the reversibility and reliability of the evolutionary mapping.
[0074] This two-stage intermediate channel mechanism, combined with the mapping of evolutionary and degradation channels, not only further improves the resolution and capacity of the channel, but also further ensures the maximum application of channel reliability in the polarization structure, and further optimizes the bit allocation strategy and bit error rate performance.
[0075] This invention establishes a bidirectional mapping relationship. First, a mapping relationship is established from the original output to the evolved output, dividing continuous signal values into different symbol intervals. Then, an inverse mapping relationship is established from the evolved output back to the original output, ensuring that the evolved channel can accurately reflect the statistical characteristics of the original channel, thereby ensuring that the evolution process does not affect the characteristics of the original channel.
[0076] 4) Flexible coding construction framework based on reliability fusion method In underground magnetic induction wireless communication systems, achieving efficient communication requires accurate assessment of sub-channel reliability and rational resource allocation, based on channel degradation modeling methods and channel evolution and enhancement symbol mapping strategies. During system operation, the transceiver coils transmit signals in a complex underground environment. Due to environmental factors, the characteristics of each sub-channel vary significantly; therefore, accurate estimation of sub-channel reliability is crucial for coding design and improving communication performance.
[0077] Regarding the reliability fusion method, this embodiment combines the discretization of the full transition probability matrix with entropy-weighted calibration of the upper and lower bounds of the Barthelite parameters. First, a baseline value is calculated using a finite-iteration Monte Carlo method to provide a reference for boundary adjustments. Based on the decoder's received symbols... With bit signal-to-noise ratio To characterize the reliability differences between sub-channels, Integrate into the discretization process.
[0078] The continuous output channel is transformed into a finite output representation through a likelihood ratio mapping guided by channel capacity, where the follow-likelihood ratio is:
[0079] in, .
[0080] To measure and classify the reliability of sub-channels, it is necessary to quantify the statistical characteristics of the channel output symbols. Based on the established channel model, the received symbols... The log-likelihood ratio (LLR) is:
[0081] The corresponding channel capacity function is derived as follows:
[0082] In this embodiment, a capacity uniform quantization strategy is introduced to achieve the mapping of output symbols to a finite alphabet. Let the total quantization interval be... Then the boundary can be obtained by solving the inverse capacity function. satisfy This operation maps a continuous space to a set of capacity segments. This provides a standardized basis for subsequent derivation of the Bach parameters and merging of output symbols. More preferably, the calculation of the... The transition probability of each output interval:
[0083] Additive Gaussian noise and input offset Under these conditions, and more preferably, a transition matrix can be constructed. To reduce redundancy, capacity distortion is minimized through information entropy-driven optimization.
[0084] in, By iteratively merging adjacent output intervals, the simplified alphabet is ensured to retain the statistical characteristics of the original channel.
[0085] This embodiment estimates the upper and lower bounds of the Barton parameters for degraded and evolved channels based on a discretized model, for degraded sub-channels. Based on the joint probability distribution of the discretized channel transfer function and the polar code, the upper bound of the Bach parameter can be obtained as follows:
[0086] To improve the accuracy of reliability estimation, a flexible and adjustable weight fusion mechanism is introduced. The upper and lower bounds of the Bartholin's parameter are linearly combined to form the calibrated Bartholin's parameter, expressed as: Among them, weight and The conditional entropy difference is dynamically adjusted. Specifically, the Parshall parameter obtained from a relatively small number of Monte Carlo simulations is used as the baseline value. The conditional entropy difference before and after discretization is calculated based on the weights, and a normalization coefficient is introduced. and Ensure that the weights are within a controllable range.
[0087] For weight optimization, the minimum mean square error (MMSE) criterion is used to calibrate the weights. The difference in output probability density between the original channel and the discrete channel is calculated by integration, defined as:
[0088] in, To further optimize the numerical stability factor, an objective function is constructed. Iterative optimization using gradient descent method and until convergence to the preset accuracy. The final expression after optimizing the Bach parameters is obtained as follows:
[0089] To achieve flexible allocation of information bits and freeze bits, all sub-channels are first configured according to the calibrated Parshall parameters. Sort in ascending order and select the first... One sub-channel is used as the information bit, and the remaining sub-channels are fixed as frozen bits.
[0090] Based on this, real-time path loss can also be calculated. The expression for flexibly adjusting the bitrate is:
[0091] This allows for a trade-off between performance and resource consumption under different channel conditions in underground magnetic induction wireless communication systems.
[0092] This invention proposes a degradation modeling method for unequally distributed sub-channels. By utilizing intermediate degradation channels and an output alphabet discretization mechanism, continuous channel outputs are mapped to a finite set of symbols. Combined with a likelihood ratio-driven symbol merging strategy, the complexity of channel modeling is reduced. This invention designs a channel evolution modeling and enhanced symbol mapping strategy. By introducing reversible mapping intermediate channels, the resolution and capacity expression accuracy of sub-channels are improved. Finally, a dynamic reliability fusion method and sub-channel scheduling mechanism are constructed. By combining upper and lower bound calibration of the Bach parameter and code rate adjustment, the allocation of frozen information bits is optimized and the bit error rate is reduced.
[0093] Through the above technical solution, the present invention effectively solves the problem of inaccurate reliability assessment of traditional polar codes in non-uniform channels. Under the premise of ensuring controllable computational complexity, it significantly improves the robustness and communication distance of the system in highly dynamic magnetic induction environments.
[0094] The polar code construction method provided by this invention is characterized in terms of computational complexity as follows: It has significantly lower computational complexity than the Monte Carlo method under the same communication performance metrics. Specifically, such as Figure 2 The image shows a scatter plot of the upper bound, lower bound, and Bartholomew's parameters obtained by the Monte Carlo algorithm. Figure 3 The image shows a scatter plot of the Barcol parameters obtained by the Monte Carlo algorithm and the Barcol parameters after MMSE correction; as shown. Figure 4 The image shows a scatter plot comparing the Parshall parameters after correction using MMSE at different code lengths; as shown. Figure 5 The diagram shows a simulation comparison of communication performance at different code rates and code lengths; for example... Figure 6 The diagram shows a simulation comparison of reliable communication distance and communication performance under different code rates and code lengths. Simulation results show that after calibration... The mean squared error of the results compared to Monta Carlo is less than 5%. At a code length of 1024 and a code rate of 0.75, the bit error rate is improved by 1.5 dB, the frame error rate is improved by 1.8 dB, and the reliable communication distance is extended by 0.5 meters in high signal-to-noise ratio scenarios.
[0095] In summary, this invention aims to address the diverse needs and complex channel conditions encountered in underground magnetic induction wireless communication (MI-WUC). The proposed method successfully achieves a balance between communication performance and computational complexity by introducing a channel likelihood ratio mechanism and allowing adjustable sub-channel structures and output quantization strategies. This method introduces a channel degradation and evolution mapping mechanism by fusing channel statistical characteristics, and obtains a set of flexibly adjustable coding construction frameworks based on the mapped Bach parameters, optimizing the original coding construction scheme under underground channels and significantly improving the reliability and robustness of the underground magnetic induction communication system. Specifically, it includes: constructing a non-stationary magnetic induction channel model to quantify the impact of the heterogeneity of the geological medium on channel gain, and realizing channel modeling by defining a time-varying conductor distribution function; proposing a degradation modeling and symbol optimization strategy for unequally distributed sub-channels, and optimizing channel reliability through probability mapping and likelihood ratio ranking strategies; utilizing a channel evolution mechanism to improve sub-channel resolution and capacity differentiation accuracy, enhancing the reliability of information bits; and proposing upper and lower bounds for the Bach parameters obtained based on the mapping mechanism, providing a coding construction framework that balances communication performance and computational complexity. Experimental results show that the method of the present invention can effectively improve the robustness of the communication system, reduce computational complexity, and extend the communication distance of the system. It significantly improves the reliability and adaptability of the underground magnetic induction wireless communication system in dynamic magnetic induction channels and is suitable for resource-constrained real-time communication scenarios.
[0096] In a second aspect, the present invention provides a polar code construction system for underground non-stationary channels, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor executes the polar code construction method provided in the first aspect of the present invention when executing the computer program.
[0097] The related technical solutions are the same as the polar code construction method provided in the first aspect of this invention, and will not be described in detail here.
[0098] Thirdly, the present invention also provides a computer-readable storage medium comprising a stored computer program, wherein the computer program, when executed by a processor, controls the device where the storage medium is located to execute the polar code construction method provided in the first aspect of the present invention.
[0099] The related technical solutions are the same as the polar code construction method provided in the first aspect of this invention, and will not be described in detail here.
[0100] Fourthly, the invention also provides a computer program product, including a computer program / instruction that, when executed by a processor, implements the polar code construction method provided in the first aspect of the invention.
[0101] The related technical solutions are the same as the polar code construction method provided in the first aspect of this invention, and will not be described in detail here.
[0102] Fifthly, the invention also provides an information processing terminal equipped with the polar code construction system provided in the second aspect of the invention.
[0103] The relevant technical solutions are the same as the polar code construction system provided in the second aspect, and will not be repeated here.
[0104] Sixthly, the invention also provides a network information processing platform equipped with the polar code construction system provided in the second aspect of the invention.
[0105] The relevant technical solutions are the same as the polar code construction system provided in the second aspect, and will not be repeated here.
[0106] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements 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 polar codes for underground non-stationary channels, characterized in that, Used in underground magnetic induction communication systems; The underground magnetic induction communication system includes: a transmitting coil and a receiving coil with parallel cross-sections and placed coaxially; the polar code construction method includes: The number of iterations in the Monte Carlo algorithm is set to a preset number. The Monte Carlo algorithm is then used to calculate the Barcol parameters of each sub-channel in the original channel, which serve as the reference values for the Barcol parameters of each sub-channel. The original channel of the underground magnetic induction communication system is degraded and mapped to obtain a degraded channel, thereby mapping the continuous channel output of the original channel to a finite symbol set. The degraded channel is then evolved and mapped to obtain an evolved channel, thereby establishing an inverse mapping relationship from the output of the evolved channel back to the output of the original channel. The Barton parameters of each sub-channel in the degraded channel are calculated as upper bounds of the Barton parameters of each sub-channel. The Barton parameters of each sub-channel in the evolved channel are also calculated as lower bounds of the Barton parameters of each sub-channel. The upper and lower bounds of the Barton parameters of each sub-channel are weighted and summed to obtain the calibration value of the Barton parameters of that sub-channel. The weights are optimized by minimizing the difference loss between the calibrated value of the Barthelium parameter of each sub-channel and the corresponding reference value, and the calibrated value of the Barthelium parameter of each sub-channel obtained based on the optimal weight is taken as its final value. The final values of the Parvian parameters for each subchannel are sorted, and the subchannel indices corresponding to the top RN Parvian parameter final values with the largest values are used as frozen bit indices, while the subchannel indices corresponding to the remaining Parvian parameter final values are used as information bit indices; R is the code rate of the polar code; N is the length of the original channel; Signal to be transmitted X The transmit bit under the middle freeze bit index is set to 0, and the transmit bit under the information bit index is set to 1, in order to... X Encode the vector to obtain the encoded vector, and then combine the encoded vector with the generator matrix G of the polar code of length N. N Multiplying these together yields the polar code encoding of the underground magnetic induction communication system channel.
2. The polar code construction method according to claim 1, characterized in that, The degradation mapping of the original channel of the underground magnetic induction communication system includes: A1. Calculation Each received bit in The log-likelihood ratio of each transmitted bit is calculated, and the resulting log-likelihood ratio is divided into L equal-width intervals. These L equal-width intervals are then compared with L distinct primary symbols. One-to-one correspondence; will The corresponding partition is divided into L subsets of transmitted bits, each corresponding to one of L equal-width intervals; The corresponding data is divided into L subsets of received bits, each corresponding to one of L equally wide intervals; L ≥ 2. ; For a pre-defined set of discrete numerical values; initialize the set. Initialize the degraded channel Q; A2. Determine if the iteration deadline condition is met. If yes, proceed to A4; otherwise, calculate the merging cost of every two adjacent primary symbols in the current set S; where, and Merger costs ; Indicates a merge operation; ; The number of primary symbols in the current set S; Primary symbols The specific parameters of the Bartholomew's Law are as follows: ; and They are respectively The number of transmitted bits with a value of 0 and the number of transmitted bits with a value of 1 in the corresponding subset of transmitted bits; The new symbol after merging, its Bartholin's parameter ; and They are respectively The corresponding subset of transmitted bits and The number of transmitted bits with values of 0 and 1 in the merged set of the corresponding subset of transmitted bits; and The output bits are respectively the output bits when the input bit is 0 and the output bit is 1 under the current degraded channel. The transition probability; ; The deadline conditions include: The number of polar code information bits is less than or equal to the number of bits, and the capacity loss between the current degraded channel and the original channel is greater than or equal to the corresponding upper bound; The upper bound of the capacity loss between the current degraded channel and the original channel is: Where K is the number of the currently received subset of bits; This is the current k-th received bit subset; for Received bits in The marginal density function; Let KL divergence be a metric. For the original channel, the input is The output for each transmitted bit is the received bit. The vector formed by the transition probabilities; For the current degraded channel, the input is The output is when each bit is transmitted. The vector formed by the transition probabilities; for The corresponding primary symbols; A3. Merge the two adjacent primary symbols with the minimum merging cost in set S, merge the intervals corresponding to these two adjacent primary symbols, merge the transmit bit subsets corresponding to these two adjacent primary symbols, and merge the receive bit subsets corresponding to these two adjacent primary symbols; update the degraded channel. Proceed to A2; To represent the i-th subchannel, this includes the probability of the transmitted bit in the i-th time slot mapping to the primary symbol; specifically, the transmitted bit in the i-th time slot... Mapped to the corresponding primary symbols The probability of: ; The input under the original channel is The output is The probability value at that time; To receive bits as preset Mapped to primary symbols The probability of; A4. Treat the current degraded channel as the degraded channel after degrading the original channel, and use the current set S as the primary symbol output set.
3. The polar code construction method according to claim 2, characterized in that, The evolutionary mapping of the degraded channel includes performing the following on each sub-channel C in the degraded channel: M primary symbols are selected from the primary symbol output set to form a subset of the output primary symbols of subchannel C. M≥1; Perform an evolutionary mapping on sub-channel C such that the evolved sub-channel C satisfies: When the input symbol of subchannel C is 0, the preset probability is used respectively. The first M-1 symbols should be output. With probability Output the Mth symbol ; When the input symbol of subchannel C is 1, the preset probability is used respectively. The first M-1 symbols should be output. With probability Output the Mth symbol ; in, ; ; ; ; satisfy: ; This is the reliability metric for the m-th symbol.
4. The polar code construction method according to claim 3, characterized in that, M is 2 or 3.
5. The polar code construction method according to claim 3, characterized in that, The reliability metric for the m-th symbol is represented by its likelihood ratio.
6. A polar code construction system for underground non-stationary channels, characterized in that, include: A memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the polar code construction method according to any one of claims 1-5.
7. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein when the computer program is run by a processor, it controls the device where the storage medium is located to execute the polar code construction method according to any one of claims 1-5.
8. A computer program product, characterized in that, Includes a computer program / instruction that, when executed by a processor, implements the polar code construction method according to any one of claims 1-5.
9. An information processing terminal equipped with the polar code construction system of claim 6.
10. A network information processing platform equipped with the polar code construction system of claim 6.
