Channel decoding method based on channel coding auxiliary imitation radio frequency division multiplexing

Abstract

The invention discloses a channel decoding method based on channel coding auxiliary imitated radio frequency division multiplexing, which comprises the following steps of: firstly, carrying out channel coding on an imitated radio frequency division multiplexing system sending sequence, then detecting and decoding a received frequency shift symbol by combining a sum-product iteration detector and a turbo iteration decoder based on channel coding, and outputting an estimated sending sequence. According to the method, the reliability of the channel coding assisted imitation radio frequency division multiplexing system is improved to the greatest extent, and the method is suitable for almost all channel coding and is high in expandability.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and more specifically, to a channel decoding method based on channel coding-assisted pseudo-radio frequency division multiplexing. Background Art

[0002] In an AFDM (Affine Frequency Division Multiplexing) wireless communication system, the receiving end needs to detect the demodulated signal based on the channel state information to recover the symbol sequence sent by the transmitting end. This process is called signal detection. In addition, in order to improve the reliability of the system, it is necessary to perform channel coding on the transmitted bit sequence before symbol mapping at the transmitting end, and to perform channel decoding on the bit sequence after signal detection and symbol inverse mapping at the receiving end. At present, research on AFDM waveforms has not considered channel coding. How to implement high-performance signal detection and channel coding is an urgent problem to be solved in AFDM wireless communication systems. Summary of the invention

[0003] The present invention solves the problem of how to implement high-performance signal detection and channel coding in a radio frequency division multiplexing wireless communication system and proposes a channel decoding method based on channel coding assisted radio frequency division multiplexing.

[0004] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0005] The present invention proposes a channel decoding method based on channel coding assisted RFDM, comprising:

[0006] S1: Obtain the prior probability of receiving frequency domain symbols and sending symbols;

[0007] S2: perform signal detection and calculation on the received frequency domain symbols according to the prior probability of the transmitted symbols to obtain a bit log-likelihood ratio sequence;

[0008] S3: Obtain a decoded bit log-likelihood ratio sequence and an estimated transmitted bit sequence according to the bit log-likelihood ratio sequence;

[0009] S4: Obtain a symbol log likelihood ratio according to the decoded bit log likelihood ratio sequence, and calculate a priori probability estimation value of the transmitted symbol according to the symbol log likelihood ratio;

[0010] S5: Update the prior probability of the transmitted symbol using the estimated value of the prior probability of the transmitted symbol, and repeat steps S2-S4; when the preset maximum number of iterations is reached, output the estimated transmitted bit sequence corresponding to the maximum number of iterations.

[0011] Preferably, in S1, the acquiring of received frequency domain symbols comprises:

[0012] Obtain an input bit sequence μ, and input μ into a channel encoder for channel coding. The channel coding includes selecting the largest part of the coding method to first encode μ to obtain a preprocessed bit sequence, and then using a set interleaving sequence to interleave the coded bit sequence to obtain a channel coded bit sequence, and performing orthogonal amplitude modulation symbol mapping on the channel coded bit sequence to obtain a frequency domain symbol x to be sent. After x is transmitted through a wireless channel, the receiving end obtains a received frequency domain symbol:

[0013] y=H eff x+w

[0014] Among them, y is the received frequency domain symbol, H eff represents the equivalent channel matrix, and w represents the frequency domain Gaussian white noise.

[0015] Preferably, in S2, performing signal detection and calculation on the received frequency domain symbols according to the prior probability of the transmitted symbols to obtain a bit log-likelihood ratio sequence includes:

[0016] Inputting the prior probability of the transmitted symbol and the received frequency domain symbol into the sum-product algorithm detector for detection, obtaining the estimated value of the transmitted frequency domain symbol, and calculating the posterior probability of the transmitted symbol according to the estimated value of the transmitted frequency domain symbol;

[0017] Perform constellation inverse mapping according to the posterior probability of the transmitted symbol to obtain an inverse mapping bit sequence, and calculate a bit log-likelihood ratio sequence according to the inverse mapping bit sequence.

[0018] Preferably, the steps of implementing the sum-product algorithm detector include:

[0019] S21: Input received frequency domain symbols, modulation constellation diagram Q, equivalent channel matrix H eff , set the number of discrete affine Fourier transform points N, the total number of paths P, and the prior probability of sending symbols;

[0020] S22: Based on the received frequency domain symbol, Q, H eff Calculate the posterior probability of the received frequency domain symbols estimated by all paths using N, obtain the posterior probability of the received frequency domain symbols of all paths, and calculate the likelihood function according to the posterior probability of the received frequency domain symbols of all paths;

[0021]

[0022] Among them, Pr{y|x[m]} represents the likelihood function, represents the posterior probability of the received frequency domain symbol of the i-th path, represents the posterior probability of the transmitted frequency domain symbol of the i-th path;

[0023] S23: Calculate the posterior probability of the transmitted frequency domain symbol according to the likelihood function, make a hard decision on the posterior probability of the transmitted frequency domain symbol, and obtain the estimated transmitted frequency domain symbol. The specific calculation formula is:

[0024]

[0025] in, represents the estimated value of the mth transmitted frequency domain symbol x[m], Pr{x[m]|y} is the posterior probability of the transmitted frequency domain symbol, which specifically means the probability of x[m] taking any orthogonal amplitude modulation symbol in the modulation constellation Q under the condition of known signal y. According to the Bayesian chain rule, we can further obtain:

[0026] Pr{x[m]|y}∝Pr{y|x[m]}Pr{x[m]}

[0027] Among them, Pr{x[m]} represents the prior probability of the transmitted symbol, and the probability that the mth transmitted frequency domain symbol x[m] takes any orthogonal amplitude modulation symbol in the constellation table Q, Pr{y|x[m]}Pr{x[m]} is:

[0028]

[0029] Where i represents the index of the i-th path, y m [i] represents the mth received frequency shift symbol of the i-th branch path,

[0030] Preferably, in S22, according to the received frequency domain symbol, Q, H eff The posterior probability of the received frequency domain symbols estimated by all paths is calculated by and N, and the posterior probability of the received frequency domain symbols of all paths is obtained, including:

[0031] Find the received frequency shift symbol y according to the constellation diagram Q m The set of transmitted frequency domain symbols All combinations of Calculate the a posteriori probability of the received frequency domain symbol of the i-th path The posterior probability of the transmitted frequency domain symbol of the i-th path is The specific calculation formulas are:

[0032]

[0033] Where N0 represents the noise power and ξ(m,i) is:

[0034]

[0035] Preferably, in S3, obtaining a decoding bit log-likelihood ratio sequence and an estimated transmission bit sequence according to the bit log-likelihood ratio sequence comprises:

[0036] The bit log-likelihood ratio sequence is deinterleaved using the inverse interleaved sequence corresponding to the set interleaved sequence to obtain the inverse interleaved bit log-likelihood ratio, and the inverse interleaved bit log-likelihood ratio is input into the optimal decoder corresponding to the encoding method of the transmitting end for decoding to obtain the decoded bit log-likelihood ratio sequence and the estimated transmitted bit sequence.

[0037] Preferably, in S4, obtaining the symbol log-likelihood ratio and the estimated value of the prior probability of the transmitted symbol according to the decoded bit log-likelihood ratio sequence comprises:

[0038] Using a set interleaving sequence to interleave a decoding bit log-likelihood ratio sequence to obtain an interleaved bit log-likelihood ratio, and performing constellation mapping on the interleaved bit log-likelihood ratio sequence to obtain a symbol log-likelihood ratio sequence;

[0039] The prior probability of the transmitted symbol is calculated based on the symbol log-likelihood ratio sequence.

[0040] The present invention proposes a channel decoding device based on channel coding assisted RFDM, the device comprising:

[0041] A data acquisition module, used to obtain a priori probabilities of received frequency domain symbols and transmitted symbols;

[0042] A signal detection module is used to perform signal detection and calculation on the received frequency domain symbols according to the prior probability of the transmitted symbols to obtain a bit log-likelihood ratio sequence;

[0043] A processing module, used for obtaining a decoding bit log-likelihood ratio sequence and an estimated transmission bit sequence according to the bit log-likelihood ratio sequence;

[0044] A transmission symbol prior probability estimation module is used to obtain a symbol log likelihood ratio according to a decoded bit log likelihood ratio sequence, and calculate a transmission symbol prior probability estimation value according to the symbol log likelihood ratio;

[0045] The transmission bit sequence estimation module is used to update the transmission symbol prior probability using the transmission symbol prior probability estimation value and return to the signal detection module; when the preset maximum number of iterations is reached, the estimated transmission bit sequence corresponding to the maximum number of iterations is output.

[0046] The present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned channel decoding method when executing the computer program.

[0047] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the channel decoding method described above are implemented.

[0048] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0049] The present invention proposes a channel decoding method based on channel coding assisted RFDM, firstly, channel coding is performed on the RFDM system transmission sequence, then the received frequency shift symbol is detected and decoded by using the joint sum-product iterative detection based on the channel coding assisted RFDM system and the turbo iterative decoder of the channel coding, and the estimated transmission sequence is output. The present invention maximizes the reliability of the RFDM system assisted by the channel coding, is applicable to almost all channel coding, and has strong scalability. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 Schematic diagram of the flow of the channel decoding method described in Example 1;

[0051] Figure 2 This is a block diagram of a channel coding radio frequency division multiplexing system in Example 2;

[0052] Figure 3 This is a channel decoding block diagram of channel coding assisted RFDM in Example 2;

[0053] Figure 4 Frame error rate simulation result diagram of the AFDM system under different coding settings in Example 2;

[0054] Figure 5 : This is a simulation result diagram of the frame error rate of the AFDM system under different numbers of branch paths in Example 2;

[0055] Figure 6 This is a schematic diagram of the computer device described in Example 2;

[0056] Figure 7 This is a schematic diagram of the structure of the channel decoding device described in Example 3. DETAILED DESCRIPTION

[0057] The drawings are for illustrative purposes only and should not be construed as limiting the present patent;

[0058] In order to better illustrate the present embodiment, some parts in the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product;

[0059] It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0060] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0061] Example 1

[0062] This embodiment provides a channel decoding method based on channel coding assisted RFDM. Figure 1 As shown, including:

[0063] S1: Obtain the prior probability of receiving frequency domain symbols and sending symbols;

[0064] S2: perform signal detection and calculation on the received frequency domain symbols according to the prior probability of the transmitted symbols to obtain a bit log-likelihood ratio sequence;

[0065] S3: Obtain a decoded bit log-likelihood ratio sequence and an estimated transmitted bit sequence according to the bit log-likelihood ratio sequence;

[0066] S4: Obtain a symbol log likelihood ratio according to the decoded bit log likelihood ratio sequence, and calculate a priori probability estimation value of the transmitted symbol according to the symbol log likelihood ratio;

[0067] S5: Update the prior probability of the transmitted symbol using the estimated value of the prior probability of the transmitted symbol, and repeat steps S2-S4; when the preset maximum number of iterations is reached, output the estimated transmitted bit sequence corresponding to the maximum number of iterations.

[0068] In the specific implementation process, the received frequency domain symbols are first detected and calculated based on the initial transmitted symbol prior probability to obtain a bit log-likelihood ratio sequence, and then the estimated value of the transmitted symbol prior probability is further calculated. Then, the transmitted symbol prior probability is updated, and the iteration is repeated multiple times. When the preset maximum number of iterations is reached, the estimated transmitted bit sequence corresponding to the maximum number of iterations is output.

[0069] Example 2

[0070] This embodiment proposes a channel decoding method based on channel coding assisted RFDM, including:

[0071] S1: Obtain the prior probability of receiving frequency domain symbols and sending symbols;

[0072] like Figure 2As shown, it is a block diagram of a channel coding radio frequency division multiplexing system, where μ is an input bit sequence, which is input into a channel encoder for channel coding, including first encoding μ according to a selected coding method. This embodiment is applicable to most coding methods. The encoded bit sequence is then interleaved (interleaving is to reorder the input sequence using a specific interleaving sequence, and the receiving end has a corresponding inverse interleaving sequence for deinterleaving). x is a signal vector to be transmitted consisting of N quadrature amplitude modulation (QAM) symbols mapped by symbols (constellation) of the channel-coded bit sequence according to the constellation table Q. The transmitter performs a point inverse discrete affine Fourier transform (IDAFT, Inverse Discrete Affine Fourier Transform) on the DAFT domain signal vector x (size is N×1) to obtain a time domain signal vector s (size is N×1), and its specific expression is:

[0073]

[0074] Wherein, n represents a discrete time index, and m represents a DAFT domain (DAFT domain may also be generally referred to as a discrete linear frequency modulation domain) index; is the time domain expression of the mth subcarrier. It can be seen that it is a chirp signal. c1 and c2 are two basic parameters of AFDM, which respectively determine the time-frequency slope and initial phase of the AFDM chirp subcarrier.

[0075] The matrix expression of formula (1) is:

[0076]

[0077] in, is the DAFT modulation matrix, size N×N, the operator H represents the conjugate transpose, the matrix Represents the N-point discrete Fourier transform matrix, the matrix is a diagonal matrix of size N×N, c=c1,c2. Before sending to the channel, a matrix of length l needs to be added in front of s. max The chirp period prefix CPP (chirp-periodic prefix), where l max is the maximum normalized delay spread of the channel.

[0078] The double dispersive wireless channel is modeled in the delay-Doppler domain as follows:

[0079]

[0080] Among them, h(τ,v) indicates that the dual-dispersive wireless channel consists of P paths, and hi , τ i and v i represent the fading coefficient, delay and Doppler of the i-th path respectively, where i = 1,…,P.

[0081] The transmitted time domain signal reaches the receiving end through the channel. The receiving end removes the CPP of the received time domain signal to obtain the time domain signal vector d (size is N×1), and its matrix expression is:

[0082]

[0083] Among them, v is an N×1 vector, which is the received Gaussian white noise item with variance N0. is the equivalent time domain channel matrix, which is N×N in size; P represents the number of channel paths, The time domain channel matrix size of the i-th path is N×N, Π N is a forward cyclic matrix that can simulate channel delay. Its specific expression is:

[0084]

[0085] Among them, Δ N is a one-step frequency shift diagonal matrix, simulating the effect of Doppler frequency shift, and its specific expression is:

[0086]

[0087] Among them, α i and l i are the digital domain normalized Doppler frequency shift and delay, corresponding to v i and τ i Perform N-point discrete affine Fourier transform (DAFT) demodulation on the received time domain vector d to obtain the DAFT domain signal vector y;

[0088] The corresponding matrix expression is:

[0089]

[0090] Substituting equation (2) and equation (5) into the above equation, the input-output relationship of AFDM in the DAFT domain is:

[0091]

[0092] in, Index indicator ind i =(α i +2Nc1l i ) N, the matrix input-output relationship between the receiving DAFT domain vector y and the sending DAFT domain vector x is:

[0093] y=HefrX+w(9)

[0094] Among them, H eff is the equivalent channel matrix, and vector w is the Gaussian white noise vector in the DAFT domain. Assume that the receiving end knows the channel state information of the transmitting end and the receiving end, that is, H eff It has been estimated by conventional channel estimation means at the receiving end. The receiving end can then perform signal detection on y to obtain the estimated transmitted signal vector Again The bit stream with inverse symbol mapping is channel decoded to obtain the estimated bit sequence Will The number of bits that are different from the input bit sequence u initially sent is divided by the total number of bits transmitted and the bit error rate of this transmission.

[0095] S2: perform signal detection and calculation on the received frequency domain symbols according to the prior probability of the transmitted symbols to obtain a bit log-likelihood ratio sequence;

[0096] Inputting the prior probability of the transmitted symbol and the received frequency domain symbol into the sum-product algorithm detector for detection, obtaining the estimated value of the transmitted frequency domain symbol, and calculating the posterior probability of the transmitted symbol according to the estimated value of the transmitted frequency domain symbol;

[0097] Perform constellation inverse mapping according to the posterior probability of the transmitted symbol to obtain an inverse mapping bit sequence, and calculate a bit log-likelihood ratio sequence according to the inverse mapping bit sequence.

[0098] The steps of implementing the sum-product algorithm detector are as follows:

[0099] S21: Input received frequency domain symbols, modulation constellation diagram Q, equivalent channel matrix H eff , set the number of discrete affine Fourier transform points N, the total number of paths P, and the prior probability of sending symbols;

[0100] S22: Based on the received frequency domain symbol, Q, H eff Calculate the posterior probability of the received frequency domain symbols estimated by all paths using N, obtain the posterior probability of the received frequency domain symbols of all paths, and calculate the likelihood function according to the posterior probability of the received frequency domain symbols of all paths;

[0101]

[0102] Among them, Pr{y|x[m]} represents the likelihood function, represents the posterior probability of the received frequency domain symbol of the i-th path, represents the posterior probability of the transmitted frequency domain symbol of the i-th path;

[0103] Find the received frequency shift symbol y according to the constellation diagram Q m The set of transmitted frequency domain symbols All combinations of Calculate the a posteriori probability of the received frequency domain symbol of the i-th path The posterior probability of the transmitted frequency domain symbol of the i-th path is The specific calculation formulas are:

[0104]

[0105] Where N0 represents the noise power and ξ(m,i) is:

[0106]

[0107] S23: Calculate the posterior probability of the transmitted frequency domain symbol according to the likelihood function, make a hard decision on the posterior probability of the transmitted frequency domain symbol, and obtain an estimated transmitted frequency domain symbol:

[0108] The Sum Product Algorithm (SPA) detector is derived from the single-symbol maximum a posteriori criterion (approximately the optimal criterion, and the multi-symbol maximum a posteriori criterion is the optimal criterion), and can achieve approximately optimal detection results. Starting from the single-symbol maximum a posteriori criterion, given the received DAFT domain vector y, let the most likely QAM symbol of the m-th transmitted DAFT domain symbol x[m] be its estimated value, denoted as [m], and the specific expression is:

[0109]

[0110] in, represents the estimated value of the mth transmitted frequency domain symbol x[m], Pr{x[m]|y} is the posterior probability of the transmitted frequency domain symbol, which specifically means the probability of x[m] taking any orthogonal amplitude modulation symbol in the modulation constellation Q under the condition of known signal y. According to the Bayesian chain rule, we can further obtain:

[0111] Pr{x[m]|y}oxPr{y|x[m]}Pr{x[m]}

[0112] Where Pr{x[m]} represents the prior probability of the transmitted symbol, which is the probability that the mth transmitted DAFT domain symbol x[m] obtains any QAM symbol in the constellation table Q. In the absence of prior information, it is generally initialized to an equal probability distribution, that is, Where |Q| represents the number of constellation points, that is, the number of types of QAM symbols that x[m] can take. From equation (8), any received DAFT domain symbol y[m] is composed of P damaged transmitted DAFT domain symbols x[m i ] composition, m i =m+ind i , i = 1,…,P. Similarly, each transmitted DAFT domain symbol x[m] will affect P received DAFT domain symbols, denoted as vector The size is N×1, and the values ​​from i+1 to P are recorded as vectors Further, the i-th symbol that affects the vector, i.e. y m The set of transmitted DAFT domain symbols for [i] (excluding x[m]) is denoted as the vector The size is (N-1) × 1. Then, using the Bayesian chain rule, we have:

[0113]

[0114] Where i represents the index of the i-th path, y m [i] represents the mth received frequency shift symbol of the i-th branch path,

[0115] y\y m represents vector y minus vector y m The vector composed of the elements in, that is, the vector y m Regarding the complement of vector y. Further using Bayes' chain rule:

[0116]

[0117] Since the channel is sparse, can be approximated to Independent of each other, there are further:

[0118]

[0119] The above can be used to estimate the transmitted DAFT domain vector It can be seen that the proposed sum-product algorithm detector finally obtains not only the estimation of the transmitted DAFT domain vector, but also the posterior probability of the detected soft information x[m], that is, the probability Pr{x[m]|y} that x[m] takes any QAM symbol in the constellation table Q under the condition of known vector y. Based on this, the subsequent channel decoding steps can be combined to further improve the bit error rate of the communication system. Since the sum-product algorithm detector starts from the single-symbol maximum a posteriori criterion, the detector achieves a detection effect that achieves near-optimal detection.

[0120] S3: Obtain a decoded bit log-likelihood ratio sequence and an estimated transmitted bit sequence according to the bit log-likelihood ratio sequence;

[0121] The bit log-likelihood ratio sequence is deinterleaved using the inverse interleaved sequence corresponding to the set interleaved sequence to obtain the inverse interleaved bit log-likelihood ratio, and the inverse interleaved bit log-likelihood ratio is input into the optimal decoder corresponding to the encoding method of the transmitting end for decoding to obtain the decoded bit log-likelihood ratio sequence and the estimated transmitted bit sequence.

[0122] S4: Obtain a symbol log likelihood ratio according to the decoded bit log likelihood ratio sequence, and calculate a priori probability estimation value of the transmitted symbol according to the symbol log likelihood ratio;

[0123] Using a set interleaving sequence to interleave a decoding bit log-likelihood ratio sequence to obtain an interleaved bit log-likelihood ratio, and performing constellation mapping on the interleaved bit log-likelihood ratio sequence to obtain a symbol log-likelihood ratio sequence;

[0124] The prior probability of the transmitted symbol is calculated based on the symbol log-likelihood ratio sequence.

[0125] S5: Update the prior probability of the transmitted symbol using the estimated value of the prior probability of the transmitted symbol, and repeat steps S2-S4; when the preset maximum number of iterations is reached, output the estimated transmitted bit sequence corresponding to the maximum number of iterations.

[0126] like Figure 3As shown, it is a channel decoding block diagram of channel coding assisted RFDM proposed in this embodiment. The received DAFT domain vector y is input into an approximately optimal detector, such as the sum-product algorithm detector described in this embodiment, and the posterior probability Pr{x[m]|y} of x[m] output by the detector is obtained by constellation inverse mapping, and the "0" and "1" posterior probabilities of the bit sequence are obtained, that is, the probability that a certain bit takes the value of "0" or "1" (the bit can only take the value of "0" or "1", and the sum of the two probabilities is 1), and its log likelihood ratio is calculated, that is, the logarithmic value of the probability of "0" divided by the probability of "1", and then the inverse interleaved sequence corresponding to the interleaved sequence of the transmitting end is used to deinterleave it, and the input bit log likelihood ratio of the decoder is obtained, and it is input into an optimal decoder corresponding to the encoding method of the transmitting end, such as a BCJR (Bahl-Cocke-Jelinek-Raviv, BCJR) decoder, to obtain the decoder output bit log likelihood ratio. Based on this output bit log-likelihood ratio sequence, the simulation transmitter performs encoding, interleaving, and constellation mapping to obtain the symbol log-likelihood ratio, which can calculate a more accurate prior probability Pr{x[m]} of x[m], and use it as the input of the sum-product algorithm detector to perform signal detection again, and repeat this cycle. Since the sum-product algorithm detector is an approximately optimal detector, the decoder used is also an approximately optimal decoder, so the proposed iterative decoder that combines sum-product iterative detection and channel coding can achieve approximately optimal detection and decoding performance.

[0127] This embodiment verifies the performance of the above-mentioned iterative decoder method of joint sum-product iterative detection and channel coding in the AFDM system through numerical simulation, with the frame error rate (FER) as the evaluation criterion of the system communication performance. The normalized delay and Doppler shift of the path are evenly distributed between [0,3] and [-3,3], respectively. When the carrier frequency is 4GHz, the number of subcarriers is 128, the interval is 500Hz, and the corresponding maximum moving speed is 540 kilometers per hour. Other main simulation parameters are shown in Table 1, where the maximum number of iterations of the sum-product algorithm detector is 10, and the ratio of the received data signal power to the noise power in the time domain is recorded as SNR (signal-to-noise ratio).

[0128] Table 1 System parameter settings

[0129] System Parameters Numeric Carrier frequency 4GHz Subcarrier spacing 500Hz Number of symbols (number of subcarriers) 128 Maximum Doppler shift 1500Hz Maximum movement speed 405kmph Number of multipaths 2 / 3 / 4 Digital modulation BPSK / 4QAM Detector Sum-Product Detector Channel coding type Convolutional Code

[0130] like Figure 4 As shown in FIG. 1 , the simulation result diagram of the frame error rate of the AFDM system under different coding settings is shown, which shows the influence of the bit signal-to-noise ratio on the frame error rate of the AFDM system under different coding settings. is the minimum mean square Euclidean distance corresponding to the convolutional code used. The larger the minimum mean square Euclidean distance, the stronger the error correction performance of the channel coding, and the higher the coding complexity of the transmitter and the decoding complexity of the receiver. For fair comparison, the case where channel coding is not used is provided. It can be seen that under the same bit signal-to-noise ratio, the frame error rate of the case using channel coding is better than that of the case without coding. In addition, as the minimum mean square Euclidean distance of the channel coding used increases, the system frame error rate is lower, which means that the principle of channel coding in the AFDM system is to maximize the minimum mean square Euclidean distance as much as possible. In addition, the degree of improvement in system reliability decreases as the minimum mean square Euclidean distance increases, while the diversity performance of the system does not improve, that is, the slope of the frame error rate curve decreases as the signal-to-noise ratio increases, which is the same as the case of no coding. It is worth mentioning that the above conclusions are not limited to the type of channel coding used, such as convolutional codes, Turbo codes, and low-density parity-check (LDPC) codes used in 5G communication systems.

[0131] like Figure 5 As shown in the figure, the simulation results of the frame error rate of the AFDM system under different numbers of paths show the influence of the bit signal-to-noise ratio on the frame error rate of the AFDM system under different numbers of paths. The number of multipaths is set to 2, 3 and 4 respectively. It can be observed that under different paths, the use of channel coding can improve the frame error rate performance of the system, while the diversity performance is not improved, that is, the slope of the frame error rate curve with the increase of signal-to-noise ratio is the same as that of the uncoded case. In addition, the frame error rate curve of the system increases significantly with the increase of the number of paths, which proves that the proposed sum-product algorithm detector can achieve near-optimal detection capability and obtain the optimal diversity gain of the system. Finally, as the number of paths increases, the improvement of the system frame error rate performance brought by channel coding decreases, that is, the coding gain of the AFDM system decreases with the increase of the number of multipaths. It is worth mentioning that the above conclusions are not limited to the type of channel coding used, such as convolutional codes, Turbo codes, and low-density parity-check (LDPC) codes used in 5G communication systems.

[0132] This embodiment provides a computer device, such as Figure 6 As shown, it includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above-mentioned channel decoding method when executing the computer program.

[0133] This embodiment further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the channel decoding method described above are implemented.

[0134] Example 3

[0135] This embodiment provides a channel decoding device based on channel coding assisted RFDM, such as Figure 7 As shown, the device comprises:

[0136] A data acquisition module, used to obtain a priori probabilities of received frequency domain symbols and transmitted symbols;

[0137] A signal detection module is used to perform signal detection and calculation on the received frequency domain symbols according to the prior probability of the transmitted symbols to obtain a bit log-likelihood ratio sequence;

[0138] A processing module, used for obtaining a decoding bit log-likelihood ratio sequence and an estimated transmission bit sequence according to the bit log-likelihood ratio sequence;

[0139] A transmission symbol prior probability estimation module is used to obtain a symbol log likelihood ratio according to a decoded bit log likelihood ratio sequence, and calculate a transmission symbol prior probability estimation value according to the symbol log likelihood ratio;

[0140] The transmission bit sequence estimation module is used to update the transmission symbol prior probability using the transmission symbol prior probability estimation value and return to the signal detection module; when the preset maximum number of iterations is reached, the estimated transmission bit sequence corresponding to the maximum number of iterations is output.

[0141] The same or similar reference numerals correspond to the same or similar components;

[0142] The terms used in the drawings to describe positional relationships are only used for illustrative purposes and should not be construed as limiting this patent;

[0143] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A channel decoding method based on channel coding assisted RFDM, characterized in that: include: S1: Obtain the prior probability of receiving frequency domain symbols and sending symbols; S2: perform signal detection and calculation on the received frequency domain symbols according to the prior probability of the transmitted symbols to obtain a bit log-likelihood ratio sequence; S3: Obtain a decoded bit log-likelihood ratio sequence and an estimated transmitted bit sequence according to the bit log-likelihood ratio sequence; S4: Obtain a symbol log likelihood ratio according to the decoded bit log likelihood ratio sequence, and calculate a priori probability estimation value of the transmitted symbol according to the symbol log likelihood ratio; S5: Update the prior probability of the transmitted symbol using the estimated value of the prior probability of the transmitted symbol, and repeat steps S2-S4; when the preset maximum number of iterations is reached, output the estimated transmitted bit sequence corresponding to the maximum number of iterations.

2. The channel decoding method based on channel coding assisted RFDM according to claim 1, characterized in that: In the S1, the obtaining of received frequency domain symbols includes: Obtain an input bit sequence μ, and input μ into a channel encoder for channel coding. The channel coding includes selecting the largest part of the coding method to first encode μ to obtain a preprocessed bit sequence, and then using a set interleaving sequence to interleave the coded bit sequence to obtain a channel coded bit sequence, and performing orthogonal amplitude modulation symbol mapping on the channel coded bit sequence to obtain a frequency domain symbol x to be sent. After x is transmitted through a wireless channel, the receiving end obtains a received frequency domain symbol: y=H eff x+w Among them, y is the received frequency domain symbol, H eff represents the equivalent channel matrix, and w represents the frequency domain Gaussian white noise.

3. The channel decoding method based on channel coding assisted RFDM according to claim 2, characterized in that: In S2, signal detection and calculation are performed on the received frequency domain symbols according to the prior probability of the transmitted symbols to obtain a bit log-likelihood ratio sequence, which includes: Inputting the prior probability of the transmitted symbol and the received frequency domain symbol into the sum-product algorithm detector for detection, obtaining the estimated value of the transmitted frequency domain symbol, and calculating the posterior probability of the transmitted symbol according to the estimated value of the transmitted frequency domain symbol; Perform constellation inverse mapping according to the posterior probability of the transmitted symbol to obtain an inverse mapping bit sequence, and calculate a bit log-likelihood ratio sequence according to the inverse mapping bit sequence.

4. The channel decoding method based on channel coding assisted RFDM according to claim 3, characterized in that: The steps of implementing the sum-product algorithm detector include: S21: Input received frequency domain symbols, modulation constellation diagram Q, equivalent channel matrix H eff , set the number of discrete affine Fourier transform points N, the total number of paths P, and the prior probability of sending symbols; S22: Based on the received frequency domain symbol, Q, H eff Calculate the posterior probability of the received frequency domain symbols estimated by all paths using N, obtain the posterior probability of the received frequency domain symbols of all paths, and calculate the likelihood function according to the posterior probability of the received frequency domain symbols of all paths; Among them, Pr{y|x[m]} represents the likelihood function, represents the posterior probability of the received frequency domain symbol of the i-th path, represents the posterior probability of the transmitted frequency domain symbol of the i-th path; S23: Calculate the posterior probability of the transmitted frequency domain symbol according to the likelihood function, make a hard decision on the posterior probability of the transmitted frequency domain symbol, and obtain the estimated transmitted frequency domain symbol. The specific calculation formula is: in, represents the estimated value of the mth transmitted frequency domain symbol x[m], Pr{x[m]|y} is the posterior probability of the transmitted frequency domain symbol, which specifically means the probability of x[m] taking any orthogonal amplitude modulation symbol in the modulation constellation Q under the condition of known signal y. According to the Bayesian chain rule, we can further obtain: Pr{x[m]|y}∝Pr{y|x[m]}Pr{x[m]} Among them, Pr{x[m]} represents the prior probability of the transmitted symbol, and the probability that the mth transmitted frequency domain symbol x[m] takes any orthogonal amplitude modulation symbol in the constellation table Q, Pr{y|x[m]}Pr{x[m]} is: Where i represents the index of the i-th path, y m [i] represents the mth received frequency shift symbol of the i-th branch path, 5. The channel decoding method based on channel coding assisted RFDM according to claim 4, characterized in that: In S22, according to the received frequency domain symbol, Q, H eff The posterior probability of the received frequency domain symbols estimated by all paths is calculated by and N, and the posterior probability of the received frequency domain symbols of all paths is obtained, including: Find the received frequency shift symbol y according to the constellation diagram Q m The set of transmitted frequency domain symbols All combinations of Calculate the a posteriori probability of the received frequency domain symbol of the i-th path The posterior probability of the transmitted frequency domain symbol of the i-th path is The specific calculation formulas are: Where N0 represents the noise power and ξ(m, i) is:

6. The channel decoding method based on channel coding assisted RFDM according to claim 3, characterized in that: In S3, obtaining a decoded bit log-likelihood ratio sequence and an estimated transmitted bit sequence according to the bit log-likelihood ratio sequence includes: The bit log-likelihood ratio sequence is deinterleaved using the inverse interleaved sequence corresponding to the set interleaved sequence to obtain the inverse interleaved bit log-likelihood ratio, and the inverse interleaved bit log-likelihood ratio is input into the optimal decoder corresponding to the encoding method of the transmitting end for decoding to obtain the decoded bit log-likelihood ratio sequence and the estimated transmitted bit sequence.

7. The channel decoding method based on channel coding assisted RFDM according to claim 6, characterized in that: In S4, obtaining the symbol log-likelihood ratio and the estimated value of the prior probability of the transmitted symbol according to the decoded bit log-likelihood ratio sequence includes: Using a set interleaving sequence to interleave a decoding bit log-likelihood ratio sequence to obtain an interleaved bit log-likelihood ratio, and performing constellation mapping on the interleaved bit log-likelihood ratio sequence to obtain a symbol log-likelihood ratio sequence; The prior probability of the transmitted symbol is calculated based on the symbol log-likelihood ratio sequence.

8. A channel decoding device based on channel coding assisted RFDM, characterized in that: The device comprises: A data acquisition module, used to obtain a priori probabilities of received frequency domain symbols and transmitted symbols; A signal detection module is used to perform signal detection and calculation on the received frequency domain symbols according to the prior probability of the transmitted symbols to obtain a bit log-likelihood ratio sequence; A processing module, used for obtaining a decoding bit log-likelihood ratio sequence and an estimated transmission bit sequence according to the bit log-likelihood ratio sequence; A transmission symbol prior probability estimation module is used to obtain a symbol log likelihood ratio according to a decoded bit log likelihood ratio sequence, and calculate a transmission symbol prior probability estimation value according to the symbol log likelihood ratio; The transmission bit sequence estimation module is used to update the transmission symbol prior probability using the transmission symbol prior probability estimation value and return to the signal detection module; when the preset maximum number of iterations is reached, the estimated transmission bit sequence corresponding to the maximum number of iterations is output.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the channel decoding method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the channel decoding method according to any one of claims 1 to 7 are implemented.

Images