A coding index modulation method, apparatus, and device based on resource separation transmission

Abstract

This invention provides a coding index modulation method, apparatus, and device based on resource-separated transmission. It separates the information bits (information bits in the original scheme) and check bits for physical layer resource-separated transmission (information bits are carried in MDCSK modulation symbols, and check bits are carried in time slot indices). Simultaneously, a joint soft-decision iterative mechanism is implemented at the decoding end. Without adding extra redundancy or reconstructing the frame structure, it achieves cascading error avoidance, dynamic bit rate adaptation, and improved system error correction capabilities.

Description

Technical Field

[0001] This invention relates to the field of communication coding and modulation, and in particular to a coding index modulation method, apparatus and device based on resource separation transmission. Background Technology

[0002] Chaotic communication technology, relying on the boundedness, non-periodicity, and noise-like characteristics of chaotic signals, possesses advantages such as resistance to eavesdropping and multipath interference, attracting significant attention in the field of wireless communication. Based on whether chaotic synchronization is required, chaotic communication schemes can be divided into coherent and incoherent categories. Coherent schemes require chaotic carrier synchronization at the receiver; although theoretically superior, perfect synchronization is difficult to achieve, limiting engineering applications. Incoherent schemes do not require synchronization or channel estimation, exhibiting stronger robustness; among them, DCSK and its improved versions are research hotspots. Traditional DCSK suffers from low spectral efficiency due to a high reference time slot ratio. MDCSK achieves efficient transmission by utilizing combinations of chaotic sequences and orthogonal sequences. Indexed modulation (IM) uses the activation state of physical resources in the communication system as an independent information carrier, significantly improving spectral efficiency. Therefore, the paper "Design and Performance Analysis of a New M-Ary Differential Chaos Shift Keying With Index Modulation" [2020 IEEE Transactions on Wireless Communications: vol. 19, no. 2, pp. 846-858] proposes the IM-MDCSK modulation scheme, and the specific modulation and transmission scheme is described below:

[0003] First, at the transmitting end, the bits are divided into blocks, and the random stream of input information bits is divided into blocks. The sub-block of position, where ( , It is a down-value function. It is the total number of time slots. (The number of active time slots) is used as an index mapping bit to select the active time slot; ( (Representing the number of bits transmitted in each time slot) is the modulation mapping bits, and the bit vector of the sub-block is... .

[0004] Using the mapping rule from natural numbers to k combinations, the mapping bits are... The unique correspondence is the combination of active time slots. Then... According to each The positions are grouped together, and then the constellation point mapping rules are used. Each group of bits is mapped to M-ary constellation points, and the constellation symbols are combined with the chaotic sequence to generate a discrete baseband signal. ( Sampling factor, It is a chaotic sequence. (It is an orthogonal sequence obtained by Hilbert transforming a chaotic sequence). Finally, frame assembly is performed; the frame structure is a combination of a reference time slot and a data time slot. The reference time slot is the original chaotic sequence, while the information part is a combination of all-zero signals (idle time slot) and discrete baseband signals (active time slot).

[0005] At the receiving end, received Then, it is restored to the reference time slot according to the frame structure. Information time slots (of which) (Activation slot). Reference slot. , No. Information slots At this point, we can perform correlation operations between the reference signal and its orthogonal form and all possible information-carrying signals. The decision values ​​for the in-phase and quadrature branches in each time slot are as follows:

[0006]

[0007] In the formula, Hilbert is the Hilbert transform, and T is the transpose operation. From the above two equations, we can obtain... .set up Then, the set can be obtained through the demapping algorithm. In Consistent with the maximum value Index. Then, set It is selected The vector of maximum value (active slot). (Through...) Combining the bits can demodulate the index portion, and then... Mapping to the nearest constellation point yields the demodulated original information bits. Summary of the Invention

[0008] The inventors discovered that, in order to improve the spectral efficiency of incoherent chaotic communication, the traditional IM-MDCSK scheme divides information bits into mapping bits and modulation bits, which are carried in the time slot index and MDCSK modulation symbols respectively. By using time slot index modulation to expand the number of bits transmitted per frame, the spectral efficiency is improved at a fixed code rate. However, this method does not distinguish between information and check resources. The mixed transmission of mapping bits and modulation bits can easily lead to a chain of errors from index bit errors to all modulation bits errors. At the same time, there is no independent check redundancy, and it relies solely on hard decision detection, which causes the bit error rate to deteriorate sharply at low signal-to-noise ratios. Moreover, when adjusting the code rate, the time slot parameters need to be redesigned, which cannot achieve timely dynamic adaptation according to the channel state. Furthermore, the adaptation operation is complex and inflexible, making it difficult to match the dynamic requirements of actual communication.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] The first embodiment of the present invention provides a coding index modulation method based on resource separation transmission, comprising:

[0011] S1. Perform LDPC encoding on the input information bits to generate a complete codeword, and separate the codeword to obtain the bit parity bit and bit information bit;

[0012] S2. Map the bit parity bits to an index combination of active time slots, and generate a discrete baseband signal by modulating the bit information bits using a chaotic sequence.

[0013] S3. The discrete baseband signal and the original chaotic sequence are combined according to the frame structure to form a transmission frame, which is then transmitted through the channel;

[0014] S4. The receiving end performs frame synchronization, separates the reference part and the information part, calculates the energy of each information time slot, and then calculates and generates dynamic parameters and candidate list, and initializes the prior log-likelihood ratio, damping coefficient and number of iterations.

[0015] S5. Input the candidate list, dynamic parameters, and prior log-likelihood ratio into the soft index detector to obtain the posterior log-likelihood ratio of the check bits; input the candidate list and prior log-likelihood ratio into the MDCSK soft demodulator to obtain the posterior log-likelihood ratio of the information bits; concatenate the posterior log-likelihood ratio of the check bits and the posterior log-likelihood ratio of the information bits to obtain the posterior log-likelihood ratio of the entire sequence.

[0016] S6. The posterior log-likelihood ratio of the entire sequence is fed into the LDPC decoder for decoding and verification. If the verification is successful, the decoding result is output.

[0017] S7. If the verification fails, calculate the extrinsic information, multiply the extrinsic information by the damping coefficient to update the prior log-likelihood ratio, separate the verification bits and information bits, and feed them back to step S5 for iteration until the decoding is successful or the maximum number of iterations is reached.

[0018] A second embodiment of the present invention provides a coding index modulation apparatus based on resource separation transmission, comprising:

[0019] The resource separation module is used to perform LDPC encoding on the input information bits to generate a complete codeword, and then separate the codeword to obtain the bit parity bit and the bit information bit.

[0020] The index mapping module is used to map the bit parity bits to an index combination of active time slots and to generate a discrete baseband signal by modulating the bit information bits through chaotic sequence.

[0021] The frame assembly module is used to combine the discrete baseband signal and the original chaotic sequence according to the frame structure to form a transmission frame, which is then transmitted through the channel.

[0022] The parameter adaptive module is used for frame synchronization at the receiver, separating the reference part and the information part, calculating the energy of each information time slot, and then calculating and generating dynamic parameters and candidate lists, and initializing the prior log-likelihood ratio, damping coefficient and number of iterations.

[0023] The splicing unit is used to input the candidate list, dynamic parameters, and prior log-likelihood ratio into the soft index detector to obtain the posterior log-likelihood ratio of the check bits; input the candidate list and prior log-likelihood ratio into the MDCSK soft demodulator to obtain the posterior log-likelihood ratio of the information bits; and splice the posterior log-likelihood ratio of the check bits and the posterior log-likelihood ratio of the information bits to obtain the posterior log-likelihood ratio of the entire sequence.

[0024] The first verification unit is used to send the posterior log-likelihood ratio of the entire sequence into the LDPC decoder for decoding and verification. If the verification is successful, the decoding result is output.

[0025] The second verification unit is used to calculate external information when verification fails, multiply the external information by the damping coefficient to update the prior log-likelihood ratio, separate the verification bits and information bits, and feed them back to the splicing unit for iteration until decoding is successful or the maximum number of iterations is reached.

[0026] The third embodiment of the present invention provides a coding index modulation device based on resource separation transmission, characterized in that it includes a memory and a processor, wherein the memory stores a computer program, and the computer program can be executed by the processor to implement a coding index modulation method based on resource separation transmission as described in any of the above embodiments.

[0027] Based on the coding index modulation method, apparatus, and device based on resource separation transmission provided by this invention, information bits (information bits in the original scheme) and parity bits are transmitted separately at the physical layer (information bits are carried in MDCSK modulation symbols, and parity bits are carried in time slot indexes). Simultaneously, a joint soft-decision iterative mechanism is implemented at the decoding end. Without adding extra redundancy or reconstructing the frame structure, this achieves avoidance of cascading errors, dynamic bit rate adaptation, and improved system error correction capabilities. Attached Figure Description

[0028] Figure 1 This is a flowchart illustrating a coding index modulation method based on resource separation transmission provided in the first embodiment of the present invention;

[0029] Figure 2 This invention provides a transmitter for an indexed multi-level differential chaotic shift keying modulation communication scheme based on resource separation transmission.

[0030] Figure 3 This is a block diagram of the joint optimization algorithm for demodulation and decoding provided by the present invention;

[0031] Figure 4 This is a schematic diagram of a coding index modulation device based on resource separation transmission provided in the second embodiment of the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] To better understand the technical solution of the present invention, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0034] Please see Figure 1 and Figure 2 The first embodiment of the present invention provides a coding index modulation method based on resource separation transmission, comprising:

[0035] S1. Perform LDPC encoding on the input information bits to generate a complete codeword, and separate the codeword to obtain the bit parity bit and bit information bit;

[0036] At the transmitting end, the input information bits are first processed. LDPC encoding is performed, and a complete codeword containing information bits and parity bits is generated by an LDPC encoder. Where b is the original information bit and p is the parity bit. After encoding, the information bit b and the parity bit p in the complete codeword are separated to obtain the parity bit and the information bit.

[0037] The determination of the number of parity bits and the number of information bits is closely related to the frame structure parameters. Specifically, the total number of time slots N in the information portion of the frame structure and the number of time slots that need to be activated are first determined based on the system design requirements. Then calculate the selection from N total time slots. The number of all possible combinations R for each active time slot, and the number of parity bits. The combination number R is obtained through the formula The calculation yielded, where Using the floor function, this formula ensures that the parity bit can be uniquely mapped to one of the R possible time slot combinations. The number of information bits is determined by the number of active time slots. Number of bits transmitted with each active time slot The product is determined, that is, the number of information bits is... ,in, The number of bits transmitted for each active time slot.

[0038] To illustrate with a specific example, suppose we are given an information sequence 1001. After encoding by an LDPC encoder, we obtain the transmission codeword 100110. Separating this codeword yields the information bit b=1001 and the check bit p=10. In this example, we select a total number of time slots N=4 and an active time slot number... =2, modulation order M=4, at this time the number of parity bits =2, number of information bits =4, which meets the system parameter setting requirements. Through this resource separation method, the check bit will be used for subsequent time slot index mapping, while the information bit will be used for chaotic modulation, realizing the separation of check bit and information bit in physical layer transmission resources.

[0039] S2. Map the bit parity bits to an index combination of active time slots, and generate a discrete baseband signal by modulating the bit information bits using a chaotic sequence.

[0040] After codeword separation is completed, the parity bit and information bit need to be processed differently to achieve separate transmission of the parity bit and information bit in the physical layer transmission resources.

[0041] For the processing of the parity bit, a preset mapping rule from natural numbers to k combinations is used to uniquely map the parity bit to a combination selected from N total time slots. Each active time slot is a combination of several active time slots. Specifically, each parity bit value corresponds to a unique combination of active time slots, and the corresponding active time slot can be determined by looking up a table or using a combination mapping algorithm. The location of the activated time slot. Using the previous example as an illustration, when the total number of time slots N=4, the number of activated time slots... When =2, we assume that 00 corresponds to the activation slot. Right now , , , There are no corresponding combinations for the other two time slots.

[0042] For the processing of information bits, a chaotic sequence of length θ is first generated using a chaos generator. Where θ is the sampling factor, and then a Hilbert transform is performed on the chaotic sequence to obtain a sequence orthogonal to it. Next, Bit information bit b according to each Bits are grouped into A group, where M is the modulation order, represents the number of bits that can be carried in each active time slot using M-ary modulation. Each group corresponds to one active time slot. Each group of bits is mapped to an M-ary constellation point, represented as... ,in For the in-phase components of the constellation points, The orthogonal components of the constellation points are represented by the in-phase and orthogonal components. The in-phase and orthogonal components of the constellation points are multiplied by the chaotic sequence and its orthogonal sequence, respectively, and then superimposed to obtain the discrete baseband signal of the nth activation slot. .

[0043] Continuing with the previous example, let's assume a chaotic sequence. =[-0.2455,0.1018,-0.6060,0.7497], after Hilbert transform, yields an orthogonal sequence. =[0.3240,0.1803, -0.3240,-0.1803]. The information bits are divided into two groups and mapped to constellation points. Assume the constellation point mapping rule is... , , , And we assume the chaotic sequence is =[-0.2455,0.1018,-0.6060,0.7497], then =[0.3240,0.1803, -0.3240,-0.1803], and then we can obtain the second time slot. =[-0.5695,-0.0785,-0.2820,0.9300], the 4th time slot =[0.5695,0.0785,0.2820,-0.9300] Through the above processing, the check bit information is carried in the index combination of the active time slot, while the information bit information is carried in the chaotic modulation symbol of each active time slot, realizing the resource separation transmission of the two.

[0044] S3. The discrete baseband signal and the original chaotic sequence are combined according to the frame structure to form a transmission frame, which is then transmitted through the channel;

[0045] The transmission frame consists of two parts: a reference time slot and N data time slots. The reference time slot is located at the beginning of the transmission frame and carries the original chaotic sequence. This is used as a reference signal for incoherent demodulation at the receiver. N data time slots immediately follow the reference time slot, and the active time slot indices are combined according to the parity bit mapping. Each activation time slot is filled with the corresponding discrete baseband signal. And the remaining The idle time slots are filled with all-zero signals. Through this frame structure design, the parity bit information is implicit in the positional distribution of the active and idle time slots, while the information bit information is carried in the modulation symbols within the active time slots.

[0046] Continuing with the previous example, when the total number of time slots N=4 and the number of active time slots... =2. When the parity bit p=10 corresponds to the active time slot combination {1,4}, the assembly process of the transmitted frame is as follows: The reference time slot carries the original chaotic sequence. =[-0.2455,0.1018,-0.6060,0.7497]; The first data time slot is the active time slot, filled with the discrete baseband signal corresponding to the first group of information bits. =[-0.5695,-0.0785,-0.2820,0.9300]; The second data time slot is an idle time slot, filled with all zeros [0,0,0,0]; The third data time slot is an idle time slot, filled with all zeros [0,0,0,0]; The fourth data time slot is an active time slot, filled with the discrete baseband signal corresponding to the second group of information bits. =[0.5695,0.0785,0.2820,-0.9300]. The above reference time slot and four data time slots are arranged sequentially, and then the transmit frame is transmitted through the wireless channel. During transmission, the transmit frame will be affected by channel noise and fading. The signal received by the receiver is the result of the convolution of the transmitted signal and the channel response with added noise.

[0047] S4. The receiving end performs frame synchronization, separates the reference part and the information part, calculates the energy of each information time slot, and then calculates and generates dynamic parameters and candidate list, and initializes the prior log-likelihood ratio, damping coefficient and number of iterations.

[0048] Please combine Figure 3 After receiving the signal transmitted through the channel, the receiving end first performs frame synchronization, separating the received signal into a reference part according to the preset frame structure. and N information parts Reference section The corresponding reference time slot at the transmitting end contains the chaotic sequence after transmission through the channel; the information part The corresponding N data time slots at the transmitting end, among which Each active time slot contains the discrete baseband signal after transmission through the channel. Each idle time slot contains channel noise.

[0049] After completing frame synchronization and signal separation, the reference section is used. The decision value is obtained by performing correlation operations between its orthogonal form and each information part. According to the judgment Calculate the energy of each information time slot The method for calculating energy and the decision quantity of the original system. Similarly, active time slots, carrying the modulation signal, have higher energy values, while idle time slots, containing only noise, have lower energy values. Energy detection can preliminarily distinguish between active and idle time slots, providing observational information for subsequent soft-decision decoding.

[0050] After obtaining the energy of each time slot, the dynamic parameter α is calculated and the candidate list S is generated through the parameter adaptation module, while the prior log-likelihood ratio is initialized. The zero vector, the initial value of the damping coefficient λ, and the iteration counter are set. The calculation process of the dynamic parameter α and the candidate list S is as follows:

[0051] First, calculate the average energy across all time slots. and standard deviation The expression for the mean is: , Where N is the total number of time slots, Let be the energy of the i-th time slot, and be the mean. Reflects the overall energy level, standard deviation It reflects the degree of energy dispersion in each time slot.

[0052] Then calculate the adaptive weights. ,in These are the preset base weight coefficients. The adaptive weight α is used to dynamically adjust the influence of slot energy observations on the posterior probability calculation of candidate combinations. When the value of α is large, it indicates good channel quality and high energy discrimination between active and idle slots. In this case, the observed energy value has a greater impact on the posterior probability calculation, and the system trusts the energy detection results more. When the value of α is small, it indicates poor channel quality and low energy discrimination. In this case, the prior log-likelihood ratio... It has a significant impact on the calculation of posterior probability, and the system relies more on prior information accumulated during the iteration process.

[0053] Finally, a dynamic candidate list S is generated. All time slots are sorted in descending order of energy value, and the slots are ranked according to their order of energy value. Energy per time slot Based on this, calculate the energy threshold. Where δ is the attenuation coefficient, which can be dynamically adjusted according to the channel state. Time slots with energy higher than a threshold T are included in the candidate range, and a candidate list S is generated using these candidate time slots. The candidate list S contains h candidate combinations, each candidate combination... Corresponding to selection from candidate time slots One possible scheme for each activation slot, each candidate combination Corresponding unique check bit sequence By setting an energy threshold to filter candidate time slots, the computational complexity of traversing all combinations is avoided, while retaining potential active time slot combinations, thus achieving a balance between computational efficiency and detection performance.

[0054] Continuing with the previous example, assume the receiving end separates the reference portion. =[-0.24,0.11,-0.7,0.74], and 4 information parts:

[0055] =[-0.1,0.02,-0.03,-0.05] =[-0.56,-0.08,-0.28,0.92] =[-0.01,-0.03,-0.13,0.08] =[0.57,0.07,0.3,-0.87]. Then calculate... , Then I got ,at this time that is Since this assumption is that the channel condition is good, soft decision can be performed directly, and the verification will succeed on the first try; assuming Then calculate according to the above steps. , Calculate adaptive weights Then arrange the energy in descending order. We set the attenuation coefficient to be With the second highest energy value ( Using 2 as a baseline, the energy threshold T is calculated to avoid missing potential activation slots. ,and These could be potential activation slots. Then we can determine the candidate slots. At this point, the candidate set S contains three candidate combinations, namely... , as well as Invalid combinations. Select the valid combinations to form the final candidate set S, i.e. .

[0056] S5. Input the candidate list, dynamic parameters, and prior log-likelihood ratio into the soft index detector to obtain the posterior log-likelihood ratio of the check bits; input the candidate list and prior log-likelihood ratio into the MDCSK soft demodulator to obtain the posterior log-likelihood ratio of the information bits; concatenate the posterior log-likelihood ratio of the check bits and the posterior log-likelihood ratio of the information bits to obtain the posterior log-likelihood ratio of the entire sequence.

[0057] After obtaining the dynamic parameter α and the candidate list S, the candidate list S, the dynamic parameter α, and the prior log-likelihood ratio are combined. The soft index detector performs soft decision-making on the check bits, while simultaneously comparing the candidate list S with the prior log-likelihood ratio. The input to the MDCSK soft demodulator performs soft decision-making on the information bits, and finally the posterior log-likelihood ratio of the parity bits is calculated. And the posterior log-likelihood ratio of information bits The posterior log-likelihood ratio of the entire sequence is obtained by concatenating the sequences. .

[0058] The calculation process of the soft index detector is as follows: First, it utilizes the observed time slot energy information and the prior log-likelihood ratio of the check bit. Calculate each candidate combination in the candidate list S Nonnormalized measure Its expression is In this formula, This represents the total energy of the t time slots in the j-th candidate combination, where t is numerically equal to the number of activated time slots. This item reflects the degree of matching between candidate combinations and energy observations; combinations with higher energy receive larger metric values. This maps 1 to +1 and 0 to -1 in the check bit, which facilitates the subsequent calculation of the log-likelihood ratio. For the check bit sequence The prior log-likelihood ratio of the k-th position; The term is used to quantify whether candidate combinations conform to prior probability tendencies, so that combinations that better conform to prior information receive higher posterior metrics. Then, based on the non-normalized metric... Calculate the posterior probability of each candidate combination. , where h is the total number of candidate combinations in the candidate list S. Finally, the ratio of the sum of the posterior probabilities of each check bit being 1 to the sum of the posterior probabilities of it being 0 in all candidate combinations is calculated, and the posterior log-likelihood ratio of the k-th check bit is calculated. , where the numerator represents the sum of the posterior probabilities of the k-th check bit being 1 in all candidate combinations, and the denominator represents the sum of the posterior probabilities of the k-th check bit being 0 in all candidate combinations.

[0059] The calculation process of the MDCSK soft demodulator is as follows: First, the received signal corresponding to the candidate time slot combination is correlated with the reference signal to calculate the in-phase branch decision value. Orthogonal branch judgment This leads to the decision variable z. Unlike traditional schemes that use minimum Euclidean distance for hard decision mapping of constellation points, this embodiment uses the Max-Log-MAP algorithm for soft decision-making, calculating the decision variables for different candidate combinations S. The intermediate value of each information bit Its expression is .

[0060] In this formula, Let t be the set of constellation points where the t-th bit is 1. Let t be the set of constellation points where the t-th bit is 0. For noise variance, express The t-th information bit in a set of information bits. express The values ​​of the bits other than the t-th bit in the information bits. for The corresponding prior log-likelihood ratio. Taking a quaternary constellation as an example, when calculating the first information bit... intermediate quantity At that time, it is necessary to find the set of all constellation points where the first information bit is 1. and the set of constellation points where the first information bit is 0 Calculate the Euclidean distance between the decision variable z and each constellation point, and combine this with the second information bit. Prior information A weighted average is applied, and the difference between the maximum values ​​in the two sets is taken as the intermediate value for the first information bit. Finally, the posterior log-likelihood ratio of the final information bits is calculated by weighted summation. ,in The posterior probabilities of candidate combinations calculated by the soft index detector are comprehensively considered through this weighted summation operation, resulting in more reliable soft information bits.

[0061] After completing the soft decision for the check bit and information bit, the posterior log-likelihood ratio of the information bit is calculated. The posterior log-likelihood ratio of the parity bit The original codewords are concatenated together in the order they appear to be, and the posterior log-likelihood ratio of the entire sequence is obtained, which is then used as the input to the LDPC decoder.

[0062] Continuing with the previous example, during the first iteration... Calculate if it is 0. Similarly, we can obtain Then calculate Next, the posterior probabilities of the two check bits are calculated. , .

[0063] Based on the calculation of information bits Assuming that the calculation yields... Down Same direction branch , Assuming =0.15, the correspondence between constellation points is At this point, we need to find all the constellation points where the first information bit is 1, that is... and Then calculate the received z( )and distance = = , ( The first index 1 indicates the first information bit, the second index 1 indicates that the value of that information bit is 1, and the third index... (representing constellation points), similarly, z and... distance = =-130.8, because =0, If it is 0, then take and The larger value, -52.0, is used as the first part of the calculation. Then, the calculation is performed for the case where the first information bit is 0, following the same process as above, resulting in... =-191.2, =-270.0, similarly, take the larger of the two, -191.2, as the calculated value for the second part, and then calculate. =-52.0-(-191.2)=139.2, (the probability of the first information bit being 1 is high). The second information bit can be calculated similarly to the above process. =-78.8. Next, we can calculate... The corresponding third information bit and the fourth information bit .and Down (First information bit and second information bit) )and Down The results were exactly the same. Down corresponding and Then we calculate the final = = Finally and Combining them gives the total. .

[0064] S6. The posterior log-likelihood ratio of the entire sequence is fed into the LDPC decoder for decoding and verification. If the verification is successful, the decoding result is output.

[0065] To obtain the posterior log-likelihood ratio of the entire sequence Then, it is fed into an LDPC decoder for decoding. The LDPC decoder uses iterative decoding methods such as the belief propagation algorithm or the minimum sum algorithm, utilizing the parity-check matrix of the LDPC code to determine the posterior log-likelihood ratio of the input. The process involves gradually correcting the soft information of each bit through information transfer between the variable node and the check node, and outputting the decoded posterior log-likelihood value. .

[0066] After decoding, the decoding result is verified using the check equation of the LDPC code. Specifically, it is verified based on the posterior log-likelihood value. Hard decision is performed on each bit to obtain a codeword estimate. This codeword estimate is then substituted into the parity-check matrix of the LDPC code for verification. If all parity-check equations are satisfied (i.e., the product of the parity-check matrix and the codeword estimate is an all-zero vector), the verification is successful, the decoding process ends, and the information bits obtained from the decision are output as the final decoding result. If any parity-check equations are not satisfied, the verification fails, the current decoding result contains errors, and further iterative processing is needed to correct the decoding result.

[0067] Assume the posterior log-likelihood ratio of the input to the LDPC decoder is... The corresponding hard decision result is the codeword estimate 100110. This codeword is substituted into the parity check matrix of the LDPC code for verification. If all parity check equations are satisfied, the decoding is successful, and information bit 1001 is output as the final decoding result. Under good channel conditions, the first decoding can pass the verification without iterative processing, and the system completes reliable transmission with low computational complexity.

[0068] S7. If the verification fails, calculate the extrinsic information, multiply the extrinsic information by the damping coefficient to update the prior log-likelihood ratio, separate the verification bits and information bits, and feed them back to step S5 for iteration until the decoding is successful or the maximum number of iterations is reached.

[0069] If the decoding result output by the LDPC decoder fails to be verified, it indicates that there is an error in the current decoding and further correction is needed through iterative processing. The iterative control unit first determines whether the current iteration count has reached the preset maximum iteration count. If the maximum iteration count has been reached, it directly checks the posterior log-likelihood value output by the LDPC decoder. Make a hard decision and use the decision result as the final output; if the maximum number of iterations has not been reached, proceed to the iterative update process.

[0070] The core of iterative updates is calculating extrinsic information and feeding it back to the soft index detector and the MDCSK soft demodulator. Extrinsic information The calculation formula is: External information This represents additional information mined by the LDPC decoder through the verification constraint relationship. This information was not obtained by the soft index detector and MDCSK soft demodulator in the current iteration. Feeding it back can help improve the soft decision performance in the next round.

[0071] To ensure stable convergence of the iterative process, damping of external information is necessary. Multiplying by the damping coefficient λ yields the weighted extrinsic information λ· Use it as the prior log-likelihood ratio for the next iteration. The damping coefficient λ ranges from 0 to 1 and is used to control the feedback strength of external information to avoid oscillations or divergence during the iteration process.

[0072] The damping coefficient λ employs an adaptive adjustment strategy, dynamically updating based on the convergence state during the iteration process. First, the monitoring indicators are calculated. ,in This represents the number of check equations that were not satisfied in this round. This represents the number of unsatisfied check equations in the previous round. The monitoring index Δ reflects the trend of decoding performance. If Δ < 0, it indicates that the number of unsatisfied check equations is decreasing and the iteration is converging. In this case, the damping coefficient λ = min(0.95, λ + 0.05) should be increased to accelerate the convergence speed. If Δ ≥ 0, it indicates that the iteration is stuck in oscillation and the number of unsatisfied check equations has not decreased. In this case, the damping coefficient λ = max(0.3, λ × 0.6) should be decreased to enhance stability. If Δ > 0 twice consecutively, it indicates that the iteration has deteriorated severely. The damping coefficient λ should be set to 0 to stop the feedback of external information, and the LDPC decoder should be allowed to complete the subsequent decoding on its own.

[0073] Complete the prior log-likelihood ratio After the update, the data is separated into two parts according to the check bit and the information bit, and fed back to the soft index detector and the MDCSK soft demodulator, respectively. The soft index detector uses the updated check bit prior log-likelihood ratio to recalculate the nonnormalized metric and posterior probability of candidate combinations, thereby obtaining a more accurate check bit posterior log-likelihood ratio. The MDCSK soft demodulator uses the updated prior log-likelihood ratio of the information bits to perform a new Max-Log-MAP soft decision, obtaining a more reliable posterior log-likelihood ratio of the information bits. The updated version and The posterior log-likelihood ratio of the concatenated sequence The data is then fed back into the LDPC decoder for decoding and verification. This iterative process is repeated until decoding and verification are successful or the maximum number of iterations is reached.

[0074] Please see Figure 4 The second embodiment of the present invention provides a coding index modulation apparatus based on resource separation transmission, comprising:

[0075] The resource separation module 201 is used to perform LDPC encoding on the input information bits to generate a complete codeword, and then separate the codeword to obtain the bit parity bit and the bit information bit.

[0076] The index mapping module 202 is used to map the bit parity bit to an index combination of active time slots and to generate a discrete baseband signal by modulating the bit information bit through a chaotic sequence.

[0077] The frame assembly module 203 is used to combine the discrete baseband signal and the original chaotic sequence according to the frame structure to form a transmission frame, which is then transmitted through the channel.

[0078] The parameter adaptive module 204 is used for frame synchronization at the receiver, separating the reference part and the information part, calculating the energy of each information time slot, and then calculating and generating dynamic parameters and candidate lists, and initializing the prior log-likelihood ratio, damping coefficient and number of iterations.

[0079] The splicing unit 205 is used to input the candidate list, dynamic parameters, and prior log-likelihood ratio into the soft index detector to obtain the posterior log-likelihood ratio of the check bit; input the candidate list and prior log-likelihood ratio into the MDCSK soft demodulator to obtain the posterior log-likelihood ratio of the information bit; and splice the posterior log-likelihood ratio of the check bit and the posterior log-likelihood ratio of the information bit to obtain the posterior log-likelihood ratio of the entire sequence.

[0080] The first verification unit 206 is used to send the posterior log-likelihood ratio of the entire sequence into the LDPC decoder for decoding and verification. If the verification is successful, the decoding result is output.

[0081] The second verification unit 207 is used to calculate external information when verification fails, multiply the external information by the damping coefficient to update the prior log-likelihood ratio, separate the verification bits and information bits, and feed them back to the splicing unit for iteration until decoding is successful or the maximum number of iterations is reached.

[0082] The third embodiment of the present invention provides a coding index modulation device based on resource separation transmission, including a memory and a processor. The memory stores a computer program, which can be executed by the processor to implement a coding index modulation method based on resource separation transmission as described in any of the above embodiments.

[0083] The fourth embodiment of the present invention provides a computer-readable storage medium storing a computer program that can be executed by a processor of the device in which the computer-readable storage medium is located, so as to implement a method for classifying fallopian tube images as described in any of the above claims.

[0084] Exemplary examples show that the computer program described in the third and fourth embodiments of the present invention can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in implementing a resource-separated transmission-based coded index modulation apparatus. For example, the apparatus described in the second embodiment of the present invention.

[0085] The processor referred to can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the resource-separated transmission-based coded index modulation method, connecting various parts of the method through various interfaces and lines.

[0086] The memory can be used to store the computer program and / or modules. The processor, by running or executing the computer program and / or modules stored in the memory, and by calling the data stored in the memory, implements various functions of a resource-separated transmission-based coded index modulation method. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, text conversion function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, text message data, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0087] If the implemented module is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0088] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0089] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method of encoding index modulation based on resource separation transmission, characterized in that, include: S1. Perform LDPC encoding on the input information bits to generate a complete codeword, and separate the codeword to obtain the bit parity bit and bit information bit; S2. Map the bit parity bits to an index combination of active time slots, and generate a discrete baseband signal by modulating the bit information bits using a chaotic sequence. S3. The discrete baseband signal and the original chaotic sequence are combined according to the frame structure to form a transmission frame, which is then transmitted through the channel; S4. The receiving end performs frame synchronization, separates the reference part and the information part, calculates the energy of each information time slot, and then calculates and generates dynamic parameters and a candidate list, and initializes the prior log-likelihood ratio, damping coefficient and number of iterations. Specifically, the calculation and generation of dynamic parameters and candidate list are as follows: Compute the mean of all the time slot energies and the standard deviation whose expression is , wherein N is the total number of slots, Ei is the energy of the ith slot. Computing adaptive weights wherein, is a base weight coefficient; All time slots are sorted in energy descending order, and the energy of the first time slot is taken as the reference to calculate the threshold wherein, is the decay coefficient; a candidate list S is generated with time slots whose energy is higher than the threshold T, and the candidate list S contains h candidate combinations, each candidate combination corresponds to a unique check bit sequence , is the number of check bit​​ S5. Input the candidate list, dynamic parameters, and prior log-likelihood ratio into the soft index detector to obtain the posterior log-likelihood ratio of the check bits; input the candidate list and prior log-likelihood ratio into the MDCSK soft demodulator to obtain the posterior log-likelihood ratio of the information bits; concatenate the posterior log-likelihood ratio of the check bits and the posterior log-likelihood ratio of the information bits to obtain the posterior log-likelihood ratio of the entire sequence. S6. The posterior log-likelihood ratio of the entire sequence is fed into the LDPC decoder for decoding and verification. If the verification is successful, the decoding result is output. S7. If the verification fails, calculate the extrinsic information, multiply the extrinsic information by the damping coefficient to update the prior log-likelihood ratio, separate the verification bits and information bits, and feed them back to step S5 for iteration until the decoding is successful or the maximum number of iterations is reached.

2. The encoding index modulation method based on resource separation transmission according to claim 1, characterized in that, The method for determining the check bit and the information bit is as follows: The number of combinations R of the active time slots is calculated according to the total number N of time slots of the information part in the frame structure and the number of active time slots , and the number of check bit , wherein is a floor function; the number of information bit is , wherein, is the number of bits transmitted in each active time slot.

3. The coding index modulation method based on resource separation transmission according to claim 1, characterized in that, The step of generating a discrete baseband signal by modulating the bit information bits using a chaotic sequence is specifically as follows: Generating chaotic sequences using a chaos generator Orthogonal sequences are obtained through Hilbert transform. ;Place information bit b according to Bits are grouped into Groups, mapping each group to M-ary constellation points. The discrete baseband signal is obtained from the above: ; in, For the same phase component of the constellation, For orthogonal components, is the sampling factor, and M is the modulation order.

4. The coding index modulation method based on resource separation transmission according to claim 1, characterized in that, The step of combining the discrete baseband signal and the original chaotic sequence according to a frame structure to form a transmission frame specifically involves: The transmission frame includes a reference time slot and N data time slots, wherein the reference time slot carries the original chaotic sequence. The active time slots in the N data time slots carry discrete baseband signals. Idle time slots are filled with all-zero signals.

5. The coding index modulation method based on resource separation transmission according to claim 1, characterized in that, The step of inputting the candidate list, dynamic parameters, and prior log-likelihood ratio into the soft index detector to obtain the posterior log-likelihood ratio of the check bit is specifically as follows: Calculate each candidate combination in the candidate list S Nonnormalized measure Its expression is: ; in, The total energy of the j-th candidate combination in the t time slots; This maps 1 to +1 and 0 to -1 in the check bit. For the check bit sequence The prior log-likelihood ratio of the k-th position; Calculate the posterior probability of each candidate combination. Calculate the posterior log-likelihood ratio of the k-th parity bit: h is the total number of candidate combinations in the candidate list S.

6. The coding index modulation method based on resource separation transmission according to claim 1, characterized in that, The process of inputting the candidate list and prior log-likelihood ratio into the MDCSK soft demodulator to obtain the posterior log-likelihood ratio of the information bits is as follows: The decision value of the in-phase branch is obtained by combining the candidate time slots. Orthogonal branch judgment Thus, the decision variable z is obtained; The median value of each information bit under different candidate combinations is calculated using the Max-Log-MAP algorithm. : ;in, Let t be the set of constellation points where the t-th bit is 1. Let t be the set of constellation points where the t-th bit is 0. For noise variance, express The t-th information bit in a set of information bits. express In a set of information bits, the values ​​of the bits other than the t-th bit are... for The corresponding prior log-likelihood ratio; The final posterior log-likelihood ratio of the information bits is calculated by weighted summation: .

7. The coding index modulation method based on resource separation transmission according to claim 1, characterized in that, The method for adjusting the damping coefficient λ includes: Calculate monitoring indicators ,in This represents the number of check equations that were not satisfied in this round. This represents the number of verification equations that were not satisfied in the previous round. Adaptively adjust the damping coefficient λ: If Δ < 0, it indicates that the iteration has converged, so let λ = min(0.95, λ + 0.05); if Δ ≥ 0, it indicates that the iteration oscillates, so let λ = max(0.3, λ × 0.6); if Δ > 0 twice in a row, it indicates that the iteration has deteriorated, so let λ = 0.

8. A coding index modulation apparatus based on resource separation transmission, characterized in that, include: The resource separation module is used to perform LDPC encoding on the input information bits to generate a complete codeword, and then separate the codeword to obtain the bit parity bit and the bit information bit. The index mapping module is used to map the bit parity bits to an index combination of active time slots and to generate a discrete baseband signal by modulating the bit information bits through chaotic sequence. The frame assembly module is used to combine the discrete baseband signal and the original chaotic sequence according to the frame structure to form a transmission frame, which is then transmitted through the channel. The parameter adaptive module is used for frame synchronization at the receiver, separating the reference part and the information part, calculating the energy of each information time slot, and then calculating and generating dynamic parameters and a candidate list, and initializing the prior log-likelihood ratio, damping coefficient, and number of iterations. Specifically, the calculation and generation of dynamic parameters and the candidate list involves: Calculate the average energy across all time slots. and standard deviation Its expression is , Where N is the total number of time slots, The energy of the i-th time slot; Calculate adaptive weights ,in, Basic weighting coefficients; All time slots are arranged in descending order of energy, starting with the first... Energy per time slot Calculate the threshold based on the baseline ,in, The attenuation coefficient is used; a candidate list S is generated using time slots with energy higher than a threshold T, wherein the candidate list S contains h candidate combinations, and each candidate combination... Corresponding unique check bit sequence , Number of check bits; The splicing unit is used to input the candidate list, dynamic parameters, and prior log-likelihood ratio into the soft index detector to obtain the posterior log-likelihood ratio of the check bits; input the candidate list and prior log-likelihood ratio into the MDCSK soft demodulator to obtain the posterior log-likelihood ratio of the information bits; and splice the posterior log-likelihood ratio of the check bits and the posterior log-likelihood ratio of the information bits to obtain the posterior log-likelihood ratio of the entire sequence. The first verification unit is used to send the posterior log-likelihood ratio of the entire sequence into the LDPC decoder for decoding and verification. If the verification is successful, the decoding result is output. The second verification unit is used to calculate external information when verification fails, multiply the external information by the damping coefficient to update the prior log-likelihood ratio, separate the verification bits and information bits, and feed them back to the splicing unit for iteration until decoding is successful or the maximum number of iterations is reached.

9. A coding index modulation device based on resource separation transmission, characterized in that, The system includes a memory and a processor, wherein the memory stores a computer program that can be executed by the processor to implement a resource separation transmission-based coded index modulation method as described in any one of claims 1 to 7.

Images