Non-full-rank-encoding-mapping-based general-order constellation endogenous probability shaping method and system

Abstract

The application discloses a non-full rank system mapping-based general order constellation endogenous probability forming method and system, relates to the optical transmission technical field, and comprises the following steps: acquiring a binary bit stream to be transmitted, grouping the binary bit stream to be transmitted based on a preset bit number to obtain a binary bit group, performing a system conversion on the binary bit group to obtain a high-system legal state block, mapping each state in the high-system legal state block based on a preset state and constellation mapping relationship to obtain a state corresponding constellation point, and generating a complex vector to be sent based on the state corresponding constellation point; performing a joint maximum likelihood search on the complex vector to be sent based on a pre-constructed legal codebook set to obtain an optimal legal state block, and returning the optimal legal state block to the binary bit stream to be transmitted, so that demodulation is realized.

Description

Technical Field

[0001] This invention relates to the field of optical transmission technology, specifically to a method and system for shaping the endogenous probability of a generalized constellation based on non-full-rank mapping. Background Technology

[0002] As modern communication systems increasingly demand high spectral efficiency and high energy efficiency, traditional physical layer architectures are gradually revealing structural bottlenecks.

[0003] Existing probabilistic shaping schemes heavily rely on external distributed matchers and have not constructed multidimensional block mapping architectures that cross radix (base conversion). Furthermore, they are limited by single-dimensional decision-making methods, which increases the coupling complexity and processing latency of the system and makes it difficult to meet the core requirements of low-latency transmission in the future. Summary of the Invention

[0004] To address the shortcomings mentioned in the background, the present invention aims to provide a method and system for endogenous probability shaping of generalized constellations based on non-full-rank base mapping. By truncating the natural state generated by the underlying base conversion, it achieves extremely simple endogenous probability shaping without relying on an external distributed matcher. At the same time, it combines the legal codebook of the receiving end with maximum likelihood detection to solve the risk of illegal block out-of-bounds judgment caused by traditional single-dimensional independent judgment.

[0005] Firstly, the objective of this invention can be achieved through the following technical solution: a method for shaping the endogenous probability of a generalized constellation based on a non-full-rank mapping, the method comprising the following steps: Obtain the binary bit stream to be transmitted, group the binary bit stream to be transmitted into binary bit groups based on a preset number of bits, and convert the binary bit groups into different bases to obtain a high-base valid state block. Each state of the high-base valid state block is taken from zero to a preset number of bits minus one. Each state in the high-radix legal state block is mapped based on the preset state-constellation mapping relationship to obtain the constellation point corresponding to the state, and a complex vector to be sent is generated based on the constellation point corresponding to the state. Based on a pre-built set of valid codebooks, a joint maximum likelihood search is performed on the complex vector to be transmitted to obtain the optimal valid state block. The optimal valid state block is then returned to the binary bit stream to be transmitted to achieve demodulation.

[0006] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: the high-radix valid state block includes a number of states, a block length, and a preset number of input bits, wherein the preset number of input bits is less than or equal to the product of the number of blocks and the logarithm of the number of states to base 2.

[0007] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: within the high-radix valid state block, a controllable finite block truncation relationship is formed between the high-radix state space and the preset valid input space.

[0008] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: within the finite block truncation relationship, each state of the high-radix legal state block exhibits a probability deviation from uniform distribution.

[0009] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: the preset state and the constellation mapping relationship are as follows: States with high probability of occurrence are preferentially mapped to constellation points with low average energy and strong decision robustness, while states with low probability of occurrence are mapped to constellation points with high average energy.

[0010] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: the high-radix valid state block is as follows: Each state All belong to sets The complex vector to be transmitted is loaded onto the corresponding subcarrier to form an OFDM frequency domain symbol. An IFFT transform is performed on the OFDM frequency domain symbol, a cyclic prefix is ​​added, and a parallel-to-serial transformation is performed to form the OFDM time domain signal to be transmitted. The OFDM symbol can be represented as: .

[0011] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: the sending vector corresponding to the legal input number m in the pre-constructed legal codebook set is: In the formula, m represents the valid input number, and m is taken from the set of valid input numbers; This represents the state value of the valid input number m in the l-th state bit after being expanded in base K; Representation and State The corresponding complex constellation points; L represents the state block length, and the transmission vector consists of L complex constellation points.

[0012] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: the optimal legal state block is as follows: In the formula, y represents the complex received vector obtained by the receiver, c(m) represents the transmitted vector corresponding to the valid input number m, ||·||2 denoted by , where arg min represents the squared Euclidean distance, and arg min represents selecting the number with the smallest distance from the set of valid input numbers. The optimal valid state block is the state block obtained by reverse transformation of the selected number.

[0013] Secondly, in order to achieve the above objectives, this invention discloses a generalized constellation endogenous probability shaping system based on non-full-rank mapping, comprising: The data acquisition module is used to acquire the binary bit stream to be transmitted, group the binary bit stream to be transmitted into binary bit groups based on a preset number of bits, and convert the binary bit groups into a base to obtain a high-base valid state block. Each state of the high-base valid state block is taken from zero to a preset number of bits minus one. The data processing module is used to map each state in the high-radix legal state block based on the preset state-constellation mapping relationship to obtain the constellation point corresponding to the state, and generate a complex vector to be sent based on the constellation point corresponding to the state. The output module is used to perform a joint maximum likelihood search on the complex vector to be transmitted based on a pre-built set of legal codebooks to obtain the optimal legal state block, and return the optimal legal state block to the binary bit stream to be transmitted to achieve demodulation.

[0014] The beneficial effects of this invention are: This invention constructs a multidimensional block mapping framework across cardinality by introducing an adjustable number of input bits. It directly induces a non-uniform probability distribution by utilizing the natural state truncation under high-dimensional space projection. Combined with multidimensional joint maximum likelihood detection based on a set of legal codebooks at the receiver, it achieves endogenous minimal shaping and high-reliability reception without matching, thus realizing flexible transmission that balances high spectral efficiency and high reliability under various channel conditions. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 This is a schematic diagram of the overall structure of the present invention; Figure 3 This invention is different A diagram showing the comparison of system utilization under different parameter combinations; Figure 4 This invention is different A bar chart illustrating the state probabilities under different parameter combinations; Figure 5This is a schematic diagram of the K=9 and K=8 state-constellation mapping relationship of the present invention; Figure 6 This is a schematic diagram of the receiver constellation of the present invention; Figure 7 This is a schematic diagram of the bit error rate under different signal-to-noise ratios (SNRs) of K=9, L=4, N=12, 11, and 10 according to the present invention. Figure 8 This is a schematic diagram of the code rate under K=8, L=4, N=12, 11, 10 and conventional 8QAM modulation. Figure 9 This is a schematic diagram of the system structure of the present invention. Detailed Implementation

[0016] 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.

[0017] Example 1: like Figure 1 As shown, the method for shaping the endogenous probability of a generalized constellation based on a non-full-rank mapping includes the following steps: S101: Obtain the binary bit stream to be transmitted, group the binary bit stream to be transmitted into binary bit groups based on a preset number of bits, convert the binary bit groups into different bases to obtain a high-base valid state block, wherein each state of the high-base valid state block is taken from zero to a preset number of bits minus one. A high-base valid state block includes a number of states, a block length, and a preset number of input bits, wherein the preset number of input bits is less than or equal to the product of the number of blocks and the logarithm of the number of states to base 2. Within the high-base valid state block, a controllable finite block truncation relationship is formed between the high-base state space and the preset valid input space. Within the finite block truncation relationship, each state of the high-base valid state block exhibits a probability deviation from a uniform distribution.

[0018] Specifically, this invention constructs a high-ary state space with K states and L block length at the transmitting end, and sets the number of input bits N to no more than [a certain value]. The system uses adjustable parameters to map a fixed-length binary input N to a K-ary valid state block of length L, thus forming a constrained valid transmission set. Secondly, by utilizing the finite block truncation relationship between the input space and the high-ary state space, the probability of each state appearing in the valid transmission block naturally forms a non-uniform and hierarchically differentiated distribution structure, thereby establishing a unified relationship between utilization and state probability distribution. Furthermore, constellation points are configured according to the frequency of each state's occurrence, providing a basis for geometric mapping and subsequent optimization. Finally, at the receiver, instead of using the traditional method of independent decision-making per subcarrier followed by patching illegal blocks, joint maximum likelihood detection is performed directly on the valid codebook set composed of valid transmission blocks to recover the most probable valid transmission block, thus avoiding out-of-bounds errors and performance loss caused by illegal block patching.

[0019] like Figure 2 As shown, assuming the number of states is K and the subcarrier block length is L, a transmission block can be denoted as: Each state bit in the K-ary representation is taken from 0 to K-1, and its specific value is derived from the input N binary bits. A K-ary state block of length L can represent a complete state space of size K. L On the other hand, each time the input terminal inputs N binary bits, the number of possible values ​​is 2^N. N To ensure that each N-bit binary input can be uniquely mapped to a K-ary state block of length L, and to avoid the input space being larger than the available K-ary state space, leading to unrepresentable data or mapping conflicts, the following must be satisfied: 2 N ≤ K L Taking the logarithm to the base 2 of both sides of the above inequality, we get: N ≤ Llog2 K This invention allows designers to set the input bit length N to satisfy, given K and L. Any positive integer. Therefore, the total number of valid inputs to the system is... The sending end only uses the first M valid high-radix state blocks and does not use the remaining state blocks, while the complete state space size is... Thus, a controllable finite block truncation relationship is formed between the high-radix state space and the legal input space. This invention argues that this truncation relationship is not simply a mapping loss, but rather an endogenous probability hierarchy formation mechanism that can be utilized by the system. Subsequently, a state-constellation mapping relationship is established. Specifically, a constellation set containing K complex constellation points is pre-constructed. Each constellation point c d With a status number One-to-one correspondence. For the sent block Any state s in i Its value ranges from 0 to K-1, therefore it can be determined according to the mapping function. , will state si Mapped to the corresponding complex constellation points Therefore, a K-ary state block of length L can be mapped to a complex transmission vector of length L: .

[0020] Specifically, the present invention will be further illustrated below through embodiments: To further verify the probabilistic hierarchy formation capability of the ternary parameter radix conversion framework proposed in this invention under ideal statistical conditions, this embodiment first starts from the high-radix state space, the legal input space, and the finite block prefix truncation relation, and then examines different... Statistical analysis was performed on the system utilization and state probability distribution under parameter combinations.

[0021] In this embodiment, three representative parameter combinations are selected for analysis. The first category is the non-two-integer power verification group, which is selected as follows: K=9, L=4, Choose N=12,11,10 The second category is the block length variation verification group, which selects: K=9, L=5, Choose N=15 K=9, L=6, Choose N=19 The third category is the power-law base control group, selected as follows: K=8, L=4, Choose N=12,11,10 When K=8, L=4, N=12, we have: This means that the valid input space completely matches the complete high-ary state space, and there is no prefix truncation. This parameter point can be used as a benchmark for uniform distribution to illustrate that when prefix truncation does not exist, the probability of each state should be close to a uniform distribution.

[0022] For each set of parameters mentioned above, this embodiment first calculates the system utilization rate: This measures the utilization of the amount of legitimate input information under a given combination of parameters relative to the theoretical capacity of the complete high-radix state space. The calculation results are as follows: Figure 3 As shown.

[0023] As can be seen from the figure, different Parameter combinations correspond to different utilization levels, where the closer N is to... The higher the system utilization, the better; and when N is actively reduced, the system retains more unused state space, providing greater design margin for subsequent probability hierarchy formation and legal codebook constraint detection.

[0024] Subsequently, the probability of occurrence of each state number was calculated. The state probability histogram for each representative parameter group is shown below. Figure 4 As shown in the figure, when the valid input space perfectly matches the complete state space, the probability of each state is nearly uniform; however, when the valid input space only covers the prefix of the complete state space, the probability of each state begins to deviate from a uniform distribution and forms a hierarchical probability structure.

[0025] As demonstrated by the above ideal statistical verification, the ternary parameter framework proposed in this invention can naturally form a state probability hierarchy solely through base conversion and legal prefix truncation without relying on an external probability matcher; simultaneously, indicators such as utilization rate and state probability distribution can be obtained from... The results were obtained through precise calculations. These results provide parameter basis and theoretical support for subsequent specific constellation point configurations, OFDM transmission statistical verification, and bit error rate performance analysis.

[0026] Furthermore, as demonstrated above, the number of non-two-integer powers of state can be determined by the maximum number of input bits. Natural prefix truncation is formed, inducing a non-uniform state probability distribution. Simultaneously, by actively reducing N, the system can further adjust the size of the legal codebook and the state probability distribution. Based on these results, this invention selects representative values ​​K=9, L=4, N=12, 11, 10 as the non-two-integer power verification group, and K=8, L=4, N=12, 11, 10 as the two-power base control group. Further verification is then conducted in specific constellation mappings and OFDM transceiver links to demonstrate whether the theoretical probability hierarchy can be transformed into actual constellation point access frequencies and differences in bit error rate performance.

[0027] This embodiment selects three representative parameter combinations for OFDM system simulation.

[0028] The first category is the verification group for non-two-integer powers, and the following is selected: The second category is the power-law base control group, selected as follows: L=4 corresponds to each state block containing 4 state bits, which are embedded with 4 valid subcarriers for joint detection; N=12, 11, 10 are used to observe the impact of changes in the number of input bits on bit error rate performance under the same K and L conditions.

[0029] When K=8, L=4, N=12, we have: Therefore, the valid input space perfectly matches the complete state space, and the probability of each state occurring is theoretically nearly uniform, without producing endogenous probability shaping caused by prefix truncation. This parameter point serves as a full-rank unshaped benchmark and is compared with traditional direct modulation and demodulation methods.

[0030] The OFDM system parameters used in this embodiment are shown in Table 1 below: Table 1 OFDM System Parameter Table S102: Map each state in the high-radix legal state block according to the preset state-constellation mapping relationship to obtain the constellation point corresponding to the state, and generate a complex vector to be sent based on the constellation point corresponding to the state. The default state and constellation mapping relationship are as follows: States with high probability of occurrence are preferentially mapped to constellation points with low average energy and strong decision robustness, while states with low probability of occurrence are mapped to constellation points with high average energy.

[0031] The complex vector to be transmitted is loaded onto the corresponding subcarrier to form an OFDM frequency domain symbol. An IFFT transform is performed on the OFDM frequency domain symbol, a cyclic prefix is ​​added, and a parallel-to-serial transformation is performed to form the OFDM time domain signal to be transmitted. Specifically, the transmitting end constellation mapping and OFDM modulation process is as follows: In this embodiment, the transmitter first constructs a corresponding generalized complex constellation for different state numbers K based on the obtained state probability distribution results, and then completes the mapping from legal state blocks to OFDM transmit signals. For the parameter group with K=9, a 9-point complex constellation is constructed; for the parameter group with K=8, an 8-point complex constellation is constructed. Let the state number be... Each state number corresponds to a complex constellation point cd. Constellation point configuration follows a probability-driven geometric mapping principle. For parameter sets with a clear probability hierarchy, states with higher probabilities are preferentially mapped to geometric positions with lower average energy and stronger decision robustness, while states with lower probabilities are mapped to relatively peripheral positions or positions with higher average energy. In this way, the state probability hierarchy induced by the aforementioned number system conversion and legal space truncation can be further transformed into differences in energy distribution and access frequency in the actual constellation geometry, thus providing a foundation for subsequent transmission performance optimization.

[0032] In this embodiment, the two state numbers K=9 and K=8 are adopted as follows: Figure 5 The diagram illustrates the constellation construction method. As can be seen from the diagram, different state numbers have been mapped to corresponding complex constellation points, thus establishing a correspondence between "state number - constellation point".

[0033] The input binary bitstream is first grouped into blocks of N bits each, and then converted into a valid 4-bit high-radix state block using a base conversion: Each state All belong to sets Subsequently, based on the aforementioned state-constellation mapping relationship, each state is mapped to a corresponding constellation point. The aforementioned complex transmit vector is loaded onto the corresponding four subcarriers to form an OFDM frequency domain symbol. This paper selects 5000 OFDM symbols for system simulation verification. Then, an IFFT transform is performed on this frequency domain symbol, a cyclic prefix is ​​added, and a parallel-to-serial transformation is performed to form the OFDM time domain signal to be transmitted.

[0034] S103: Based on the pre-constructed set of legal codebooks, perform a joint maximum likelihood search on the complex vector to be transmitted to obtain the optimal legal state block, and return the optimal legal state block to the binary bit stream to be transmitted to achieve demodulation.

[0035] The signal reception and demodulation process is as follows: At the receiving end, the received signal undergoes serial-to-parallel conversion, cyclic prefix removal, and FFT transformation to obtain the corresponding receiving matrix. Each received OFDM symbol can be represented as: At the receiving end, a valid codebook set identical to that of the sending end is first constructed. Specifically, since the sending end inputs N binary bits each time, the range of valid input number m is... For each valid input number m, the sending end converts m into a K-ary state block of length L according to the same base conversion rules as described above: in Subsequently, based on the pre-defined state-constellation mapping relationship, the state block is mapped to the corresponding complex transmission vector: The set of valid codebooks is composed of the complex transmission vectors corresponding to all valid input numbers m∈{0,1,...,2^N-1}. Therefore, the size of the set of valid codebooks is Its setting is based on the N-bit valid binary input space allowed by the sender. This set only contains complex vectors that the sender may actually transmit, and does not include truncated vectors from the complete K-ary state space. An unused state block.

[0036] At the receiving end, let the received vector obtained after FFT and equalization be: , The receiving end then directly uses the aforementioned set of valid codebooks. A joint maximum likelihood search is performed to recover the optimal valid input number according to the following criteria: Subsequently, the estimated The signal is converted back to the corresponding N-bit binary bits, thus completing the demodulation.

[0037] Figure 6 This is the receiver constellation diagram with SNR=3, L=4, N=11, K=9, and K=8. As can be seen from the diagram, after adding the AWGN channel, noise spreads around the theoretical constellation point; the sample density near different constellation points is not exactly the same, but corresponds to the aforementioned theoretical state probability distribution.

[0038] To further verify the performance of this invention in a real transmission link, this experiment compared the bit error rate of different parameters under different signal-to-noise ratio conditions. First, the parameter groups K=9, L=4, N=12, 11, and 10 were compared, and the results are as follows: Figure 7 As shown in the figure, the system's bit error rate decreases continuously with increasing SNR. Furthermore, within the SNR range of -3 to 3, the smaller N is, the lower the system's bit error rate, indicating that actively reducing the number of input bits can improve reception robustness under low SNR conditions by enhancing the constraints of the legal codebook and the probability hierarchy.

[0039] Secondly, the parameter sets K=8, L=4, N=12, 11, 10 were analyzed, and the results are as follows: Figure 8 As shown in the figure, the bit error rate (BER) pattern is consistent with that when K=9. Meanwhile, K=8 and N=12 represent a complete matching baseline, with a nearly uniform state distribution. Under these parameters, the BER of this scheme was specifically compared with that of the conventional 8QAM direct mapping modulation and demodulation scheme. As can be seen from the figure, under the same signal-to-noise ratio (SNR), the BER performance of this scheme is slightly better than that of the traditional direct mapping demodulation scheme. This indicates that even under the boundary condition of a full-rank state space (i.e., without probabilistic shaping gain), the radix conversion mapping and multi-dimensional valid codebook joint detection mechanism relied upon by this system still possess independent value as a fundamental general-purpose transmission architecture.

[0040] Example 2: To achieve the above objective, such as Figure 9 As shown, based on Embodiment 1, this invention discloses a generalized constellation endogenous probability shaping system based on non-full-rank mapping, comprising: Data acquisition module 11 is used to acquire the binary bit stream to be transmitted, group the binary bit stream to be transmitted into binary bit groups based on a preset number of bits, and convert the binary bit groups into a base to obtain a high base legal state block. Each state of the high base legal state block is taken from zero to a preset number of bits minus one. Data processing module 12 is used to map each state in the high-radix legal state block based on the preset state-constellation mapping relationship to obtain the constellation point corresponding to the state, and generate a complex vector to be sent based on the constellation point corresponding to the state. Output module 13 is used to perform a joint maximum likelihood search on the complex vector to be transmitted based on a pre-built set of legal codebooks to obtain the optimal legal state block, and return the optimal legal state block to the binary bit stream to be transmitted to achieve demodulation.

[0041] Based on the same inventive concept, this invention also provides a computer device, comprising: one or more processors, and a memory for storing one or more computer programs; the programs include program instructions, and the processor executes the program instructions stored in the memory. The processor may 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. It is the computing and control core of the terminal, used to implement one or more instructions, specifically for loading and executing one or more instructions stored in a computer storage medium to implement the above-described method.

[0042] It should be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium storing a computer program, which, when executed by a processor, performs the above-described method. This storage medium can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0043] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0044] The foregoing has shown and described the basic principles, main features, and advantages of this disclosure. Those skilled in the art should understand that this disclosure is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of this disclosure. Various changes and modifications can be made to this disclosure without departing from its spirit and scope, and all such changes and modifications fall within the scope of this disclosure as claimed.

Claims

1. A method for shaping the endogenous probability of a generalized constellation based on a non-full-rank base mapping, characterized in that, The method includes the following steps: Obtain the binary bit stream to be transmitted, group the binary bit stream to be transmitted into binary bit groups based on a preset number of bits, and convert the binary bit groups into different bases to obtain a high-base valid state block. Each state of the high-base valid state block is taken from zero to a preset number of bits minus one. Each state in the high-radix legal state block is mapped based on the preset state-constellation mapping relationship to obtain the constellation point corresponding to the state, and a complex vector to be sent is generated based on the constellation point corresponding to the state. Based on a pre-built set of valid codebooks, a joint maximum likelihood search is performed on the complex vector to be transmitted to obtain the optimal valid state block. The optimal valid state block is then returned to the binary bit stream to be transmitted to achieve demodulation.

2. The method for shaping the endogenous probability of a generalized constellation based on a non-full-rank mapping according to claim 1, characterized in that, The high-radix valid state block includes the number of states, the block length, and a preset number of input bits, wherein the preset number of input bits is less than or equal to the product of the number of blocks and the logarithm of the number of states to base 2; The following are the legal status blocks in high base: Each state All belong to sets .

3. The method for shaping the endogenous probability of a generalized constellation based on a non-full-rank mapping according to claim 2, characterized in that, Within the higher-radix valid state block, a controllable finite block truncation relationship is formed between the higher-radix state space and the preset valid input space.

4. The method for shaping the endogenous probability of a generalized constellation based on a non-full-rank mapping according to claim 3, characterized in that, Within the finite block truncation relation, the states of each state of the high-radix legal state block exhibit a probability deviation from a uniform distribution.

5. The method for shaping the endogenous probability of a generalized constellation based on a non-full-rank mapping according to claim 1, characterized in that, The preset states and their corresponding constellation mappings are as follows: States with high probability of occurrence are preferentially mapped to constellation points with low average energy and strong decision robustness, while states with low probability of occurrence are mapped to constellation points with high average energy.

6. The method for shaping the endogenous probability of a generalized constellation based on a non-full-rank mapping according to claim 1, characterized in that, The complex vector to be transmitted is loaded onto the corresponding subcarrier to form an OFDM frequency domain symbol. An IFFT transform is performed on the OFDM frequency domain symbol, a cyclic prefix is ​​added, and a parallel-to-serial transformation is performed to form the OFDM time domain signal to be transmitted. The OFDM symbol can be represented as: 。 7. The method for shaping the endogenous probability of a generalized constellation based on a non-full-rank mapping according to claim 1, characterized in that, The sending vector corresponding to the legal input number m in the pre-constructed legal codebook set is: In the formula, m represents the valid input number, and m is taken from the set of valid input numbers; This represents the state value of the valid input number m in the l-th state bit after being expanded in base K; Representation and State The corresponding complex constellation points; L represents the state block length, and the transmission vector consists of L complex constellation points.

8. The method for shaping the endogenous probability of a generalized constellation based on a non-full-rank mapping according to claim 1, characterized in that, The optimal legal state block is as follows: In the formula, y represents the complex received vector obtained by the receiver, c(m) represents the transmitted vector corresponding to the valid input number m, ||·|| 2 denoted by , where arg min represents the squared Euclidean distance, and arg min represents selecting the number with the smallest distance from the set of valid input numbers. The optimal valid state block is the state block obtained by reverse transformation of the selected number.

9. A generic constellation endogenous probability shaping system based on non-full-rank base mapping, employing the generic constellation endogenous probability shaping method based on non-full-rank base mapping as described in any one of claims 1 to 8, characterized in that... include: The data acquisition module is used to acquire the binary bit stream to be transmitted, group the binary bit stream to be transmitted into binary bit groups based on a preset number of bits, and convert the binary bit groups into a base to obtain a high-base valid state block. Each state of the high-base valid state block is taken from zero to a preset number of bits minus one. The data processing module is used to map each state in the high-radix legal state block based on the preset state-constellation mapping relationship to obtain the constellation point corresponding to the state, and generate a complex vector to be sent based on the constellation point corresponding to the state. The output module is used to perform a joint maximum likelihood search on the complex vector to be transmitted based on a pre-built set of legal codebooks to obtain the optimal legal state block, and return the optimal legal state block to the binary bit stream to be transmitted to achieve demodulation.

10. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, The memory stores a computer program that can run on the processor. When the processor loads and executes the computer program, it employs the endogenous probability shaping method of the generalized constellation based on non-full rank mapping as described in any one of claims 1 to 8.

Images