Four-dimensional multi-component phase modulation transmission and receiving method based on source entropy driving symbol extraction

Abstract

The invention discloses a four-dimensional multi-component phase modulation transmission and receiving method based on information source entropy driving symbol extraction in the technical field of high-dimensional phase modulation, and the method comprises the steps: obtaining the information source entropy of a target information source, calculating the number of information source symbols, and generating a simplified four-dimensional symbol unit; determining a bit index of each symbol to construct a mapping table; selecting the number of constellations, and distributing a corresponding Huffman code for each constellation point; performing interleaving processing on the bit sequence after Huffman coding by combining a bit interleaver, converting the bit sequence into a corresponding decimal numerical value, and performing bit-to-symbol mapping by matching a corresponding bit index through a mapping table to obtain a mapped symbol sequence; and modulating the mapped symbol sequence to a target wave band for transmission. According to the method, the spectrum efficiency and the transmission rate of a wireless communication system are remarkably improved, the symbol mapping efficiency is improved, the anti-interference capability of the system is effectively enhanced, the stability of data transmission is improved, and the error rate is also remarkably reduced.

Description

Technical Field

[0001] This invention relates to a four-dimensional multi-component phase modulation transmission and reception method based on source entropy-driven symbol extraction, belonging to the field of high-dimensional phase modulation technology. Background Technology

[0002] Global internet protocols and internet traffic have seen significant growth over the past two decades, with compound annual growth rates of 30% and 52% respectively, demonstrating a dramatic increase in the number of internet applications, devices, and connections. While traditional modulation methods can meet certain basic requirements, the increasing complexity of application scenarios, especially in environments with high-speed data transmission and high noise interference, necessitates optimizing modulation formats and signal design to improve system performance, which has become a crucial direction for current communication technology development. Phase modulation (PM), as a highly robust modulation method, has been applied in various communication fields. However, with the continuous increase in data rate demands, researchers have begun exploring the combination of non-uniform signal (NS) design and high-dimensional phase modulation to overcome the limitations of simple modulation methods and provide more efficient and reliable solutions in complex and ever-changing environments.

[0003] Phase modulation is a modulation method that uses the instantaneous phase change of a carrier wave to represent information. It can improve the performance of long-distance data transmission systems and offers better fault tolerance and robustness against various transmission impairments. Among various phase modulation formats, Binary Phase Shift Keying (BPSK) and Quadrature Phase Shift Keying (QPSK) data formats have become top candidates for wireless communication applications such as satellite communication, Wi-Fi, and LTE due to their high spectral efficiency and strong noise immunity. However, simple phase modulation formats can only be applied to low-speed communication links and low-power device environments. Therefore, in applications with extremely high wireless communication performance requirements, it may be necessary to introduce more complex modulation formats to optimize BPSK and QPSK signals to adapt to high-speed data transmission needs. When wireless communication systems need to support multiple different types of services or applications within limited spectrum resources, it may be necessary to multiplex multiple data formats into a single signal. In the past few years, researchers have already implemented the multiplexing of different signal formats in highly nonlinear fiber (HNLF). However, this multiplexing method has limitations in spectral efficiency, especially in scenarios requiring the processing of more data streams and higher data rates. Simple multiplexed signals may not be sufficient to meet the demands for high spectral efficiency and high-speed transmission. To address these issues, a higher-dimensional signal design is needed. Furthermore, by adjusting the modulation scheme based on the entropy characteristics of the signal source, signal energy and information density can be allocated more effectively, thereby improving spectral efficiency, reducing redundancy, and enhancing the system's transmission efficiency and anti-interference capabilities.

[0004] To further improve the spectral efficiency and transmission efficiency of the system, combining non-uniform signal design with high-dimensional phase modulation techniques is particularly important. Non-uniform signal design, as an optimization scheme, can more efficiently utilize limited spectral resources and improve the system's anti-interference capability. Based on the different probability distributions of signal symbols, non-uniform signal mapping assigns different energy or amplitude to each symbol to transmit information more effectively. The idea of ​​an optimal non-uniform signal design for Gaussian channels was first proposed in the 1990s and has gradually become a research hotspot in the field of communications in recent years. Huffman coding (HC), as a classic lossless source coding method, minimizes the average code length by assigning different length codes to different symbols and optimizes it according to the probability of symbol occurrence, giving shorter codes to higher-frequency symbols and longer codes to lower-frequency symbols. This method can significantly improve the spectral efficiency and energy utilization efficiency of the system, especially in high-noise and high-interference environments. However, traditional non-uniform signal design techniques are mainly aimed at simple low-dimensional signal structures. In practical applications, as the requirements for transmission rate and signal quality continue to increase, a single coding and modulation scheme may not be able to effectively meet the dual requirements of system capacity and anti-interference performance. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide a four-dimensional multi-component phase modulation transmission and reception method based on source entropy-driven symbol extraction.

[0006] To solve the above-mentioned technical problems, the present invention is implemented using the following technical solution.

[0007] In a first aspect, the present invention discloses a four-dimensional multi-component phase modulation transmission method based on source entropy-driven symbol extraction, comprising: Obtain the source entropy of the target information source, calculate the number of source symbols based on the information entropy, and drive the four-dimensional multi-component phase modulation module to generate simplified four-dimensional symbol units based on the number of source symbols; The bit index of each symbol is determined based on the random sequence and the simplified four-dimensional symbol unit, and a mapping table is constructed based on the bit index of each symbol. Assign a corresponding Huffman code to each constellation point in the simplified four-dimensional symbol unit to obtain the Huffman-coded bit sequence; The Huffman-coded bit sequence is interleaved using a bit interleaver to generate a pseudo-random binary sequence. The pseudo-random binary sequence is then converted into a corresponding decimal value. The corresponding bit index is matched with the mapping table to map bits to symbols, resulting in a mapped symbol sequence. The mapped symbol sequence is modulated onto the target band for transmission.

[0008] Furthermore, the calculation of the number of source symbols based on information entropy includes: Obtain the information entropy (SE) of the information source; Using n=2 SE×D Calculate the number of source symbols, where n represents the number of source symbols and D represents the number of dimensions.

[0009] Furthermore, the four-dimensional multi-component phase modulation module aggregates a set of three QPSK symbols and one BPSK symbol to form a four-dimensional symbol unit, which has a maximum of 128 constellation points. The four-dimensional symbol unit is represented as a vector (s1, s2, s3, s4), where component si represents a modulation symbol, i=1,2,3,4. The first three modulation symbols in the vector (s1, s2, s3, s4) are QPSK modulated, with each symbol carrying 2 bits of QPSK phase information, and the fourth modulation symbol is BPSK modulated, carrying 1 bit of BPSK phase information. Based on the calculated number of source symbols n, n optimal constellation points are selected from 128 constellation points to form a simplified constellation point set, which constitutes the simplified four-dimensional symbol unit.

[0010] Further, determining the bit index of each symbol based on the random sequence and the four-dimensional symbol unit includes: Based on the introduced random sequence seed and combined with the pseudo-random sequence generation algorithm, the bit index of each modulation symbol is determined.

[0011] Furthermore, the formula for calculating the code length of the Huffman code is as follows: ; In the formula, C(x) i () represents the i-th symbol x in the information source x. i The encoding length, P(x) i () represents the i-th symbol x in the information source x. i The probability of occurrence; x = {x1, x2, ..., x I}, and its corresponding probability distribution is P = {P(x1), P(x2), ..., P(x...} I )}, i=1,2,……,I, where I represents the total number of symbols in source x.

[0012] Furthermore, the bit interleaving process is performed on the Huffman-encoded bit sequence using a bit interleaver to generate a pseudo-random binary sequence. This pseudo-random binary sequence is then converted into corresponding decimal values. A mapping table is used to match the corresponding bit indices to perform bit-to-symbol mapping, resulting in a mapped symbol sequence, including: The Huffman-coded bit sequence is interleaved using bit interleaving techniques to generate a pseudo-random binary sequence, represented as follows: ; In the formula, B Huff B represents the bit sequence after Huffman coding. int This represents a pseudo-random binary sequence after interleaving, and Interleave(·) represents the operation function for bit interleaving the input bit sequence; The pseudo-random binary sequence B after interleaving int Convert to decimal value D(B) int ), and through the mapping table, the decimal value D(B) int By mapping the constellation symbols in the simplified four-dimensional symbol unit to the corresponding constellation symbols, the mapped constellation symbol sequence is obtained, represented as: ; In the formula, S k Map(·) represents the mapped constellation symbol, and Map(·) represents the mapping function from bit sequence to constellation symbol.

[0013] Secondly, this invention also discloses a four-dimensional multi-component phase modulation reception method based on source entropy-driven symbol extraction. Receive the target band signal transmitted based on the four-dimensional multi-component phase modulation transmission method based on source entropy-driven symbol extraction as described in the first aspect; The received signal in the target band is demodulated, and then the corresponding symbol is found based on the demodulated bit index to complete the inverse bit-to-symbol mapping process. The original bitstream is recovered through Huffman decoding and deinterleaving.

[0014] Thirdly, the present invention also discloses a computer-readable storage medium for storing one or more programs, characterized in that the one or more programs include instructions that, when executed by a computing device, cause the computing device to perform the method of the first aspect.

[0015] Fourthly, the present invention also discloses a computer device, characterized in that it comprises, One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for performing the method described in the first aspect.

[0016] The beneficial effects achieved by this invention are as follows: This invention significantly improves the spectral efficiency and transmission rate of wireless communication systems by introducing a four-dimensional multi-component phase modulation transmission scheme driven by source entropy for symbol extraction. Furthermore, it enhances the efficiency of symbol mapping through Huffman coding, effectively strengthening the system's anti-interference capability. Simultaneously, the combination of pseudo-random sequences and bit interleaving techniques improves data transmission stability and significantly reduces the bit error rate. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall process of the present invention; Figure 2 It is a four-dimensional symbol unit and a multi-component constellation diagram. Figure 2 (a) is a schematic diagram of a four-dimensional symbol unit structure, where Dimension represents the dimension; (b) is a four-dimensional multi-component constellation map (4D Symbols Map). Figure 3 It is a four-dimensional multi-component constellation diagram. Figure 3 In the diagram, (a) shows the BPSK phase of π / 4 in the four-dimensional constellation diagram, and (b) shows the BPSK phase of 5π / 4 in the four-dimensional constellation diagram. The phase of symbol 1 / 2 / 3 is the phase of the four-dimensional symbol 1 / 2 / 3. Figure 4 This is a visualization analysis of the impact of source entropy on symbol extraction. Source Entropy represents the source entropy. Figure 4 (I) is the dimension reduction graph of the four-dimensional symbols extracted under SE=1.2, (II) is the dimension reduction graph of the four-dimensional symbols extracted under SE=1.4, and (III) is the dimension reduction graph of the four-dimensional symbols extracted under SE=1.6. Figure 5 It is a T-SNE dimensionality reduction graph based on source entropy-driven extraction of four-dimensional symbol units. Figure 5 (a) is the original four-dimensional symbol distribution map, and (b) is the symbol distribution map extracted based on source entropy.

[0018] Figure 6 This is a schematic diagram of the Huffman tree structure constructed under different source entropy values. Figure 6 (a) is the Huffman tree with SE=1.2, (b) is the Huffman tree with SE=1.4, and (c) is the Huffman tree with SE=1.6. Figure 7 This is a flowchart of the bit interleaver's operation. Interleraver is the interleaver, Uniform binary bitinput is the uniform binary bit input, Input is the input sequence, and Output is the output sequence. Figure 8 It is a flowchart of digital signal processing and symbol mapping; Figure 9 This is a system settings screenshot. Figure 9 In the diagram, (a) represents the settings at the sending end, and (b) represents the settings at the receiving end. Figure 10 It is a dimension-reduced projection image extracted from the four-dimensional symbols at the receiving end. Figure 10 (a) is a three-dimensional projection of the BPSK symbol phase in π / 4 dimension, and (b) is a three-dimensional projection of the BPSK symbol phase in 5π / 4 dimension. Figure 11 It is a constellation diagram of a four-dimensional multi-component phase modulation receiver; Figure 12 This is a comparison chart of Huffman coding, with the horizontal axis representing the signal-to-noise ratio (SNR (dB)) and the vertical axis representing the bit error rate (BER). Figure 13 This is a comparison chart of bit error rates under different source entropy settings; Figure 14 This is a comparison chart of the bit error rate performance of four-dimensional multi-component phase modulation and QPSK when SE=1.2. Detailed Implementation

[0019] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0020] Example 1: This example introduces a four-dimensional multi-component phase modulation transmission method based on source entropy-driven symbol extraction, including: Obtain the source entropy of the target information source, calculate the number of source symbols based on the information entropy, and drive the four-dimensional multi-component phase modulation module to generate simplified four-dimensional symbol units based on the number of source symbols; The bit index of each symbol is determined based on the random sequence and the simplified four-dimensional symbol unit, and a mapping table is constructed based on the bit index of each symbol. Assign a corresponding Huffman code to each constellation point in the simplified four-dimensional symbol unit to obtain the Huffman-coded bit sequence; The Huffman-coded bit sequence is interleaved using a bit interleaver to generate a pseudo-random binary sequence. The pseudo-random binary sequence is then converted into a corresponding decimal value. The corresponding bit index is matched with the mapping table to map bits to symbols, resulting in a mapped symbol sequence. The mapped symbol sequence is modulated onto the target band for transmission.

[0021] The calculation process for the number of source symbols is as follows: Obtain the information entropy (SE) of the information source; Using n=2 SE×D Calculate the number of source symbols, where n represents the number of source symbols and D represents the number of dimensions.

[0022] The four-dimensional multi-component phase modulation module aggregates a set of three QPSK symbols and one BPSK symbol to form a four-dimensional symbol unit. A QPSK signal has four different phase states, and a BPSK signal has two different phase states, resulting in 4*4*4*2=128 possibilities. Therefore, the entire set has 128 possible symbol points. The four-dimensional symbol unit is represented as a vector (s1, s2, s3, s4), where component si represents a modulation symbol, i=1,2,3,4. The first three modulation symbols in the vector (s1, s2, s3, s4) are QPSK modulated, with each symbol carrying 2 bits of QPSK phase information. The fourth modulation symbol is BPSK modulated, carrying 1 bit of BPSK phase information.

[0023] Formulan=2 SE×D The function of this function is to determine the number of symbols to be retained in the constellation diagram based on the source entropy value SE, thereby driving the construction of the subsequent four-dimensional symbol unit. The four-dimensional symbol unit itself consists of 3 QPSK symbols and 1 BPSK symbol, with a total of 128 possible symbol points (i.e., constellation points). When the source entropy is low, the information requirement is small, and it is not necessary to use all symbol points. Therefore, after calculating n using the formula, n optimal symbol points are selected from the 128 symbol points in the total set to form a simplified constellation point set, constituting the simplified four-dimensional symbol unit. This ensures that the number of constellation points matches the information content of the source and improves the Euclidean distance between symbols and anti-interference performance. Subsequently, the mapping table and Huffman coding are both built based on these n symbol points, achieving efficient bit-to-symbol mapping. Therefore, n actually plays a core role in "determining the final number of constellation points and the size of the symbol subset".

[0024] The core of selecting an appropriate number of constellation points is to match the constellation diagram's capacity with the source information content based on the source entropy (SE) and system dimension (D), thereby reducing the bit error rate while ensuring spectral efficiency. Specifically, the formula n=2 is first used...SE×D The required number of symbols, n, is calculated. This number reflects the number of different symbol states that the constellation diagram needs to support. When the SE value is low, n is small, and the system selects n symbols with large Euclidean distances and the highest discriminative power from the entire set of constellation points (e.g., a set of 128 constellation points) to form a sparse constellation diagram to enhance anti-interference capabilities. When the SE value is high, n is large, and the system retains more constellation points to improve information carrying capacity and spectral efficiency. The final selection process is implemented based on the entire set of symbol units using a source entropy-driven symbol selection algorithm. That is, according to the probability distribution and importance of symbols based on SE, symbols with large information content and strong representativeness are selected first, so that the constellation diagram can both meet the source entropy requirements and maintain stable transmission performance in noisy environments.

[0025] The determination of the bit index of each symbol based on the random sequence and the four-dimensional symbol unit includes: Based on the introduced random sequence seed and combined with the pseudo-random sequence generation algorithm, the bit index of each modulation symbol is determined.

[0026] The formula for calculating the Huffman code length is as follows: ; In the formula, C(x) i () represents the i-th symbol x in the information source x. i The encoding length, P(x) i () represents the i-th symbol x in the information source x. i The probability of occurrence; x = {x1, x2, ..., x I}, and its corresponding probability distribution is P = {P(x1), P(x2), ..., P(x...} I )}, i=1,2,……,I, where I represents the total number of symbols in source x.

[0027] The bit interleaving process, combined with a bit interleaver, interleaves the Huffman-coded bit sequence to generate a pseudo-random binary sequence. This pseudo-random binary sequence is then converted to its corresponding decimal value. A mapping table is used to match the corresponding bit indices to perform a bit-to-symbol mapping, resulting in a mapped symbol sequence, including: The Huffman-coded bit sequence is interleaved using bit interleaving techniques to generate a pseudo-random binary sequence, represented as follows: ; In the formula, B Huff B represents the bit sequence after Huffman coding. intThis represents a pseudo-random binary sequence after interleaving. `Interleave(·)` is the operation function that performs bit interleaving on the input bit sequence. Its function is to rearrange the original bit sequence (here, the Huffman-coded bit sequence) according to a pre-designed interleaving rule, so that adjacent bits are dispersed as much as possible during transmission, thereby mitigating potential burst errors and improving the system's anti-interference capability and reliability in noisy or interference environments. In other words, `Interleave(·)` stands for "interleaving transformation function," which rearranges the input bit sequence according to a pseudo-random interleaving matrix or permutation rule, outputting a new pseudo-random binary sequence.

[0028] The pseudo-random binary sequence B after interleaving int Convert to decimal value D(B) int ), and through the mapping table, the decimal value D(B) int By mapping the constellation symbols in the simplified four-dimensional symbol unit to the corresponding constellation symbols, the mapped constellation symbol sequence is obtained, represented as: ; In the formula, S k The first function represents the mapped constellation symbol; the second function, Map(·), represents the mapping function from bit sequence to constellation symbol. Its function is to map a decimal value D (Bint) to a specific constellation point (i.e., constellation symbol Sk) in the constellation diagram based on the previously established mapping table. In other words, Map(·) converts the encoded integer index value into specific constellation symbol coordinates by looking up a table or rule, thus achieving a one-to-one correspondence between the bit sequence and the actual modulation symbol.

[0029] Example 2, based on the same inventive concept as Example 1, introduces a four-dimensional multi-component phase modulation receiving method based on source entropy-driven symbol extraction, to receive the target band signal transmitted by the transmission method described in Example 1; The received signal in the target band is demodulated, and then the corresponding symbol is found based on the demodulated bit index to complete the inverse bit-to-symbol mapping process. The original bitstream is recovered through Huffman decoding and deinterleaving.

[0030] Example 3, based on the same inventive concept as Example 1, introduces a four-dimensional multi-component phase modulation transmission method based on source entropy-driven symbol extraction. The process is as follows: Figure 1 As shown, it includes the following modules: system flow module, source entropy driving module, and four-dimensional multi-component phase modulation module.

[0031] The process is as follows: First, the system aggregates three QPSK signals with one BPSK signal to form a four-dimensional symbol unit. To ensure the uniqueness and randomness of the symbols, the system introduces a random sequence seed to generate a pseudo-random sequence and determine the bit index of each symbol. Simultaneously, the system selects the number of constellation points for symbol extraction based on a preset source entropy value, thus determining the required extraction length. Then, the system selects the corresponding constellation points and assigns a corresponding Huffman code to each point. Next, the system uses a bit interleaver to interleave the Huffman-coded bit sequence, generating a pseudo-random binary sequence as input for subsequent modulation. The system then converts the generated binary sequence into corresponding decimal values ​​and maps bits to symbols by matching the corresponding bit sequences. After mapping, the system modulates the mapped symbol sequence through the modulation module and transmits it via a W-band channel. At the receiving end, the system first demodulates the received W-band signal and then finds the corresponding symbol based on the demodulated bit index, thus completing the reverse bit-to-symbol mapping process. Ultimately, through decoding and deinterleaving, the system recovers the original bitstream, achieving complete data transmission.

[0032] The four-dimensional multi-component phase modulation module (4DMC) is formed by aggregating a set of three QPSK symbols with one BPSK symbol to create a four-dimensional symbol unit. This four-dimensional symbol unit can be represented as a vector (s1, s2, s3, s4), where each component si (i = 1, 2, 3, 4) represents a modulation symbol, such as... Figure 2 As shown in (a). Specifically, the first three symbols are derived from QPSK modulation, each carrying 2 bits of QPSK phase information, while the fourth symbol uses BPSK modulation, carrying 1 bit of BPSK phase information. In this way, we transform a traditional two-dimensional constellation diagram into a four-dimensional multi-component constellation diagram, as shown in (a). Figure 2 As shown in (b).

[0033] Formulan=2 SE×DThe function of this method is to determine the number of symbols to be retained in the constellation diagram based on the source entropy value (SE), thereby driving the construction of the subsequent four-dimensional symbol unit. The four-dimensional symbol unit itself consists of 3 QPSK symbols and 1 BPSK symbol, with a total of 128 possible symbol points (i.e., constellation points). When the source entropy is low, the information requirement is small, and it is not necessary to use all symbol points. Therefore, after calculating n using the formula, the n optimal symbol points are selected from the 128 symbol points in the total set to form a simplified constellation point set, constituting the simplified four-dimensional symbol unit. This ensures that the number of constellation points matches the information content of the source and improves the Euclidean distance between symbols and anti-interference performance. Subsequently, the mapping table and Huffman coding are both built based on these n symbol points, achieving efficient bit-to-symbol mapping. Therefore, n actually plays a core role in "determining the final number of constellation points and the size of the symbol subset."

[0034] The design of four-dimensional symbol units not only breaks through the limitations of two-dimensional constellation diagrams, but also significantly improves the dimensionality and information density of signals by encoding and transmitting information in four-dimensional space. Compared to traditional two-dimensional modulation methods, this four-dimensional representation can transmit more information simultaneously because data propagates in a higher-dimensional space, allowing the system to transmit signals at a higher rate while improving its resistance to noise and signal interference. This expansion from two-dimensional to four-dimensional significantly improves the system's spectral efficiency and data carrying capacity. Furthermore, the construction of four-dimensional symbols also means more precise control over the phase and amplitude of signals, thereby optimizing transmission quality. Because the combination space for each symbol is much wider, the system can more effectively utilize limited channel resources, reduce the bit error rate, and improve the reliability and efficiency of data transmission.

[0035] To ensure the uniqueness and randomness of each symbol in the communication system, a random sequence seed technique is employed in the system design. This technique, by introducing a random sequence seed and combining it with a pseudo-random sequence generation algorithm, determines the bit index of each symbol, effectively enhancing the randomness of symbol distribution. This method not only improves the security of the communication system but also enhances its anti-interference capability. The pseudo-random sequence ensures that each symbol's position in the constellation diagram is unique. Figure 3 The constellation diagram shown illustrates the four-dimensional symbol mapping scheme, reflecting the application of random sequence seeds in the four-dimensional symbol unit. It represents the constellation points determined by the pseudo-random bit indices generated by the random sequence seeds. These constellation points are uniformly distributed in four-dimensional space, demonstrating a balanced and efficient symbol utilization strategy. Figure 3 In (a), the blue dots represent constellation points in the fourth dimension with a BPSK phase of π / 4. Figure 3In (b), the red dots represent constellation points with a phase of 5π / 4 in the BPSK in the fourth dimension. The numbers in the figure represent constellation numbers determined by a random seed, and the three dimensions of the two three-dimensional coordinates represent different phase information of the three QPSKs. Furthermore, Figure 3 The four-dimensional multi-component phase-modulated signal corresponding to sequence number 1 is shown, which is presented in QPSK form. This flexible symbol configuration method enables the communication system to maintain high efficiency and stability in complex environments, optimizing the overall communication performance.

[0036] Source entropy-driven module: To ensure the efficiency and accuracy of the communication system, the system first selects an appropriate symbol extraction method based on the source entropy value. The source entropy value (i.e., the average information content of the source) directly affects the amount of information that each symbol can carry, thus determining the structure of the constellation diagram, including the number of symbols and the length of their arrangement. For a discrete multi-symbol memoryless stationary source x = {x1, x2, ..., x...} I}, and its corresponding probability distribution is P = {P(x1), P(x2), ..., P(x...} I The formula for calculating information entropy is as follows: (1); Where: SE(x) is the information entropy of the information source, P(x) i ) is the symbol x i The probability of occurrence is given by n, where n is the number of source symbols. Specifically, the higher the entropy of the source, the more bits each symbol carries, which has a significant impact on the number of points in the constellation diagram, the spacing between points, and the difficulty of decoding the signal.

[0037] For the four-dimensional symbol unit set in the four-dimensional multi-component phase modulation module, its information entropy is: (2); Because there are three QPSK signals and one BPSK signal, the maximum information entropy of this combined signal is 1.75 bits. Therefore, when configuring symbols, it is first necessary to accurately calculate and determine the basic structure of the constellation diagram to ensure that data transmission efficiency can be maximized in the target channel environment.

[0038] After determining the basic symbol configuration, the following formula, where D is the symbol dimension, reflects the relationship between the source entropy SE and the number of source symbols n: n=2 SE×D (3); This formula shows that there is an exponential relationship between the number of symbols in a constellation diagram and the source entropy and symbol dimension. When ensuring optimal data transmission efficiency under given bandwidth and signal-to-noise ratio conditions, this calculation formula provides a foundation for subsequent steps such as symbol mapping, coding, and modulation. Figure 4 As shown. Figure 4 The graph on the left depicts the dynamic relationship between source entropy and the number of source symbols. The curves in the graph show that as the number of symbols increases, the source entropy initially increases rapidly, then levels off, approaching the upper limit of 1.75. The graph uses dots of different colors to mark three specific source entropy values: SE = 1.2, 1.4, and 1.6, corresponding to different symbol number thresholds. The three groups on the right (I, II, III) each include two reduced-dimensional 3D scatter plots for each SE value, corresponding to the symbol extraction situation at different source entropies.

[0039] This invention extracts the core symbol set from the four-dimensional symbol units of the entire set based on information entropy theory. Figure 5 This paper presents a comparison between the original distribution of four-dimensional symbol units in a communication system and the symbol distribution after being filtered by source entropy theory. Figure 5 Figure (a) shows the original distribution of all four-dimensional symbol units. Each point represents a four-dimensional symbol, and a total of 128 symbol points are densely distributed in a specific region. These symbols have not undergone any form of screening or optimization and fully reflect the original state of the information source in high-dimensional space. Figure 5 (b) shows the symbols extracted from the raw data based on source entropy. Each red dot represents a filtered symbol, which is relatively rare in information transmission. Figure 5 The significant reduction in (a) indicates that, after source entropy calculation, those symbols that are more critical to the system's information transmission were retained. Furthermore, the spatial distribution of symbols is more dispersed, and the Euclidean distance is larger, reflecting the balance and optimization of the information-carrying capacity of the symbols selected by source entropy.

[0040] Next, the system selects appropriate constellation points and assigns a Huffman code to each point. Huffman coding is a classic lossless data compression algorithm that dynamically allocates the code length based on the frequency of each symbol in the data stream using variable-length coding. For each symbol xi, its Huffman code length C(xi) can be determined by its probability P(xi), and the specific calculation formula is as follows: (4); Where C(xi) is the encoding length of symbol xi, and P(xi) is the probability of symbol xi appearing. Huffman coding optimizes the compression ratio and transmission efficiency of the data stream by minimizing the expected length of the symbol encoding, so that high-probability symbols use shorter encodings, while low-probability symbols use longer encodings, thereby maximizing encoding efficiency. Figure 6The Huffman trees constructed under different source entropy values ​​are shown, corresponding to SE=1.2, 1.4, and 1.6 respectively. Each Huffman tree uses different nodes to represent hierarchical structures of different depths, optimizing the signal coding length and thus reducing redundancy in overall data transmission.

[0041] Building upon Huffman coding, the system further applies bit interleaving technology to process the encoded bit sequence. Bit interleaving is a technique used to improve data immunity to interference. By rearranging the order of data bits, adjacent bits no longer depend on their proximity in time, increasing the diversity of the data stream and thus improving the system's tolerance to bursty channel interference. The interleaved bit sequence generates a pseudo-random binary sequence, which serves as the input to the modulation process, through the following process: (5); Among them, B Huff It is the bit sequence after Huffman coding, B int It is a bit sequence that has undergone interleaving. Interleaving technology effectively disperses errors during transmission, making the impact of bit errors more even, thereby enhancing the overall reliability of the system. Figure 7 This diagram illustrates the structure and working principle of a bit interleaver, showing the rearrangement of input bits within the interleaver. Each row represents a time step, and the columns indicate different bit positions. The input bit sequence is a uniform binary bit input. After passing through the interleaver, the bit order is rearranged according to Huffman coding to enhance the resistance of the coded data stream to channel interference.

[0042] Interleaved bit sequence B int It will be further converted into a decimal value D(B) int These decimal values ​​are mapped to symbols in the constellation diagram using a mapping table, thus achieving a bit-to-symbol mapping. This process can be represented by the following formula: (6); Among them, D(B) int ) represents the interleaved binary sequence B int The process of converting to decimal values, S k This represents the corresponding constellation symbol. In this process, the system utilizes the lookup function of the mapping table to quickly and efficiently convert the bit sequence into symbols suitable for modulation, thus providing symbol input for subsequent modulation processing. After completing the symbol mapping, the system outputs the resulting symbol sequence S. k The input is processed by the modulation module for phase modulation, resulting in the following output: Figure 8 The system flow is shown below.

[0043] In summary, the entire communication system successfully optimized data transmission efficiency and effectively improved the system's anti-interference capability through multiple steps, including precise symbol selection, Huffman coding, bit interleaving, and modulation. While ensuring a high data transmission rate, it also ensured signal quality and stability under the special conditions of the W-band channel, thus providing strong technical support for high-speed wireless communication.

[0044] Specific example: Verification of IM / DD system transmission scheme based on terahertz W-band.

[0045] (1) Experimental system setup: Figure 9 The experimental setup of the terahertz W-band IM / DD system is shown. At the transmitter (Tx), the symbol extraction process of four-dimensional multi-component phase modulation is driven by source entropy and completed in offline digital signal processing (DSP). The processed signal is then converted into an analog radio frequency (RF) signal and output through an arbitrary waveform generator (AWG). The experiment utilizes the beat frequency effect in nonlinear optics to generate the W-band signal through a photon-assisted beat frequency method. Specifically, when two light waves overlap in space, their electromagnetic field oscillations interfere, thereby generating a new frequency component, i.e., the frequency difference between the two light waves. The experiment utilizes the optical frequency difference between laser 1 and laser 2 to generate the W-band signal. Subsequently, the RF signal output from the arbitrary waveform generator is amplified by an electronic amplifier (EA), then up-converted by a Mach-Zehnder modulator (MZM), and loaded onto a continuous optical carrier in the W-band. The signal is amplified and power-adjusted by an erbium-doped fiber amplifier (EDFA) and a variable optical attenuator (VOA), then converted into an electrical signal by a photodetector (PD), further amplified by a low-noise amplifier (LNA), and transmitted through a horn antenna (CA). In the experimental setup, the distance between the transmitting and receiving antennas is 20 meters. At the receiver (Rx), the signal is received by the horn antenna and amplified by the LNA, then detected and amplified by an envelope detector to recover the original modulated signal. Finally, a mixed-signal oscilloscope (MSO) is used to capture the signal, and signal processing is performed on an offline DSP at the receiver.

[0046] (2) Noise diagram verification: In the performance evaluation of digital communication systems, analyzing the constellation diagram received after passing through an additive white Gaussian noise channel is an extremely effective verification method, which can intuitively show the changes in the signal after being interfered with by noise.

[0047] Four-dimensional multi-component phase modulation utilizes the concept of multi-dimensional modulation, where three QPSK symbols and one BPSK symbol are aggregated to form a four-dimensional symbol unit. Figure 10 The image shows the dimension-reduced projection map obtained by extracting four-dimensional symbols at the receiver after passing through an additive white Gaussian noise channel. Figure 10 (a) and Figure 10 Figure (b) illustrates the changes of the BPSK symbol in dimension four across the three dimensions of the QPSK symbol in its active and deactivated states. This visually demonstrates and validates the potential of this scheme to improve the performance of communication systems, particularly its advantages in signal discrimination and interference immunity.

[0048] Figure 11 This diagram showcases a four-dimensional multi-component phase modulation constellation obtained at the receiver through filtering and demodulation techniques after passing through an additive white Gaussian noise channel. This constellation diagram visually demonstrates the signal performance after transmission via W-band terahertz, effectively validating the innovativeness of this patent's modulation technology. This constellation diagram not only reflects the stability and reliability of the signal after transmission through the high-frequency terahertz band but also demonstrates the advantages of multi-component phase modulation configurations under complex channel conditions. This technology provides important technical support and verification for future applications in a wider range of terahertz fields, such as high-speed wireless communication and long-distance space communication.

[0049] (3) Bit error rate verification: The present invention verifies the system performance from three aspects: whether or not Huffman coding is combined, changing the source entropy setting and comparing with standard QPSK. The bit error rate curve is plotted using Matlab simulation to verify the rationality and effectiveness of the present invention.

[0050] This invention determines the number of symbols to be extracted based on source entropy, and uses this number to obtain the Huffman coding for the constellation configuration. Combined with bit interleaving technology, it generates a binary sequence that meets the source entropy requirements. This method improves coding efficiency while effectively reducing data transmission redundancy. Figure 12 The bit error rate (BER) performance with and without Huffman coding is obtained under the condition that the source entropy (SE) is 1.2. With the increase of the signal-to-noise ratio (SNR), the BER in both cases shows a significant decreasing trend, but overall, the system using Huffman coding exhibits a lower BER, especially in regions with higher SNR, where its performance advantage is particularly significant. This indicates that optimizing the symbol extraction sequence through Huffman coding can not only increase the data compression ratio but also reduce transmission errors while maintaining high data transmission quality, significantly improving the reliability and efficiency of communication systems in complex signal environments.

[0051] When the source entropy is set to a low value, fewer constellation points can be configured. This sparse multi-component constellation, due to its larger Euclidean distance, can resist more noise interference during signal transmission, thereby improving the overall performance of the system. Figure 13The figure shows the bit error rate (BER) curves when the source entropy is set to 1.2, 1.4, and 1.6, respectively. We can clearly see that the BER is relatively low under the low source entropy configuration, and the symbol extraction under low source entropy achieves a more significant improvement in BER as the signal-to-noise ratio (SNR) increases. Therefore, multi-component symbol extraction based on source entropy is particularly suitable for applications requiring high signal stability and data transmission efficiency, providing an effective method for improving the performance of communication systems.

[0052] Figure 14 This paper presents a comparison of the bit error rates (BER) of four-dimensional multi-component phase modulation (4D) after symbol extraction driven by source entropy with that of traditional QPSK modulation. Figure 14 As can be observed, the bit error rate (BER) of both modulation methods decreases with increasing SNR, demonstrating that improved signal quality significantly reduces the BER. Specifically, the four-dimensional multi-component phase modulation system exhibits a lower BER than QPSK across the entire BER range, especially at high BER values, where this difference is more pronounced. This indicates that, under the condition of a source entropy of 1.2, a carefully designed symbol extraction and modulation strategy can effectively improve the noise immunity of the communication system, thereby providing more reliable data transmission.

[0053] Example 4, based on the same inventive concept as other examples, describes a computer-readable storage medium storing one or more programs, the one or more programs including instructions that, when executed by a computing device, cause the computing device to perform the method described in Example 1.

[0054] Example 5, based on the same inventive concept as other examples, describes a computer device including one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for performing the method described in Example 3.

[0055] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0056] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0057] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0058] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0059] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A four-dimensional multi-component phase modulation transmission method based on source entropy-driven symbol extraction, characterized in that, include: Obtain the source entropy of the target information source, calculate the number of source symbols based on the information entropy, and drive the four-dimensional multi-component phase modulation module to generate simplified four-dimensional symbol units based on the number of source symbols; The bit index of each symbol is determined based on the random sequence and the simplified four-dimensional symbol unit, and a mapping table is constructed based on the bit index of each symbol. Assign a corresponding Huffman code to each constellation point in the simplified four-dimensional symbol unit to obtain the Huffman-coded bit sequence; The Huffman-coded bit sequence is interleaved using a bit interleaver to generate a pseudo-random binary sequence. The pseudo-random binary sequence is then converted into a corresponding decimal value. The corresponding bit index is matched with the mapping table to map bits to symbols, resulting in a mapped symbol sequence. The mapped symbol sequence is modulated onto the target band for transmission.

2. The four-dimensional multi-component phase modulation transmission method based on source entropy-driven symbol extraction according to claim 1, characterized in that, The calculation of the number of source symbols based on information entropy includes: Obtain the information entropy (SE) of the information source; Using n=2 SE×D Calculate the number of source symbols, where n represents the number of source symbols and D represents the number of dimensions.

3. The four-dimensional multi-component phase modulation transmission method based on source entropy-driven symbol extraction according to claim 2, characterized in that, The four-dimensional multi-component phase modulation module aggregates a set of three QPSK symbols and one BPSK symbol to form a four-dimensional symbol unit, which has a maximum of 128 constellation points. The four-dimensional symbol unit is represented as a vector (s1, s2, s3, s4), where the component si represents a modulation symbol, i=1,2,3,4. The first three modulation symbols in the vector (s1, s2, s3, s4) are QPSK modulated, with each symbol carrying 2 bits of QPSK phase information, and the fourth modulation symbol is BPSK modulated, carrying 1 bit of BPSK phase information. Based on the calculated number of source symbols n, n optimal constellation points are selected from 128 constellation points to form a simplified constellation point set, which constitutes the simplified four-dimensional symbol unit.

4. The four-dimensional multi-component phase modulation transmission method based on source entropy-driven symbol extraction according to claim 3, characterized in that, The determination of the bit index of each symbol based on the random sequence and the four-dimensional symbol unit includes: Based on the introduced random sequence seed and combined with the pseudo-random sequence generation algorithm, the bit index of each modulation symbol is determined.

5. The four-dimensional multi-component phase modulation transmission method based on source entropy-driven symbol extraction according to claim 1, characterized in that, The formula for calculating the Huffman code length is as follows: ； In the formula, C(x) i () represents the i-th symbol x in the information source x. i The encoding length, P(x) i () represents the i-th symbol x in the information source x. i The probability of occurrence; x = {x1, x2, ..., x I }, and its corresponding probability distribution is P = {P(x1), P(x2), ..., P(x...} I )}, i=1,2,……,I, where I represents the total number of symbols in source x.

6. The four-dimensional multi-component phase modulation transmission method based on source entropy-driven symbol extraction according to claim 5, characterized in that, The bit interleaving process, combined with a bit interleaver, interleaves the Huffman-coded bit sequence to generate a pseudo-random binary sequence. This pseudo-random binary sequence is then converted to its corresponding decimal value. A mapping table is used to match the corresponding bit indices to perform a bit-to-symbol mapping, resulting in a mapped symbol sequence, including: The Huffman-coded bit sequence is interleaved using bit interleaving techniques to generate a pseudo-random binary sequence, represented as follows: ； In the formula, B Huff B represents the bit sequence after Huffman coding. int This represents a pseudo-random binary sequence after interleaving, and Interleave(·) represents the operation function for bit interleaving the input bit sequence; The pseudo-random binary sequence B after interleaving int Convert to decimal value D(B) int ), and through the mapping table, the decimal value D(B) int By mapping the constellation symbols in the simplified four-dimensional symbol unit to the corresponding constellation symbols, the mapped constellation symbol sequence is obtained, represented as: ； In the formula, S k Map(·) represents the mapped constellation symbol, and Map(·) represents the mapping function from bit sequence to constellation symbol.

7. A four-dimensional multi-component phase modulation receiving method based on source entropy-driven symbol extraction, characterized in that, Receive a target band signal transmitted using the four-dimensional multi-component phase modulation transmission method based on source entropy-driven symbol extraction as described in any one of claims 1-6; The received signal in the target band is demodulated, and then the corresponding symbol is found based on the demodulated bit index to complete the inverse bit-to-symbol mapping process. The original bitstream is recovered through Huffman decoding and deinterleaving.

8. A computer-readable storage medium for storing one or more programs, characterized in that, The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform any of the methods of claims 1 to 6.

9. A computer device, characterized in that, include, One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing the method of any of claims 1 to 6.

Images