Channel coding method and device adaptive to semantic communication, equipment and storage medium

Abstract

The invention belongs to the technical field of radio channel coding, and discloses a channel coding method adaptive to semantic communication, the method comprises a sending end and a receiving end, after the sending end obtains an integer sequence, the integer sequence is converted into a binary sequence arranged according to coding digits according to preset coding parameters, and the binary sequence is sent to the receiving end; and generating a target code through high-low value mapping, and converting the target code into a bit sequence for transmission. And the receiving end performs grouping reduction on the integer according to the mapping bit number, judges the bit values group by group according to the distance between the reduction result and the theoretical value, and reconstructs the original integer sequence. According to the scheme, the binary result of the original numerical value is amplified bit by bit at the sending end, the signal difference degree is remarkably increased, the enough numerical value distinction degree is still maintained in the physical layer distortion environment, and the receiving end reduces the influence of transmission noise on semantic decoding by comparing the integer to be decoded with the relative distance between the theoretical high value and the theoretical low value.

Description

Technical Field

[0001] This application relates to the field of radio channel coding technology, and in particular to channel coding methods, apparatus, devices and storage media adapted to semantic communication. Background Technology

[0002] Existing wireless communication systems (such as 4G / 5G) are designed with error-free transmission as their primary goal. The channel coding in these systems is an error-free correction coding method. However, once the transmission errors exceed the correction range of the error-correcting coding, the existing channel coding will become ineffective. Furthermore, when the transmission errors exceed the correction range of the error-correcting coding, the so-called "cliff effect" occurs, meaning the transmission error rate increases exponentially. Therefore, transmission codewords that fail physical layer verification need to be retransmitted or discarded and will not be sent to the upper layer at the receiving end.

[0003] Error correction in semantic communication (source-channel joint coding) is performed at the upper layer, allowing for bit errors at the physical layer. Data with transmission errors is sent to the upper layer at the receiving end for semantic decoding. Therefore, channel coding for semantic communication should aim to reduce the size of transmission errors, rather than aiming for error-free channel coding. Therefore, this invention proposes an integer channel coding method to reduce transmission errors. Summary of the Invention

[0004] The main objective of this application is to provide a channel coding method, apparatus, device, and storage medium adapted to semantic communication. To achieve the above objective, this application proposes a channel coding method adapted to semantic communication, the method comprising: Obtain the integer sequence to be transmitted; The integers in the integer sequence are encoded sequentially according to the channel coding parameters to obtain the coding result. The channel coding parameters include the minimum integer bit width, the number of modulation bits, the number of mapping bits, and the number of coding bits. The encoding result is converted into binary to obtain the target bits corresponding to each integer. The target bits are arranged sequentially, and the arranged bit sequence is transmitted to the receiving end.

[0005] Optionally, before encoding each integer in the integer sequence sequentially, the method further includes: The number of modulation bits is determined according to the preset signal modulation method; Determine the range of integer values ​​based on the integer sequence to be transmitted; Based on the integer value range, determine the minimum integer bit width value n that satisfies the condition, such that the integer range [0, 2^n-1] completely contains the integer value range; The maximum allowed number of bits for encoding is determined based on the minimum integer bit width value, wherein the minimum integer bit width value is greater than or equal to the maximum allowed number of bits for encoding; The modulation bit number and the encoding bit number are synchronized to the receiving end.

[0006] Optionally, the step of encoding each integer in the integer sequence sequentially according to the channel coding parameters to obtain the encoding result includes: Convert each integer in the integer sequence into a binary representation, wherein the number of bits in the binary representation is the same as the minimum integer bit width value; Based on the number of bits in the encoding, the binary representation of the integer is split into a front part and a back part of the encoding; The bits of the encoding front section are encoded sequentially from the most significant bit to the least significant bit. Bits with a value of 0 are assigned the first extreme value, and bits with a value of 1 are assigned the second extreme value, thus obtaining the encoding front section sequence. The last part is transformed according to a preset rule to obtain the last value; The encoded front sequence and the last value are concatenated in sequence to obtain the encoded processing result.

[0007] Optionally, the first extreme value is generally chosen to be 0, while the second extreme value can vary depending on the different constellation mapping schemes. However, the maximum possible value of the second extreme value cannot exceed [a certain value]. Q is the number of modulation bits.

[0008] Optionally, the number of mapped bits is determined based on the selected second extreme value. The value of the second extreme value is generally... q is the number of mapped bits. According to the second extreme value constraint, the number of mapped bits... Q.

[0009] Optionally, the channel coding method for adapting semantic communication further includes: The set of valid codewords for the encoded output is determined based on the encoding parameters. The set of valid codewords is as follows: Where J = nM-1, q is the number of modulation bits, M is the number of encoding bits, and n is the small integer bit width value.

[0010] The set of valid codewords is encoded into valid codeword description information, and the valid codeword description information and the number of mapped bits are sent to the receiving end, or are negotiated and saved in advance at both the sending and receiving ends.

[0011] Optionally, a channel coding method adapted for semantic communication is applied at a receiving end, the channel coding method adapted for semantic communication comprising: Receive the bit sequence sent by the transmitter and obtain the minimum integer bit width value, the number of modulation bits, the number of mapped bits, the number of coded bits, and the description information of the effective codeword set; The bit sequence is grouped according to the number of mapped bits to obtain the integer to be decoded corresponding to each group of bits; Perform error correction on the integer to be decoded based on the valid codeword description information, and output the corrected integer; The corrected integers are grouped according to the number of coded bits, each group of integers is decoded to obtain a bit value sequence, and the bit value sequences are merged to generate the restored integer sequence.

[0012] Optionally, the step of performing error correction on the integer to be decoded based on the valid codeword description information and outputting a corrected integer includes: Based on the number of encoded bits, the array to be decoded is divided into a pre-encoded part and a post-encoded part; For each integer in the encoding front-end portion, the distance between the integer and the first extreme value and the second extreme value is calculated respectively, and the received integer is corrected to the extreme value that is closer to it, thereby obtaining the corrected encoding front-end portion; For the last integer value, calculate the distance between the integer and each value recorded in the set of valid codewords, and calibrate the integer corresponding to the last value to the nearest neighbor codeword, thereby obtaining the corrected last value; The corrected encoding front part is concatenated with the corrected ending value to obtain the corrected integer.

[0013] Optionally, the step of calculating the numerical distance between the integer corresponding to the last value and the values ​​recorded in the set of valid codewords, and calibrating the integer corresponding to the last value to the nearest neighbor codeword to obtain the corrected last value, includes: Calculate the absolute distance between the received integer corresponding to the last value and each valid codeword in the set of valid codewords; Determine the minimum value from all absolute distances, and take the valid codeword corresponding to the minimum value as the nearest neighbor codeword; The received integer value corresponding to the last value is corrected to the nearest neighbor codeword to obtain the corrected last value.

[0014] Furthermore, to achieve the above objectives, this application also proposes a channel coding apparatus adapted for semantic communication, the channel coding apparatus for semantic communication comprising: The data relay module is used to acquire the integer sequence to be transmitted; The data encoding module is used to encode each integer in the integer sequence sequentially according to the channel coding parameters to obtain the encoding result. The channel coding parameters include the minimum integer bit width, the number of modulation bits, the number of mapping bits, and the number of coding bits. The data modulation module is used to perform binary conversion processing on the encoding processing result to obtain the target bits corresponding to each integer bit; The data transmission module is used to arrange the target bits sequentially and transmit the arranged bit sequence to the receiving end.

[0015] Furthermore, to achieve the above objectives, this application also proposes a channel coding device adapted for semantic communication, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the channel coding method adapted for semantic communication as described above.

[0016] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the channel coding method for adaptive semantic communication as described above.

[0017] The one or more technical solutions proposed in this application have at least the following technical effects: After the transmitting end obtains the integer sequence to be transmitted, it converts the integer sequence into a binary sequence arranged according to the minimum integer bit width value according to the preset encoding parameters. High and low bits are mapped through each bit of the binary sequence to generate the target code and convert it into a bit sequence for transmission. The receiving end performs group reconstruction according to the number of mapped bits, and decides the bit values ​​group by group based on the distance between the reconstruction result and the effective codeword. The decided bit values ​​are then used to construct a sequence to reconstruct the original integer sequence. This solution significantly increases the signal difference by amplifying the binary result of the original value bit by bit at the transmitting end, maintaining sufficient numerical distinguishability even under physical layer distortion. The receiving end reduces the impact of transmission noise on semantic decoding by comparing the relative distance between the integer to be decoded and the effective codeword. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0019] To more clearly illustrate the technical solutions in the embodiments of this application 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.

[0020] Figure 1 A flowchart is provided for Embodiment 1 of the channel coding method adapted to semantic communication in this application; Figure 2A flowchart illustrating the second embodiment of the channel coding method adapted to semantic communication in this application; Figure 3 A schematic diagram of the module structure of a channel coding device adapted for semantic communication in the embodiments of this application; Figure 4 This is a schematic diagram of the device structure of the hardware operating environment involved in the channel coding method adapted to semantic communication in the embodiments of this application.

[0021] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0022] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0023] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0024] In this embodiment, for ease of description, the following description will focus on a channel coding device adapted to semantic communication.

[0025] Existing wireless communication systems (such as 4G / 5G) are designed with error-free transmission as their primary goal. The channel coding in these systems is an error-free correction coding method. However, once the transmission errors exceed the correction range of the error-correcting coding, the existing channel coding will become ineffective. Furthermore, when the transmission errors exceed the correction range of the error-correcting coding, the so-called "cliff effect" occurs, meaning the transmission error rate increases exponentially. Therefore, transmission codewords that fail physical layer verification need to be retransmitted or discarded and will not be sent to the upper layer at the receiving end.

[0026] Error correction in semantic communication (source-channel joint coding) is performed at the upper layer, allowing for bit errors at the physical layer. Data with transmission errors is sent to the upper layer at the receiving end for semantic decoding. Therefore, channel coding for semantic communication should aim to reduce the size of transmission errors, rather than aiming for error-free channel coding. Therefore, this invention proposes an integer channel coding method to reduce transmission errors.

[0027] It should be noted that the execution subject in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or a channel coding device adapted for semantic communication that can achieve the above functions. The following description uses a channel coding device adapted for semantic communication as the execution subject to illustrate this embodiment and the subsequent embodiments.

[0028] Based on this, embodiments of this application provide a channel coding method adapted to semantic communication, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the channel coding method adapted to semantic communication in this application.

[0029] In this embodiment, the channel coding method for adapting semantic communication is applied at the transmitting end and includes steps A10 to A40: Step A10: Obtain the integer sequence to be transmitted.

[0030] It should be noted that the integer sequence to be transmitted refers to the overall data unit received from the application layer or generated internally by the system that needs to be transmitted through the channel. It is usually represented as an ordered integer sequence generated after encoding application layer data streams, such as video, images, and voice. These integers are the source information that will be mapped to specific physical waveforms for transmission in subsequent steps.

[0031] Step A20: Encode each integer in the integer sequence sequentially according to the channel coding parameters to obtain the coding result.

[0032] Understandably, the core parameters of channel coding are the number of modulation bits, the number of mapping bits, the minimum integer bit width, and the number of coding bits. The number of modulation bits is determined by the modulation scheme set during signal transmission. For example, when using 16QAM modulation, each symbol carries 4 bits of information, so the number of modulation bits is 4. Since the choice of modulation scheme is related to specific anti-interference requirements and data transmission bandwidth, in specific implementation, it is necessary to first determine the modulation scheme according to the specific transmission requirements, and then determine the number of modulation bits by combining the inherent mathematical properties of the modulation scheme.

[0033] It is understandable that the number of modulation bits will determine the range of values ​​for the effective codeword and the range of values ​​for the second extreme value; the range of values ​​for the effective codeword and the second extreme value is...

[0034] It should be understood that the number of bits used for encoding is related to the numerical range of the specific integer sequence. For example, for the sequence [8, 14, 11, 3], the numerical range is 3 to 14, and the binary representation of 14 is 1110, which is a 4-bit number. Since the corresponding number of bits used for encoding should be greater than or equal to the number of bits used for the binary representation of the maximum value, the number of bits used for encoding should be less than or equal to 4 bits. In a preferred embodiment, if the number of bits used for the binary representation of the maximum value of the integer sequence is n, and the number of bits used for encoding is M, then: 2 ≤ M ≤ n.

[0035] Understandably, in the example above, the value of n depends on the maximum value of 14 in the sequence, which corresponds to the binary result 1110. Therefore, the value of n is 4, and the range of the number of bits in the encoding is 2 ≤ M ≤ 4. Thus, the possible values ​​for M are 2, 3, and 4.

[0036] It should be noted that, since the sequence [8,14,11,3] in the previous example is converted to binary as [1000,1110,1100,0011], when the value of M is not equal to the value of n, there will be unencoded bits at the end of each binary code. This unencoded bit segment is directly restored according to the original value. For example, when n=4, since the range of M is [2,4], M can take the values ​​2 and 3. When M is 3, only the first M-1 bits of each binary code are encoded. Assuming the second extreme value is chosen as 2^q-1, where q=3. Then, the first M-1 bits of 1 are converted to 2^q-1, and the bits of 0 are converted to 0. After the first M bits are encoded, the values ​​of the remaining bits are directly converted to the original values. Thus, for the sequence 1000, the first two bits are converted to 7 (bit 1) and 0 (bit 0), and the last two bits 00 are converted to the original value 0. So the result of 1000 is (7,0,0). Similarly, the result of 1110 is (7,7,2), the result of 1100 is (7,7,0), and the result of 0011 is (0,0,3). So when M is 3, n is 4, and q is 3, the result of the sequence [8,14,11,3] is [7,0,0,7,7,2,7,7,0,0,0,3]. This encoding result is grouped into three bits.

[0037] In one embodiment, the step of encoding each integer in the integer sequence sequentially according to the channel coding parameters to obtain the encoding result includes: Step A201: Convert each integer in the integer sequence into a binary representation.

[0038] It should be noted that the purpose of this embodiment is to transform a single integer (from an integer sequence) into a core process of "encoding processing result" with stronger anti-interference capabilities. The core idea is to prioritize the protection of the high-order bits of the binary integer by mapping them to both ends of the numerical range to increase the codeword spacing. This ensures the correctness of the high-order bits as much as possible in the event of transmission errors, ultimately reducing the overall error of the integer.

[0039] Step A202: Based on the number of encoded bits, split the binary representation of the integer into a front part and a back part of the encoding.

[0040] It should be noted that this step divides the n-bit binary number obtained in the previous step into two parts with different importance levels based on the number of encoding bits M. M indicates how many new integers a given original integer will ultimately be encoded into. The principle of splitting is to prioritize protecting the higher-order bits.

[0041] Specifically, the first M-1 bits of the binary string are divided into the "encoding front part", which are high-order bits. If there is a deviation, it will cause a large offset to the entire data. Therefore, they are given priority to receive stronger error correction protection. The remaining n-(M-1) bits are divided into the "tail part", which are low-importance bits and have a relatively strong ability to withstand errors.

[0042] Step A203: Assign values ​​to the encoding front part sequentially from the high bit to the low bit to obtain the encoding front part sequence.

[0043] It should be noted that in the specific implementation, the bit value of the first part of the encoding (i.e. the high bit of binary) is checked bit by bit to see if it is 0 or 1, and then the extreme value is assigned: if the bit is 0, it is mapped to the value 0; if the bit is 1, it is mapped to the second extreme value.

[0044] It should be noted that the selection of the second extreme value must take into account the number of modulation bits, and the maximum possible value of the second extreme value cannot exceed [the specified value]. Q is the number of modulation bits.

[0045] For example, when the modulation scheme is 64QAM, the number of modulation bits Q=6. We can then choose q=3, so bit 0 is mapped to 0, and bit 1 is mapped to 2^q-1, which is 7. Therefore, the distance between the mapped results is 7. This allows the receiver to compare the actual received number with the two extreme values ​​during transmission, mapping the more adjacent numbers back to their corresponding 0 or 1, even if the signal is affected by noise. This greatly protects the crucial high-order bits that determine the integer's size. The integer sequence obtained after all the high-order bits have undergone this mapping is called the "encoded front-end sequence".

[0046] Step A204: Convert the last part according to a preset rule to obtain the last value.

[0047] It is understandable that after the first M-1 bits of an n-bit binary number have been assigned (2≤M≤n), the remaining n-(M-1) bits can be regarded as the end part. The end part consists of low-order bits. Since the error of the low-order bits has a small overall impact on the final integer value, there is no need to take extreme mapping protection as with the high-order bits. Therefore, the remaining bits are directly treated as a standard binary number, and its corresponding decimal integer value is calculated as the last bit value.

[0048] It should be understood that after encoding M-1 bits, the remaining bits are encoded into the last integer. The encoding method involves using the remaining bits... Convert from binary to integer , here Its function is to adaptively scale the binary conversion result to adapt to the modulation method.

[0049] Step A205: Concatenate the encoded front sequence with the last value in sequence to obtain the encoding processing result.

[0050] Understandably, this step is responsible for integrating the scattered results from the first two steps into the final output. Specifically, it involves concatenating the "encoded front sequence," containing M-1 strongly mapped protected integers, with the "tail value," representing the remaining bits, in the original bit order (from the most significant bit to the least significant bit). The concatenation results in a new integer sequence of length M, which is the "encoded result" of the original single integer. This entire process ensures that high-priority information is given special protection.

[0051] Once the encoding parameters are determined, the set of valid codewords for the encoded sequence can be determined. The set of valid codewords is: The purpose of the valid codeword set is to restore the data at the receiving end if there is a deviation in the received value.

[0052] In one embodiment, before encoding each integer in the integer sequence sequentially, the method further includes: Step A001: Determine the number of modulation bits according to the preset signal modulation method.

[0053] It should be noted that this step is the initial stage of signal transmission preparation, and its core objective is to determine the amount of basic information that a single modulation symbol can carry. The signal modulation method is pre-configured by the communication system based on factors such as channel conditions, data rate, and power consumption requirements.

[0054] Understandably, Q refers to the order of the modulation mode. This step calculates the core parameter Q (number of modulation bits) by analyzing the physical layer modulation configuration parameter m (constellation order), which is mathematically defined as: Q = floor(log2(m)) or the equivalent relation m = 2^Q. In Quadrature Amplitude Modulation (QAM), m represents the number of discrete points in the constellation diagram (e.g., 64QAM corresponds to m=64), and each symbol can transmit Q=6 bits of information. In Phase Scaled Kinematics (PSK), Q is determined by the number of phase states (e.g., in QPSK, m=4 phases, Q=2).

[0055] Step A002: Determine the range of integer values ​​based on the integer sequence to be transmitted.

[0056] It should be noted that this step aims to provide a basis for subsequently determining the appropriate binary encoding range. The integer sequence typically represents structured data after source coding (such as semantic coding).

[0057] Understandably, the integer value range refers to the range [min, max] formed by the minimum and maximum values ​​(min and max) of all integer elements in the sequence. For example, for an integer sequence [8, 14, 11, 3] to be transmitted, its value range is [3, 14]. This range is determined to ensure that the subsequent binary encoding can represent every possible value in the sequence without distortion.

[0058] It should be understood that accurately determining the value range is a prerequisite for avoiding information loss. If the encoding range is smaller than the actual value range, large values ​​cannot be represented correctly (overflow); if the encoding range is much larger than the actual value range, it will waste bit resources and reduce encoding efficiency. Therefore, this step is fundamental to achieving efficient and lossless encoding.

[0059] Step A003: Based on the integer value range, calculate the minimum number of binary bits n required to cover the integer value range, so that the integer range [0, 2^n-1] completely contains the integer value range.

[0060] It should be noted that the core of this step is to optimize the calculation of the number of bits. The goal is to find a minimum integer n such that all integers (from 0 to 2^n - 1) that can be represented by n binary bits can completely cover the value range [min, max] determined above.

[0061] Understandably, the calculation process follows this logic: First, the range of values ​​to be covered must be at least 1 - 1 / 2 * max - 1 / 2 * min. An n-bit binary number can represent 2^n distinct integers. Therefore, n must satisfy the condition: 2^n >= (max - 1 / 2 * min). The smallest n value that satisfies this condition is the desired value.

[0062] It should be understood that the calculated n is the theoretical minimum number of bits required for efficient binary encoding. Choosing this minimum value of n, rather than a larger number, maximizes data compression and improves channel spectrum utilization.

[0063] Step A004: Determine the number of encoded bits based on the minimum number of binary bits, and synchronize the number of modulation bits and the number of encoded bits to the receiving end.

[0064] Understandably, the selection of M is consistent with the traditional channel coding principle: a larger M can resist greater interference, but it will also bring more redundancy and reduce spectral efficiency. Therefore, the selection of M needs to be determined based on channel conditions.

[0065] It should be understood that synchronizing the modulation bit count, the mapping bit count, the minimum integer bit width, and the coded bit width with the receiving end is crucial for ensuring the establishment of the communication link. The receiving end must know in advance the Q (modulation bit count) and M (coded bit width) used by the transmitting end in order to correctly perform the demodulation, symbol-to-integer inverse mapping, and decoding operations corresponding to those used by the transmitting end. This synchronization can be completed at the beginning of the session establishment through higher-layer signaling (such as the control channel in the communication protocol), or through in-band transmission in the data frame header. This step ensures consistency in the coding and modulation scheme between the transmitting and receiving parties, which is an important guarantee for the normal operation of the entire system.

[0066] Step A30: Perform binary conversion on the encoding result to obtain the target bits corresponding to each integer.

[0067] It should be noted that this step converts the intermediate integer encoding processed in the previous steps into the final binary bit form suitable for transmission in digital communication channels. Its input is the "target encoding", which is a sequence of integers (for example, after the assignment step, it may be a sequence like [0, 7, 0, 0, 0]); the output is the target bits, which is a binary sequence of 0s and 1s.

[0068] Understandably, binary conversion is an element-wise bit decomposition operation. For each integer component in the target encoded sequence, it is converted into a fixed-length binary bit string. This fixed length is determined by the value of the second extremum and is equal to the number of mapped bits, q. For example, if the number of mapped bits q = 3, then the integer value 7 (binary 111) in the target encoded sequence is converted to 111. The integer value 0 (binary 000) is converted to 000. Therefore, for the target encoded [0, 7, 0, ], after this step, a longer binary sequence 000_111_000 will be obtained. This process essentially reorganizes the numerical information, which has undergone semantic and channel gain processing, into a standard binary stream.

[0069] It should be understood that this step is crucial for achieving information format adaptation. It seamlessly converts the processing-oriented numerical representation back into the transmission-oriented binary representation, preparing the data for final physical layer modulation. This conversion is deterministic and reversible, ensuring that the receiving end can accurately recover the original target code through the reverse process.

[0070] Step A40: Arrange the target bits sequentially and transmit the arranged bit sequence to the receiving end.

[0071] It should be noted that this step is the final stage of the digital baseband processing at the transmitting end. Its task is to integrate all the target bits generated in the previous steps into a continuous data stream according to their original logical order and send it out through the physical channel.

[0072] Understandably, sequential arrangement refers to strictly following the order of integers in the target encoded sequence, and the order of the binary bit strings converted from each integer from the most significant bit (MSB) to the least significant bit (LSB) (or a pre-agreed order), concatenating all bits into a long, continuous bit sequence. Continuing the previous example, arranging 000, 111, 000, 000, 000 in sequence yields the final transmitted bit sequence 000111000000000.

[0073] It should be understood that this continuous bit sequence will then be fed into the modulator, mapped to the corresponding analog modulation symbol according to the modulation method (such as QPSK, 16-QAM) determined in step A001, and then radiated out by the antenna after up-conversion, power amplification and other radio frequency processing.

[0074] This embodiment provides a channel coding method adapted for semantic communication, including a transmitting end method and a receiving end method. The transmitting end first determines the number of encoding bits and the number of mapped bits based on the numerical range of the integer sequence to be transmitted and a preset modulation scheme. Then, it performs binary encoding on the integers, assigns a specific value to each bit of the encoded result according to the number of mapped bits, and then converts it into a binary bit stream for transmission. The receiving end performs the reverse process: it divides the received bit stream according to the number of mapped bits, performs inverse conversion to obtain the integer sequence, and decodes it based on the number of encoding bits and a distance comparison decision rule to finally reconstruct the original integer sequence.

[0075] In summary, this technical solution deeply couples the numerical characteristics of semantic information (integer sequences) with physical layer modulation parameters (mapped bit count), enabling adaptive optimization during the coding stage. The transmitter employs a dynamic bit-length determination method based on the sequence value range, avoiding redundancy caused by fixed bit widths and improving spectral efficiency. A unique assignment rule maps each binary bit to an extreme point of the modulation symbol during transmission, enhancing anti-interference capabilities. The distance comparison decision mechanism used at the receiver effectively corrects minor deviations caused by channel noise. Therefore, this method significantly improves the reliability of semantic communication.

[0076] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 The channel coding method for adapting semantic communication, applied at the receiving end, includes steps B10 to B40: Step B10: Receive the bit sequence sent by the transmitter and obtain the modulation bit number, mapping bit number, encoding bit number and effective codeword set description information.

[0077] It should be noted that this step is the initialization phase of the decoding process. The receiving end must first obtain encoding parameters that are completely consistent with those of the sending end from the communication protocol, signaling, or pre-agreed configuration. These parameters are the basis for subsequent correct demapping and error correction, ensuring that both the sending and receiving ends process data in the same context.

[0078] Understandably, this bit sequence may be transmitted via a physical channel and subject to noise interference, therefore it is not the original signal from the transmitter, but rather a version that may contain errors. Obtaining accurate parameter information is a prerequisite for all subsequent correction operations to be effective. The number of modulation bits (e.g., 2 for QPSK, 4 for 16-QAM), the coding parameter M (which determines the division between the coded part and the end part), and the set of valid codewords—these parameters together constitute a complete description of the channel coding scheme.

[0079] Step B20: Group the bit sequence according to the number of mapped bits to obtain the integer to be decoded corresponding to each group of bits.

[0080] It should be noted that this step implements the inverse mapping from the received continuous bit stream to discrete integer values. Essentially, it is the reverse process of the "symbol mapping" at the sending end. The bit sequence is grouped according to the number of mapped bits, and each group of bits is restored to an integer symbol (i.e., an element in the array to be decoded).

[0081] Step B30: Perform error correction on the integer to be decoded according to the valid codeword description information, and output the corrected integer.

[0082] It should be noted that this step utilizes structured redundancy introduced at the sending end for error correction. Its specific operation comprises two main parts: first, dividing the array into a "pre-encoding section" and a "last value" based on the encoding parameters; second, applying different error correction rules (i.e., extreme value correction and nearest neighbor codeword correction) to these two sections respectively.

[0083] Understandably, the purpose of "error correction" is to restore integer values ​​that may be distorted due to channel noise to, as far as possible, the set of valid values ​​allowed by the transmitter. For the front-end of the code, the set of valid values ​​consists of two extreme values: {0, 2^q-1}; for the end value, the set of valid values ​​is... Where J = nM-1. In the event of transmission distortion, numerical restoration can be performed using the nearest neighbor principle.

[0084] Step B40: Group the corrected integers according to the number of encoded bits, decode each group of integers to obtain a bit value sequence, merge the bit value sequences, and generate the restored integer sequence.

[0085] In practice, the entire decoding process at the receiving end includes the following steps: First, each time we take an M-bit non-negative integer sequence, represented as... Starting from the most significant bit (M bits), take one integer at a time. Decode the integer value into bits. .

[0086] The decoding method is: if ,but ;otherwise .

[0087] The last integer digit of a non-negative integer sequence Corresponding decoding From input parameters Confirmed. If ,but:

[0088] if Then first Based on the principle of closest proximity, convert it into the closest valid codeword. However, according to Calculate Then Converted to natural binary Bit binary number ,in .

[0089]

[0090] The principle based on the closest distance is as follows: Belongs to the set of valid codewords ,distance ; .

[0091] If it exists Two valid codewords and If the distances are the same, then one of them is randomly selected as... The final recovered binary array will be synthesized. Restore to integers using natural binary representation .

[0092] In one embodiment, performing error correction on the array to be decoded based on valid codeword description information and outputting a correction integer includes: Step B301: Based on the number of encoded bits, divide the array to be decoded into a front part and a rear part.

[0093] It should be noted that this step is data preprocessing before applying error correction rules. The division is based on the encoding parameter M (i.e., the number of encoding bits), which directly determines how many integers the original data is encoded into. Specifically, the first M-1 integers in the array are categorized as the "front-end encoding," which follows one error correction rule; the last integer is categorized separately as the "end value." Since at the sending end, the integers in the "front-end encoding" are directly mapped from the original bits to two extreme values ​​(such as 0 and the second extreme value), while the integers in the "end value" are intermediate values ​​obtained after scaling, a corresponding division must be performed at the decoding end to apply the correct reverse error correction strategy.

[0094] Step B302: For each integer in the encoded portion, calculate the distance between the integer and the first extreme value and the second extreme value, and correct the received integer to the extreme value that is closer to it, thereby obtaining the corrected encoded portion.

[0095] It should be noted that this step performs "extreme value approximation line-of-sight" error correction on the coding part. Its core principle is to utilize the discreteness of the values ​​in the coding part. In a noiseless channel, the transmitted result can only have one of two extreme values ​​(for example, for q-bit quantization, the first extreme value is 0, and the second extreme value is 2^q - 1). Channel noise will cause the received value to deviate from these two points, and correction is to "pull" it back to the nearest origin.

[0096] Understandably, "distance" usually refers to Euclidean distance or absolute distance, and its effectiveness rests on the assumption that noise power is not large enough to bring the received signal from one extreme closer to another. Since the coding part has only two possible valid values, the nearest neighbor method has good scene adaptability in this specific scenario.

[0097] Step B303: For the integer corresponding to the last value, calculate the numerical distance between the integer and the values ​​recorded in the set of valid codewords, and calibrate the integer corresponding to the last value to the nearest neighbor codeword, thereby obtaining the corrected last value.

[0098] It should be noted that this step performs "discrete codeword set nearest neighbor" error correction on the last value. The principle is that the effective values ​​at the end are not a continuous interval. The correction involves finding the effective codeword in this discrete set that is closest to the received value.

[0099] Step B304: Concatenate the corrected encoded portion with the corrected end portion to obtain the corrected integer.

[0100] In addition, this application introduces the Integer Error Rate (IER) metric to define the data distortion rate during data transmission. Since this encoding method does not perform error-free correction but rather reduces the error distance during transmission, this metric can be used to assist in judging the effectiveness of the encoding method. The specific algorithm is as follows: Assume the integer message sequences are as follows: and ;in That is, x and y are a sequence of non-negative integers. M is a non-negative integer. The IER (Integer Error Rate) is defined as follows:

[0101] Where q is defined as the range of integer values. The number of integers in the message sequence x. Let x be the distance between integer message sequences x and y.

[0102] In this embodiment, the received basic signal unit is demodulated to obtain a basic information sequence. This sequence is then divided into multiple information groups according to a preset modulation information amount, and numerical reconstruction is performed to obtain the numerical sequence to be processed. Next, based on preset coding length information, the original bit value sequence is parsed from the numerical sequence to be processed using a distance comparison decision mechanism. Finally, the bit value sequence is numerically converted to accurately reconstruct the original semantic data sequence from the transmitting end. This method achieves the complete inverse operation of the transmitting end's encoding process.

[0103] In summary, this technical solution effectively overcomes the impact of channel interference on data transmission through specific numerical restoration and distance comparison decision mechanisms. During the decoding process, using known coding parameters and a distance comparison-based intelligent decision criterion, deviations generated during transmission can be automatically corrected. This approach enhances the decoding process's ability to combat channel noise and improves the accuracy of information recovery. Simultaneously, the entire decoding process is closely matched with the encoding strategy at the transmitting end, ensuring the integrity and reliability of semantic information transmission and significantly improving the overall performance of the communication system.

[0104] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the channel coding method for semantic communication adapted to this application. Any simple transformations based on this technical concept are within the protection scope of this application.

[0105] This application also provides a channel coding apparatus adapted for semantic communication; please refer to [reference needed]. Figure 3 The channel coding device for adapting semantic communication includes: Data relay module 10 is used to acquire the integer sequence to be transmitted; Data encoding module 20 is used to encode each integer in the integer sequence sequentially according to channel coding parameters to obtain an encoding result. The channel coding parameters include the number of modulation bits, the number of mapping bits, and the number of coding bits. The data modulation module 30 is used to perform binary conversion processing on the encoding processing result to obtain the target bits corresponding to each integer bit; The data transmission module 40 is used to arrange the target bits sequentially and transmit the arranged bit sequence to the receiving end.

[0106] The channel coding apparatus for semantic communication adapted to this application, employing the channel coding method for semantic communication adapted to the above embodiments, can solve the technical problem in the prior art of how to design a shaping transmission-decision fusion coding method that is compatible with existing digital modulation modes and achieves high fault tolerance for abnormal amplitude offsets. Compared with the prior art, the beneficial effects of the channel coding apparatus for semantic communication adapted to this application are the same as those of the channel coding method for semantic communication adapted to the above embodiments, and other technical features in the channel coding apparatus for semantic communication adapted to the above embodiments are the same as those disclosed in the method of the above embodiments, and will not be modified here.

[0107] This application provides a channel coding device adapted for semantic communication. The channel coding device adapted for semantic communication includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the channel coding method adapted for semantic communication in the above embodiment 1.

[0108] The following is for reference. Figure 4This document illustrates a schematic diagram of a channel coding device suitable for implementing adaptive semantic communication in the embodiments of this application. The channel coding device for adaptive semantic communication in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 4 The channel coding device for adapting semantic communication shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0109] like Figure 4 As shown, the channel coding device adapted for semantic communication may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 1002 or a program loaded from storage device 1003 into random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the channel coding device adapted for semantic communication. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the channel coding device adapted for semantic communication to exchange data wirelessly or via wired communication with other devices. Although channel coding devices adapted for semantic communication with various systems are shown in the figures, it should be understood that it is not required to implement or possess all of the systems shown. More or fewer systems may be implemented alternatively.

[0110] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0111] The channel coding device for semantic communication provided in this application, employing the channel coding method for semantic communication in the above embodiments, can solve the technical problem in the prior art of how to design a shaping transmission-decision fusion coding method that is compatible with existing digital modulation modes and achieves high fault tolerance for abnormal amplitude offsets. Compared with the prior art, the beneficial effects of the channel coding device for semantic communication provided in this application are the same as those of the channel coding method for semantic communication provided in the above embodiments, and other technical features in this channel coding device for semantic communication are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0112] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0113] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0114] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the channel coding method for adaptive semantic communication in the above embodiments.

[0115] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, 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 devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), or any suitable combination thereof.

[0116] The aforementioned computer-readable storage medium may be included in a channel coding device adapted for semantic communication; or it may exist independently and not assembled into a channel coding device adapted for semantic communication.

[0117] The aforementioned computer-readable storage medium carries one or more programs. When these programs are executed by a channel coding device adapted for semantic communication, the channel coding device adapting for semantic communication causes the following: it acquires an integer sequence to be transmitted; it encodes each integer in the integer sequence sequentially according to channel coding parameters to obtain a target code corresponding to each integer, wherein the channel coding parameters include the number of modulation bits and the number of coding bits; it performs binary conversion processing on the target code to obtain target bits corresponding to each integer; it arranges the target bits sequentially and transmits the arranged bit sequence to the receiving end.

[0118] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAn) or a wide area network (WAn)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0119] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0120] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0121] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the channel coding method for adaptive semantic communication described above. This solves the technical problem in the prior art of how to design a shaping transmission-decision fusion coding method that can intelligently coordinate multiple anti-interference coding modes and achieve high fault tolerance for abnormal amplitude offsets. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the channel coding method for adaptive semantic communication provided in the above embodiments, and will not be repeated here.

[0122] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the channel coding method for adaptive semantic communication as described above.

[0123] The computer program product provided in this application solves the technical problem in the prior art of how to design a shaping transmission-decision fusion coding method that can intelligently coordinate multiple anti-interference coding modes and achieve high fault tolerance for abnormal amplitude offsets. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the channel coding method for adapting semantic communication provided in the above embodiments, and will not be repeated here.

[0124] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A channel coding method adapted for semantic communication, applied at the transmitting end, characterized in that, The channel coding method for adapting semantic communication includes: Obtain the integer sequence to be transmitted; The integers in the integer sequence are encoded sequentially according to the channel coding parameters to obtain the coding result. The channel coding parameters include the minimum integer bit width, the number of modulation bits, the number of mapping bits, and the number of coding bits. The encoding result is converted into binary to obtain the target bits corresponding to each integer. The target bits are arranged sequentially, and the arranged bit sequence is transmitted to the receiving end.

2. The channel coding method for adapting semantic communication according to claim 1, characterized in that, Before encoding each integer in the integer sequence sequentially, the process further includes: The number of modulation bits is determined according to the preset signal modulation method; Determine the range of integer values ​​based on the integer sequence to be transmitted; Based on the integer value range, determine the minimum integer bit width value n that satisfies the condition, such that the integer range [0, 2^n-1] completely contains the integer value range; The maximum allowed number of bits for encoding is determined based on the minimum integer bit width value, wherein the minimum integer bit width value is greater than or equal to the maximum allowed number of bits for encoding; The modulation bit number and the encoding bit number are synchronized to the receiving end.

3. The channel coding method for adapting semantic communication according to claim 1, characterized in that, The step of encoding each integer in the integer sequence sequentially according to the channel coding parameters to obtain the encoding result includes: Convert each integer in the integer sequence into a binary representation, wherein the number of bits in the binary representation is the same as the minimum integer bit width value; Based on the number of bits in the encoding, the binary representation of the integer is split into a part before the encoding and a part after the encoding; The bits of the encoding front section are encoded sequentially from the most significant bit to the least significant bit. Bits with a value of 0 are assigned the first extreme value, and bits with a value of 1 are assigned the second extreme value, thus obtaining the encoding sequence. The last part of the encoding is converted according to a preset rule to obtain the last value; The encoded sequence and the last value are concatenated in sequence to obtain the encoded processing result.

4. The channel coding method for adapting semantic communication according to claim 3, characterized in that, The channel coding method for adapting semantic communication further includes: Determine the set of valid codewords for the encoded output based on the encoding parameters; The number of mapped bits is determined based on the selected second extreme value; The set of valid codewords is encoded into valid codeword description information, and the valid codeword description information and the number of mapped bits are sent to the receiving end, or are negotiated and saved in advance at both the sending and receiving ends.

5. A channel coding method adapted for semantic communication, applied at a receiving end, characterized in that, The channel coding method for adapting semantic communication includes: Receive the bit sequence sent by the sender and obtain the minimum integer bit width value, the number of mapped bits, the number of encoded bits, and the description information of the effective codeword set; The bit sequence is grouped according to the number of mapped bits to obtain the integer to be decoded corresponding to each group of bits; Perform error correction on the integer to be decoded based on the valid codeword description information, and output the corrected integer; The corrected integers are grouped according to the number of coded bits, each group of integers is decoded to obtain a bit value sequence, and the bit value sequences are merged to generate the restored integer sequence.

6. The channel coding method for adapting semantic communication according to claim 5, characterized in that, The step of performing error correction on the integer to be decoded based on the valid codeword description information and outputting the corrected integer includes: Based on the number of encoded bits, the array to be decoded is divided into a pre-encoded part and a post-encoded part; For each integer in the encoding front-end portion, the distance between the integer and the first extreme value and the second extreme value is calculated respectively, and the received integer is corrected to the extreme value that is closer, thereby obtaining the corrected encoding front-end portion; For the last integer value, calculate the distance between the integer and each value recorded in the set of valid codewords, and calibrate the integer corresponding to the last value to the nearest neighbor codeword, thereby obtaining the corrected last value; The corrected encoding front part is concatenated with the corrected ending value to obtain the corrected integer.

7. The channel coding method for adapting semantic communication according to claim 6, characterized in that, The step of calculating the numerical distance between the last integer value and the values ​​recorded in the set of valid codewords, and calibrating the last integer value to the nearest neighbor codeword to obtain the corrected last value, includes: Calculate the absolute distance between the last integer value and each valid codeword in the set of valid codewords; Determine the minimum value from all absolute distances, and take the valid codeword corresponding to the minimum value as the nearest neighbor codeword; The value of the last integer is corrected to the nearest neighbor codeword to obtain the corrected last value.

8. A channel coding apparatus adapted for semantic communication, characterized in that, The channel coding device for adapting semantic communication includes: The data relay module is used to acquire the integer sequence to be transmitted; The data encoding module is used to encode each integer in the integer sequence sequentially according to the channel coding parameters to obtain the encoding result. The channel coding parameters include the minimum integer bit width, the number of modulation bits, the number of mapping bits, and the number of coding bits. The data modulation module is used to perform binary conversion processing on the encoding processing result to obtain the target bits corresponding to each integer bit; The data transmission module is used to arrange the target bits sequentially and transmit the arranged bit sequence to the receiving end.

9. A channel coding device adapted for semantic communication, characterized in that, The channel coding device for adaptive semantic communication includes: a memory, a processor, and a channel coding program for adaptive semantic communication stored in the memory and executable on the processor, wherein the channel coding program for adaptive semantic communication is configured to implement the channel coding method for adaptive semantic communication as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores a channel coding program for semantic communication adaptation, which, when executed by a processor, implements the channel coding method for semantic communication adaptation as described in any one of claims 1 to 7.

Images