Coding and decoding method for semantic channel and electronic device

Abstract

The present application provides a coding and decoding method for a semantic channel and an electronic device. The method comprises: after obtaining a semantic codebook and a synonymous mapping relationship that are pre-constructed, determining a grammar information sequence to be sent, determining a grammar code word corresponding to the grammar information sequence in the semantic codebook on the basis of the synonymous mapping relationship, and modulating the grammar code word to obtain a signal vector; and performing signal recovery processing on the signal vector on the basis of the semantic codebook and a preset decoding criterion to obtain a target synonymous set serial number for representing semantic information, on the basis of the synonymous mapping relationship, determining a target grammar information sequence set corresponding to the target synonymous set serial number, and selecting a grammar information sequence from among the target grammar information sequence set as an output result.

Description

Semantic channel encoding and decoding method and electronic device Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a semantic channel encoding and decoding method and electronic equipment. Background Art

[0002] The performance of traditional communication technologies, guided by classical information theory, has already reached its theoretical limits, making further significant breakthroughs difficult. To meet the high efficiency and reliability requirements of future wireless communication scenarios, new technologies are needed to achieve breakthroughs. Currently, semantic communication technology offers the potential to break through the Shannon limit, and semantic channel coding is a key technology for increasing transmission rates. However, the introduction of the concept of synonymy in semantic communication elevates the research focus from codewords to synonymous sets, rendering some analytical methods used in traditional channel coding schemes inapplicable. Furthermore, there is currently no theoretical framework for transitioning channel coding and decoding schemes from syntactic communication scenarios to semantic communication scenarios. Therefore, it is necessary to provide theoretical guidance and design appropriate channel coding and decoding schemes for semantic communication.

[0003] Summary of the Invention

[0004] In view of this, the purpose of this application is to propose a coding and decoding method and electronic device for a semantic channel, so as to expand the coding and decoding method of a syntactic channel into a coding and decoding method of a semantic channel, thereby reducing the probability of errors in the received semantic information in a semantic scenario and reducing the error semantic information rate.

[0005] The encoding method of the semantic channel provided in the present application includes: obtaining a pre-constructed semantic codebook and a synonymous mapping relationship; determining a grammatical information sequence to be sent, determining a grammatical codeword corresponding to the grammatical information sequence in the semantic codebook according to the synonymous mapping relationship, and modulating the grammatical codeword to obtain a signal vector.

[0006] The decoding method of the semantic channel provided in the present application includes: obtaining a pre-constructed semantic codebook and synonymous mapping relationship; performing signal recovery processing on a signal vector according to the semantic codebook and a preset decoding criterion to obtain a target synonymous set sequence number for representing semantic information; determining a target grammatical information sequence set corresponding to the target synonymous set sequence number according to the synonymous mapping relationship; and selecting a grammatical information sequence from the target grammatical information sequence set as an output result.

[0007] The encoding device of the semantic channel provided by this application includes:

[0008] Relationship parameter acquisition module, used to obtain pre-built semantic codebook and synonym mapping relationship;

[0009] A grammatical information sequence determination module, configured to determine a grammatical information sequence to be sent;

[0010] A grammatical codeword determination module is used to determine a grammatical codeword corresponding to the grammatical information sequence in the semantic codebook according to the synonymous mapping relationship;

[0011] The modulation module is used to modulate the syntax codeword to obtain a signal vector.

[0012] The decoding device of the semantic channel provided in this application includes:

[0013] Relationship parameter acquisition module, used to obtain pre-built semantic codebook and synonym mapping relationship;

[0014] A recovery module, configured to perform signal recovery processing on the signal vector according to the semantic codebook and a preset decoding criterion to obtain a target synonym set sequence number for representing semantic information;

[0015] a grammatical information sequence set determination module, configured to determine a target grammatical information sequence set corresponding to a target synonymous set sequence number according to the synonymous mapping relationship; and

[0016] The output module is used to select a grammatical information sequence from the target grammatical information sequence set as an output result.

[0017] Based on the same inventive concept, the present disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor implements the above-mentioned method when executing the computer program.

[0018] In addition, the present disclosure also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the above method.

[0019] The present disclosure also provides a computer program product, comprising: computer program instructions, which, when executed on a computer, enable the computer to execute the above method.

[0020] From the above description, it can be seen that the encoding and decoding method and electronic device of the semantic channel provided by the present application can determine the grammatical information sequence to be sent after obtaining the pre-constructed semantic codebook and synonymous mapping relationship, determine the grammatical codeword corresponding to the grammatical information sequence in the semantic codebook according to the synonymous mapping relationship, modulate the grammatical codeword to obtain a signal vector; perform signal recovery processing on the signal vector according to the semantic codebook and the preset decoding criteria to obtain the target synonymous set number for representing the semantic information, and determine the target grammatical information sequence set corresponding to the target synonymous set number according to the synonymous mapping relationship, and select a grammatical information sequence from the target grammatical information sequence set as the output result. A bridge is built between the grammatical channel encoding and decoding scheme and the semantic channel encoding and decoding scheme through the synonymous mapping relationship, so as to expand the encoding and decoding method of the grammatical channel to the encoding and decoding method of the semantic channel, thereby reducing the probability of errors in the received semantic information in the semantic scenario and reducing the error semantic information rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in this application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are merely embodiments of this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0022] FIG1 is a schematic diagram of a channel communication system in a semantic communication scenario according to an embodiment of the present application;

[0023] FIG2a is a flow chart of a method for encoding a semantic channel according to an embodiment of the present application;

[0024] FIG2 b is a flowchart of a decoding method for a semantic channel according to an embodiment of the present application;

[0025] FIG3a is a flowchart of constructing a semantic codebook and synonymous mapping relationships according to an embodiment of the present application;

[0026] FIG3 b is a schematic diagram of a synonymous mapping relationship according to an embodiment of the present application;

[0027] FIG4 is a schematic diagram of the internal structure of a channel encoder in a semantic communication scenario according to an embodiment of the present application;

[0028] FIG5 is a schematic diagram of the internal structure of a channel decoder in a semantic communication scenario according to an embodiment of the present application;

[0029] FIG6a is a flowchart of constructing a semantic codebook and synonymous mapping relationship according to an embodiment of the present application;

[0030] FIG6 b is a flowchart of another embodiment of the present application for constructing a semantic codebook and synonymous mapping relationship;

[0031] FIG7a is a flowchart of determining a syntax codeword corresponding to a syntax information sequence according to an embodiment of the present application;

[0032] FIG7 b is a flowchart of a process of semantic channel coding according to an embodiment of the present application;

[0033] FIG8a is a flowchart of a signal recovery process according to an embodiment of the present application;

[0034] FIG8 b is a flowchart of a process for semantic channel decoding based on the maximum likelihood group decoding criterion according to an embodiment of the present application;

[0035] FIG9 is a flowchart of a process for semantic channel decoding based on the logarithmic maximum likelihood group decoding criterion according to an embodiment of the present application;

[0036] FIG10 is a flowchart of a process for semantic channel decoding based on the minimum group Euclidean distance decoding criterion according to an embodiment of the present application;

[0037] FIG11 is a flowchart of a process for semantic channel decoding based on the maximum group correlation metric decoding criterion according to an embodiment of the present application;

[0038] FIG12 is a flowchart of a process for semantic channel decoding based on the minimum group Hamming distance decoding criterion according to an embodiment of the present application;

[0039] FIG13 is a flowchart of determining a target syntax information sequence set according to an embodiment of the present application;

[0040] FIG14 is a schematic diagram of verification result data of an embodiment of the present application;

[0041] FIG15a is a schematic structural diagram of a semantic channel coding apparatus according to an embodiment of the present application;

[0042] FIG15 b is a schematic structural diagram of a semantic channel decoding device according to an embodiment of the present application;

[0043] FIG16 is a schematic structural diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0044] In order to make the objectives, technical solutions and advantages of this application more clear, this application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0045] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should have the usual meanings understood by people with ordinary skills in the field to which this application belongs. The "first", "second" and similar words used in the embodiments of the present application do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0046] It should be understood herein that any number of elements in the drawings is for illustration only and not for limitation, and any naming is only for distinction and does not have any limiting meaning.

[0047] Based on the description of the above background technology, the following situations also exist in the related art:

[0048] In related technologies, the grammatical channel communication model includes the following parts:

[0049] Source: The carrier of information, which can be simply understood as the sender of information. The default output is binary syntax codeword.

[0050] Source encoder: Lossless encoding (or lossy encoding that meets the requirements) of the source output is performed to reduce the redundancy of the source output information. This can be understood as compressing the source.

[0051] Channel encoder: Encodes the output of the source encoder so that the resulting signal sequence can be better transmitted in the channel: generally, redundancy is added to enhance anti-interference capabilities.

[0052] Channel: Information is transmitted in the channel and sent to the receiving end.

[0053] Channel decoder: Decodes the received signal sequence and can recover certain transmission errors.

[0054] Source decoder: Decodes the output of the channel decoder to restore the original information sequence.

[0055] Therefore, while the channel encoder in the related art encodes syntactic codewords, in the semantic channel of semantic communication, the channel encoder encodes synonymous sets consisting of at least one syntactic codeword, rather than syntactic codewords. This makes the channel encoder in the syntactic channel communication model no longer suitable for directly encoding synonymous sets, resulting in less accurate encoding. Furthermore, the channel decoder in the syntactic channel communication model no longer suitable for directly decoding synonymous sets, resulting in less accurate decoding by the channel decoder and, consequently, less accurate transmission of semantic information.

[0056] The performance metric for evaluating semantic channel coding is the Semantic Information Error Rate (SIER), which is the ratio of the number of semantic information errors to the total number of semantic information transmitted. During a semantic information transmission, if the syntactic information sequence recovered by the receiver and the syntactic information sequence sent by the transmitter are not in the same set of syntactic information sequences after synonymous mapping, the transmission process is said to have erroneously transmitted semantic information. The total number of semantic information transmitted is the total number of transmissions, and the number of erroneously transmitted semantic information is the number of transmissions in which semantic information was erroneously transmitted.

[0057] Therefore, after the transmission object changes from a codeword to a synonym set, the encoding and decoding methods in the related technology will lead to an increase in the rate of erroneous semantic information, resulting in frequent transmission errors of semantic information.

[0058] The encoding and decoding method and electronic device of the semantic channel provided in the embodiment of the present application can determine the grammatical information sequence to be sent after obtaining the pre-built semantic codebook and synonymous mapping relationship, determine the grammatical codeword corresponding to the grammatical information sequence in the semantic codebook according to the synonymous mapping relationship, modulate the grammatical codeword to obtain a signal vector; perform signal recovery processing on the signal vector according to the semantic codebook and the preset decoding criteria to obtain the target synonymous set number for representing the semantic information, and determine the target grammatical information sequence set corresponding to the target synonymous set number according to the synonymous mapping relationship, and select a grammatical information sequence from the target grammatical information sequence set as the output result. A bridge is built between the grammatical channel encoding and decoding scheme and the semantic channel encoding and decoding scheme through the synonymous mapping relationship, so as to expand the encoding and decoding method of the grammatical channel to the encoding and decoding method of the semantic channel, thereby reducing the probability of errors in the received semantic information in the semantic scenario and reducing the error semantic information rate.

[0059] The following describes a method for encoding and decoding a semantic channel according to an exemplary embodiment of the present application with reference to the accompanying drawings.

[0060] In some embodiments, the semantic channel encoding and decoding method is applied to a channel communication system in a semantic communication scenario as shown in Figure 1. The channel communication system in the semantic communication scenario includes three modules: a relationship building module for generating a semantic codebook and establishing synonymous mappings, a semantic channel encoding module, and a semantic channel decoding module.

[0061] The input to the relational building module is the set of all possible syntactic information sequences (i.e., the syntactic information sequence set), and its output is a semantic codebook and synonymous mappings. Specifically, the relational building module generates a semantic channel coding codebook, establishes synonymous mappings between semantic information and syntactic information sequences, and synchronizes the codebook and synonymous mappings to the transmitter and receiver. The synonymous mapping is a mapping relationship between the synonymous set number representing the semantic information and the syntactic information sequence.

[0062] The input of the semantic channel coding module is a grammatical information sequence and the semantic codebook and synonymous mapping output from the module for generating the semantic codebook and establishing the synonymous mapping, and the output is a grammatical codeword.

[0063] The input of the semantic channel decoding module is the received signal vector obtained from the channel and the semantic codebook and synonymous mapping output from the module for generating semantic codebook and establishing synonymous mapping. The output is a grammatical information sequence.

[0064] In some embodiments, as shown in FIG2a , a method for encoding a semantic channel includes:

[0065] Step 201: Obtain a pre-built semantic codebook and synonym mapping relationship.

[0066] In specific implementation, as shown in Figures 3a and 3b, the semantic channel coding codebook is obtained by dividing the synonymous sets based on the maximum criterion of the Hamming distance between the minimum synonymous sets on the grammatical channel coding codebook, and the elements in the semantic channel coding codebook are synonymous sets. After the semantic codebook is created, a mapping relationship can be constructed between the synonymous set serial number representing the semantic information and the grammatical information sequence corresponding to the grammatical codeword in the synonymous set. This mapping is called a synonymous mapping. Finally, the synonymous mapping relationship of the semantic codebook is synchronized to the transmitting end and the receiving end. Among them, for the convenience of explanation, the semantic channel coding codebook is referred to as the semantic codebook, and the synonymous mapping relationship between the synonymous set serial number representing the semantic information and the grammatical information sequence is referred to as the synonymous mapping relationship, corresponding to the synonymous mapping f.

[0067] Among them, you can use Indicates the amount of semantic information, and N indicates the number of grammatical codewords. The grammatical codewords are divided into Synonymous set, i s Synonym sets contain Syntax codewords, i sThat is the i s The synonym set number of the synonym set. The grammar codeword set is denoted as C, the i-th s Synonymous set is denoted as i-th s The jth syntax codeword in the synonym set is denoted as c n (i s ,j).

[0068] Step 202: Determine the syntax information sequence to be sent.

[0069] Step 203: Determine the syntax codeword corresponding to the syntax information sequence in the semantic codebook according to the synonymous mapping relationship.

[0070] Step 204: Modulate the syntax codeword to obtain a signal vector.

[0071] In specific implementation, the encoding process at the transmitting end is shown in FIG4 . At the transmitting end, semantic information is extracted from the grammatical information sequence to be transmitted according to the synonymous mapping relationship, and then a grammatical codeword is used to carry the semantic information, and the signal vector modulated by the grammatical codeword is sent to the channel. First, according to the received synonymous mapping relationship, the grammatical information sequence to be transmitted is mapped to its corresponding synonymous set number representing the semantic information; secondly, the synonymous set corresponding to the synonymous set number is found in the semantic codebook, and a grammatical codeword is arbitrarily selected from the synonymous set as the codeword to be transmitted, or a grammatical codeword is selected from the synonymous set as the codeword to be transmitted according to a preset selection rule; finally, the grammatical codeword is modulated, and the modulated signal is transmitted to the channel. Exemplarily, the preset selection rule can be to calculate the degree of similarity between each grammatical codeword in the synonymous set and the target semantics (e.g., happy), and select the grammatical codeword according to the degree of similarity from high to low.

[0072] The modulation process is to convert the i-th s The initial signal vector after the modulation of the j-th syntax codeword with code length n in the synonymous set is recorded as x n (i s ,j), referred to as x n , i-th s The modulated signal of the kth codeword in the jth syntax codeword in the synonymous set is denoted as x (k) (i s ,j), initial signal vector x n and the channel noise z n Superposition is performed, and the final signal vector y n , that is, y n =x n +z n .

[0073] In some embodiments, as shown in FIG2b , a decoding method for a semantic channel includes:

[0074] Step 211: Obtain a pre-built semantic codebook and synonym mapping relationship.

[0075] Step 212: Perform signal recovery processing on the signal vector according to the semantic codebook and a preset decoding criterion to obtain a target synonym set number for representing semantic information.

[0076] Step 213: Determine the target syntax information sequence set corresponding to the target synonym set sequence number according to the synonym mapping relationship.

[0077] Step 214: Select a syntax information sequence from the target syntax information sequence set as an output result.

[0078] In specific implementation, the decoding process at the receiving end is shown in Figure 5. At the receiving end, the received signal is directly judged as the most likely semantic information, and then a syntax information sequence is used to carry this semantic information as the output result. Exemplarily, first, the synonym sets in the semantic codebook are traversed, and the target synonym set corresponding to the semantic information most likely to be sent is selected through the maximum likelihood group decoding criterion, and then the target synonym set sequence number corresponding to the target synonym set is determined; then, according to the synonym mapping, the target synonym set sequence number representing the semantic information is mapped to a target syntax information sequence set. Then, a syntax information sequence can be arbitrarily selected from the target syntax information sequence set as the output result, or a syntax information sequence can be selected from the target syntax information sequence set as the output result according to a preset sequence selection rule. Exemplarily, the preset sequence selection rule can be to calculate the degree of similarity between each syntax information sequence in the target syntax information sequence set and the target semantics, and select the syntax information sequence from high to low according to the degree of similarity.

[0079] In summary, the encoding and decoding method of the semantic channel provided in the embodiment of the present application can determine the grammatical information sequence to be sent after obtaining the pre-constructed semantic codebook and synonymous mapping relationship, determine the grammatical codeword corresponding to the grammatical information sequence in the semantic codebook according to the synonymous mapping relationship, modulate the grammatical codeword, and obtain a signal vector; perform signal recovery processing on the signal vector according to the semantic codebook and the preset decoding criteria to obtain the target synonymous set number for representing the semantic information, and determine the target grammatical information sequence set corresponding to the target synonymous set number according to the synonymous mapping relationship, and select a grammatical information sequence from the target grammatical information sequence set as the output result. A bridge is built between the grammatical channel encoding and decoding scheme and the semantic channel encoding and decoding scheme through the synonymous mapping relationship, so as to expand the encoding and decoding method of the grammatical channel to the encoding and decoding method of the semantic channel, thereby reducing the probability of errors in the received semantic information in the semantic scenario and reducing the error semantic information rate.

[0080] In some embodiments, as shown in FIG6a , the semantic codebook and synonym mapping relationship are constructed by the following method:

[0081] Step 601: Determine a syntax codeword corresponding to each syntax information sequence in an input syntax information sequence set.

[0082] In specific implementations, as shown in Figures 3a and 3b, the input to the syntax channel encoder should be all possible syntax information sequences, namely, the input syntax information sequence set. Each syntax information sequence in the syntax information sequence set is encoded to obtain the corresponding syntax codeword. The output set of the syntax channel encoder is a syntax channel coding codebook containing all syntax codewords. Furthermore, to maximize the distance between the divided synonymous sets, the selected syntax channel coding method should maximize the minimum Hamming distance between syntax codewords.

[0083] Therefore, when creating a semantic codebook, it is necessary to determine the syntax codeword corresponding to each syntax information sequence in the input syntax information sequence set to obtain a syntax channel coding codebook. Further processing of the syntax channel coding codebook can obtain the corresponding semantic codebook, thereby realizing a coding method that expands the syntax channel coding into the semantic channel.

[0084] Step 602: Divide the grammatical codewords into synonymous sets of the same number of semantics according to the number of semantics, and obtain a semantic codebook, wherein each codeword in the synonymous set expresses the same semantics.

[0085] In specific implementation, as shown in Figures 3a and 3b, the synonym set partitioner partitions the synonym sets in the generated grammatical channel coding codebook according to the amount of semantic information and the maximum criterion of the Hamming distance between the minimum synonym sets, so that the codewords in the same synonym set express the same semantics. The output of the synonym set partitioner is a semantic codebook, and each element in the semantic codebook is a synonym set.

[0086] Therefore, it is necessary to divide the grammatical codewords into semantically numbered synonymous sets according to the number of semantics to obtain a semantic codebook, so as to implement a coding method that expands the coding of the grammatical channel into the coding of the semantic channel.

[0087] After constructing the semantic codebook, the synonym mapping relationship will be further constructed through the following method.

[0088] Step 603: Determine the synonym set sequence number of each synonym set in the semantic codebook, and determine the grammatical information sequence corresponding to each synonym set, wherein the synonym set sequence number corresponds one-to-one to the semantic information.

[0089] In specific implementations, as shown in Figures 3a and 3b, the synonymous mapping relationship is established by mapping the synonymous set numbers representing semantic information to the grammatical information sequences. The synonymous mapping relationship must ensure that each synonymous set corresponds to at least one grammatical information sequence, and the number of grammatical codewords in the synonymous set must be the same as the number of grammatical information sequences obtained after synonymous mapping of the corresponding synonymous set numbers.

[0090] Therefore, we first need to determine the synonym set number for each synonym set in the semantic codebook. This is because the entity representation of semantic information is the synonym set number, which means that the synonym set number corresponds one-to-one with the semantic information. After determining the data on one side of the synonym mapping relationship, we need to determine the data on the other side of the synonym mapping relationship, that is, the grammatical information sequence corresponding to each synonym set. This is because the synonym mapping relationship is a mapping relationship between the synonym set number representing the semantic information and the grammatical information sequence.

[0091] For example, grammatical codewords grouped into the same synonymous set are synonymous; they express the same meaning, but differ in grammatical expression. For example, to express the meaning of "happy," one could use the grammatical codeword corresponding to "joy" or the grammatical codeword corresponding to "happy." Then, all grammatical codewords expressing "happy" are combined into a set, which is a synonymous set. This synonymous set has a one-to-one correspondence with the semantic information of "happy," and each synonymous set has a unique synonymous set sequence number. Therefore, the synonymous set sequence number of the synonymous set composed of all grammatical codewords expressing "happy" has a one-to-one correspondence with the semantic information of "happy."

[0092] Semantic information is an abstract concept. Since it is not a specific sequence or codeword, synonym set numbers are used to represent semantic information.

[0093] Step 604: Establish a corresponding relationship between the synonym set sequence number and the grammatical information sequence corresponding to the same synonym set to obtain a synonym mapping relationship.

[0094] In specific implementation, since the synonymous mapping relationship is a mapping relationship between synonymous set numbers representing semantic information and grammatical information sequences, the synonymous mapping relationship can be obtained by establishing a corresponding relationship between synonymous set numbers and grammatical information sequences corresponding to the same synonymous set.

[0095] In summary, as shown in FIG6b , the specific process of establishing a synonymous mapping relationship includes:

[0096] First, syntax channel coding is used to generate syntax codewords from all possible syntax information sequences (ie, syntax information sequence sets).

[0097] The syntax channel coding scheme encodes a syntax information sequence of length k into a syntax codeword of length n, and there is a one-to-one mapping relationship between the syntax information sequence and the syntax codeword.

[0098] Secondly, the grammatical codewords are divided into semantic number synonymous sets according to the semantic number of the semantic information, so as to realize the creation of the semantic codebook.

[0099] Among them, the synonym sets in the semantic codebook need to satisfy that any two synonym sets are disjoint, and the union of the grammatical codewords in all synonym sets is all grammatical codewords, that is,

[0100] Then calculate the group Hamming distance between any two synonymous sets, the i-th s Synonymous sets and j s The calculation formula of the Hamming distance between synonym sets is:

[0101] The minimum Hamming distance between synonymous sets in the semantic codebook is

[0102] In order to make the decision domain of each synonym set as large as possible, the minimum Hamming distance between synonym sets in the semantic codebook needs to be as large as possible. Therefore, the method of dividing synonym sets we choose should satisfy the requirement that the minimum Hamming distance d between synonym sets in the semantic codebook is d when the number of semantics is given. GH,min maximum.

[0103] Finally, a synonym mapping relationship is established between the synonym set sequence number representing the semantic information and the grammatical information sequence.

[0104] For any synonym set, from its sequence number i s The mapping to all grammatical codewords within the synonym set is a fixed mapping relationship, and each grammatical codeword has a one-to-one mapping relationship with a grammatical information sequence. Therefore, after the transition of grammatical codewords, a fixed mapping relationship exists between the synonym set number and the grammatical information sequence. By traversing all synonym sets, a synonymous mapping relationship between the synonym set number representing semantic information and the grammatical information sequence can be constructed. Ultimately, a synonymous mapping relationship between the synonym set number representing semantic information and the grammatical information sequence is established.

[0105] It should be noted that after generating the semantic codebook and establishing a synonymous mapping relationship between semantic information and syntactic information, the semantic codebook and the synonymous mapping relationship need to be synchronized to the transmitter and the receiver, so that the transmitter can implement semantic channel coding based on the semantic codebook and the synonymous mapping, and the receiver can implement semantic channel decoding based on the semantic codebook and the synonymous mapping relationship.

[0106] In some embodiments, as shown in FIG7a , determining a syntax codeword corresponding to a syntax information sequence in a semantic codebook according to a synonymous mapping relationship includes:

[0107] Step 701: Map the grammatical information sequence to be sent into a synonymous set number for representing semantic information according to a synonymous mapping relationship.

[0108] In specific implementations, as shown in Figure 4, the synonym mapper takes as input a grammatical information sequence and a synonym mapping relationship output by the semantic codebook generation and synonym mapping relationship building module. Its output is a synonym set number representing the semantic information. Therefore, the grammatical information sequence to be transmitted can be mapped to a synonym set number representing the semantic information based on the synonym mapping relationship.

[0109] For example, when the content of the grammatical information sequence is a sequence representing "happy", according to the synonymous mapping relationship, the synonymous set number expressing the semantic information of "happy" can be found by taking the grammatical codeword expressing the semantics of "happy" as a transition.

[0110] Step 702: Determine the target synonym set corresponding to the synonym set sequence number in the semantic codebook.

[0111] In specific implementations, as shown in Figure 4, the codeword generator takes as input a synonym set number representing semantic information and the semantic codebook output from the semantic codebook generation and synonym mapping modules. The output is a grammatical codeword. Since there is a one-to-one correspondence between synonym sets and synonym set numbers, once the synonym set number is determined, the target synonym set corresponding to that synonym set number can be determined in the semantic codebook. For example, the target synonym set consists of all grammatical codewords expressing the word "happy."

[0112] Step 703: Select a codeword in the target synonym set as the grammatical codeword.

[0113] In specific implementation, after determining the target synonym set, since the target synonym set is composed of grammatical codewords expressing the same semantics, a codeword can be selected from the target synonym set as the grammatical codeword.

[0114] As shown in Figure 7b, the processing flow of semantic channel coding includes:

[0115] 1. Extract semantic information from grammatical information sequences based on synonym mapping.

[0116] According to the synonymous mapping relationship between the synonymous set number representing the semantic information and the grammatical information sequence, the grammatical information sequence u to be sent is k Mapped to the synonym set number i representing semantic information s , to achieve the extraction of semantic information corresponding to the grammatical information sequence.

[0117] 2. Select a syntax codeword from the target synonym set corresponding to the semantic information to be sent according to the semantic codebook, and use the syntax codeword as the channel coding codeword for the semantic information.

[0118] The synonym set number i corresponding to the semantic information can be s Corresponding target synonym set Choose any grammatical codeword x from n (i s ,j) as the channel coding codeword of the semantic information. It is also possible to use the prior information to represent the sequence number i of the semantic information. s Corresponding target synonym set Select a syntax codeword x according to the preset selection rules n (i s ,j) as the channel coding codeword for the semantic information. For example, the preset selection rule may be to calculate the similarity between each grammatical codeword in the target synonym set and the target semantics (e.g., happy), and select grammatical codewords based on the similarity from high to low.

[0119] In some embodiments, the semantic channel decoding process is applicable to semantic communication scenarios in which a discrete memoryless channel is used and the receiver knows the channel state information. As shown in Figure 5, in the semantic channel decoding process, the input of the synonymous set decider is the received signal vector obtained from the channel and the semantic codebook output from the semantic codebook generation and synonym mapping module establishment module, and the output is a synonymous set serial number of the semantic information. Appropriate decoding criteria need to be used in the synonymous set decider to match the division method of synonymous sets in semantic channel coding. Five decoding criteria are proposed in the embodiments of the present application, namely, maximum likelihood group decoding criterion, logarithmic maximum likelihood group decoding criterion, minimum group Euclidean distance decoding, minimum group Hamming distance decoding criterion and maximum correlation metric criterion. Among them, the maximum likelihood group decoding criterion is the core criterion, and the other decoding criteria are equivalent criteria of the maximum likelihood group decoding criterion under different conditions.

[0120] The input of the synonym mapper is a synonym set number of semantic information and the obtained synonym mapping relationship, and the output is a grammatical information sequence. It should be noted that although there are synonym mappers in both the semantic channel coding module and the semantic channel decoding module, the two are not exactly the same. The synonym mapper in the semantic channel coding module maps a grammatical information sequence to a synonym set number representing semantic information, and its mapping result is deterministic; the synonym mapper in the semantic channel decoding module maps a synonym set number of semantic information to a grammatical information sequence, and its mapping result is not deterministic.

[0121] In some embodiments, the decoding criterion includes a maximum likelihood group decoding criterion. As shown in FIG8a , signal recovery processing is performed on the signal vector according to the semantic codebook and the preset decoding criterion to obtain the target synonymous set sequence number. Obtaining the target synonymous set sequence number includes:

[0122] Step 801: Determine the group likelihood probability between the signal vector and each synonym set in the semantic codebook.

[0123] Step 802: Determine the maximum group likelihood probability among the group likelihood probabilities.

[0124] Step 803: Determine the first synonymous set corresponding to the maximum group likelihood probability, and determine the synonymous set sequence number of the first synonymous set as the target synonymous set sequence number.

[0125] In specific implementation, as shown in FIG8b , the processing flow of semantic channel decoding based on the maximum likelihood group decoding criterion includes:

[0126] The signal vector y is received from the channel nAfter that. Select a non-repeated synonymous set from the semantic codebook, calculate the group likelihood probability between the synonymous set and the signal vector, and determine whether the group likelihood probability is the current maximum group likelihood probability; if it is the current maximum group likelihood probability, record the synonymous set serial number of the synonymous set, and determine whether all synonymous sets in the semantic codebook have been traversed; if it is not the current maximum group likelihood probability, directly determine whether all synonymous sets in the semantic codebook have been traversed. If the traversal process has not ended, continue to select a non-repeated synonymous set from the semantic codebook. If the traversal process is ended, output the maximum group likelihood probability. And select the synonymous set serial number of the synonymous set corresponding to the maximum group likelihood probability in the synonymous set judge. As a result of the judgment, the target synonym set number is obtained

[0127] Among them, the i s The calculation process of the group likelihood probability of a synonymous set is:

[0128] In the above obtained The largest synonym set is selected from the group likelihood probabilities of the synonym sets as the first synonym set corresponding to the maximum group likelihood probability, and the synonym set number of the first synonym set is determined as the target synonym set number, that is, the target synonym set number is set as the target synonym set number. As the decision result of the synonym set decider.

[0129] Then, use the synonym mapper to map the resulting synonym set number Mapped into a grammatical information sequence. According to the one-to-one mapping rule between semantic information and synonym set sequence number, the synonym set sequence number obtained is determined That is, it represents semantic information, and arbitrarily selects the first synonym set corresponding to it A grammatical codeword x in n (i s ,j) or according to the prior information in its corresponding synonymous set Select a syntax code word x according to the preset selection rules n (i s ,j), and then map the syntax codeword into a syntax information sequence This syntactic information sequence is the output of the semantic channel decoder.

[0130] In some embodiments, the decoding criterion includes a logarithmic maximum likelihood group decoding criterion. As shown in FIG9 , signal recovery processing is performed on the signal vector according to the semantic codebook and the preset decoding criterion to obtain the target synonymous set sequence number. Obtaining the target synonymous set sequence number includes:

[0131] Step 901: Determine the log-group likelihood between the signal vector and each synonym set in the semantic codebook.

[0132] Step 902: Determine the maximum log-group likelihood probability among the log-group likelihood probabilities.

[0133] Step 903: Determine a second synonymous set corresponding to the maximum log-group likelihood probability, and determine the synonymous set sequence number of the second synonymous set as the target synonymous set sequence number.

[0134] In specific implementation, the processing flow of semantic channel decoding based on the logarithmic maximum likelihood group decoding criterion includes:

[0135] The signal vector y is received from the channel n After that. Select a non-repeated synonymous set from the semantic codebook, calculate the logarithmic group likelihood probability between the synonymous set and the signal vector, and determine whether the logarithmic group likelihood probability is the current maximum logarithmic group likelihood probability; if it is the current maximum logarithmic group likelihood probability, record the synonymous set serial number of the synonymous set, and determine whether all synonymous sets in the semantic codebook have been traversed; if it is not the current maximum logarithmic group likelihood probability, directly determine whether all synonymous sets in the semantic codebook have been traversed. If the traversal process has not ended, continue to select a non-repeated synonymous set from the semantic codebook. If the traversal process is ended, output the maximum logarithmic group likelihood probability. And select the synonymous set serial number of the synonymous set corresponding to the maximum logarithmic group likelihood probability in the synonymous set judge. As a result of the judgment, the target synonym set number is obtained

[0136] Among them, the i s The calculation process of the log-group likelihood probability of a synonymous set is:

[0137] In the above obtained The largest synonym set is selected from the logarithmic group likelihood probabilities of the synonym sets as the second synonym set corresponding to the largest logarithmic group likelihood probability, and the synonym set number of the second synonym set is determined as the target synonym set number, that is, the target synonym set number is As the decision result of the synonym set decider.

[0138] It should be noted that the logarithmic maximum likelihood group decoding criterion is essentially no different from the maximum likelihood group decoding criterion. However, in some embodiments, the length of the syntax codeword is relatively long, resulting in a small calculated group likelihood probability, which is prone to overflow during the hardware calculation process. The logarithmic maximum likelihood group decoding criterion can effectively improve the problem of overflow that is prone to occur during the calculation process.

[0139] In some embodiments, the decoding criterion includes a minimum group Euclidean distance decoding criterion. As shown in FIG10 , signal recovery processing is performed on the signal vector according to the semantic codebook and the preset decoding criterion to obtain the target synonymous set sequence number. Obtaining the target synonymous set sequence number includes:

[0140] Step 1001: Determine the group Euclidean distance between the signal vector and each synonym set in the semantic codebook.

[0141] Step 1002: Determine the minimum group Euclidean distance among the group Euclidean distances.

[0142] Step 1003: Determine a third synonymous set corresponding to the minimum group Euclidean distance, and determine the synonymous set sequence number of the third synonymous set as the target synonymous set sequence number.

[0143] In specific implementation, the processing flow of semantic channel decoding based on the minimum group Euclidean distance decoding criterion includes:

[0144] The signal vector y is received from the channel n After. Select a non-repeated synonymous set from the semantic codebook, calculate the group Euclidean distance between the synonymous set and the signal vector, and determine whether the group Euclidean distance is the current minimum group Euclidean distance; if it is the current minimum group Euclidean distance, record the synonymous set serial number of the synonymous set, and determine whether all synonymous sets in the semantic codebook have been traversed; if it is not the current minimum group Euclidean distance, directly determine whether all synonymous sets in the semantic codebook have been traversed. If the traversal process has not ended, continue to select a non-repeated synonymous set from the semantic codebook. If the traversal process is ended, output the minimum group Euclidean distance. And select the synonymous set serial number of the synonymous set corresponding to the minimum group Euclidean distance in the synonymous set judge. As a result of the judgment, the target synonym set number is obtained

[0145] Among them, the i s The calculation process of the group Euclidean distance of a synonymous set is:

[0146] In the above obtained The smallest synonym set among the group Euclidean distances of the synonym sets is selected as the third synonym set corresponding to the minimum group Euclidean distance, and the synonym set number of the third synonym set is determined as the target synonym set number, that is, the target synonym set number As the decision result of the synonym set decider.

[0147] It should be noted that the minimum group Euclidean distance decoding criterion is an equivalent form of the maximum likelihood group decoding criterion under additive white Gaussian noise (AWGN) channel conditions.

[0148] In some embodiments, the decoding criterion includes a maximum group correlation metric decoding criterion. As shown in FIG11 , signal recovery processing is performed on the signal vector according to the semantic codebook and the preset decoding criterion to obtain the target synonymous set sequence number. Obtaining the target synonymous set sequence number includes:

[0149] Step 1101: Determine a group correlation measure between a signal vector and each synonym set in a semantic codebook.

[0150] Step 1102: Determine the maximum group correlation metric among the group correlation metrics.

[0151] Step 1103: Determine the fourth synonymous set corresponding to the maximum group correlation metric, and determine the synonymous set sequence number of the fourth synonymous set as the target synonymous set sequence number.

[0152] In specific implementation, the processing flow of semantic channel decoding based on the maximum group correlation metric decoding criterion includes:

[0153] The signal vector y is received from the channel n After that. Select a non-repeated synonymous set from the semantic codebook, calculate the group correlation measure between the synonymous set and the signal vector, and determine whether the group correlation measure is the current maximum group correlation measure; if it is the current maximum group correlation measure, record the synonymous set serial number of the synonymous set, and determine whether all synonymous sets in the semantic codebook have been traversed; if it is not the current maximum group correlation measure, directly determine whether all synonymous sets in the semantic codebook have been traversed. If the traversal process has not ended, continue to select a non-repeated synonymous set from the semantic codebook. If the traversal process is ended, output the maximum group correlation measure. And select the synonymous set serial number of the synonymous set corresponding to the maximum group correlation measure in the synonymous set judge. As a result of the judgment, the target synonym set number is obtained

[0154] Among them, the i s The calculation process of the group correlation measure of a synonymous set is:

[0155] In the above obtained The largest synonym set is selected from the group correlation measures of the synonym sets as the fourth synonym set corresponding to the largest group correlation measure, and the synonym set number of the fourth synonym set is determined as the target synonym set number, that is, the target synonym set number As the decision result of the synonym set decider.

[0156] It should be noted that the maximum group correlation metric decoding criterion is an equivalent form of the maximum likelihood group decoding criterion under the conditions of additive white Gaussian noise (AWGN) channel and normalized transmitted signal energy.

[0157] In some embodiments, the decoding criterion includes a minimum group Hamming distance decoding criterion. As shown in FIG12 , signal recovery processing is performed on the signal vector according to the semantic codebook and the preset decoding criterion to obtain the target synonymous set sequence number. Obtaining the target synonymous set sequence number includes:

[0158] Step 1201: Determine the group Hamming distance between the signal vector and each synonym set in the semantic codebook.

[0159] Step 1202: Determine the minimum group Hamming distance among the group Hamming distances.

[0160] Step 1203: Determine the fifth synonymous set corresponding to the minimum group Hamming distance, and determine the synonymous set sequence number of the fifth synonymous set as the target synonymous set sequence number.

[0161] In specific implementation, the processing flow of semantic channel decoding based on the minimum group Hamming distance decoding criterion includes:

[0162] The signal vector y is received from the channel n After that. Select a non-repeated synonymous set from the semantic codebook, calculate the group Hamming distance between the synonymous set and the signal vector, and determine whether the group Hamming distance is the current minimum group Hamming distance; if it is the current minimum group Hamming distance, record the synonymous set serial number of the synonymous set, and determine whether all synonymous sets in the semantic codebook have been traversed; if it is not the current minimum group Hamming distance, directly determine whether all synonymous sets in the semantic codebook have been traversed. If the traversal process has not ended, continue to select a non-repeated synonymous set from the semantic codebook. If the traversal process is over, output the minimum group Hamming distance. And select the synonymous set serial number of the synonymous set corresponding to the minimum group Hamming distance in the synonymous set judge. As a result of the judgment, the target synonym set number is obtained

[0163] Among them, the i s The calculation process of the group Hamming distance of a synonymous set is:

[0164] In the above obtained The smallest synonym set among the group Hamming distances of the synonym sets is selected as the third synonym set corresponding to the minimum group Hamming distance, and the synonym set number of the third synonym set is determined as the target synonym set number, that is, the target synonym set number As the decision result of the synonym set decider.

[0165] It should be noted that the minimum group Hamming distance decoding criterion is an equivalent form of the maximum likelihood group decoding criterion under binary symmetric channel (BSC) conditions.

[0166] In some embodiments, as shown in FIG13 , determining a target syntax information sequence set corresponding to a target synonym set sequence number according to a synonym mapping relationship includes:

[0167] Step 1301: Determine the selected synonym set corresponding to the target synonym set sequence number.

[0168] In specific implementation, according to the one-to-one correspondence between synonym sets and synonym set serial numbers, the target synonym set serial number can be determined. Corresponding selection synonym set

[0169] Step 1302: Determine the target syntax information sequence corresponding to each syntax codeword in the selected synonymous set according to the synonymous mapping relationship, and integrate the target syntax information sequences to obtain a target syntax information sequence set.

[0170] In specific implementation, the synonym set is determined and selected according to the synonym mapping relationship. The target syntax information sequence corresponding to each syntax codeword in , and the target syntax information sequence is integrated to obtain the target syntax information sequence set, and any syntax information sequence is selected from the target syntax information sequence set As the output result of the semantic channel decoder.

[0171] Optionally, you can also select a synonym set A grammatical codeword x randomly selected from n (i s ,j) or according to the prior information in its corresponding synonymous set Select a syntax code word x according to the preset selection rules n (i s ,j), and then map the syntax codeword into a syntax information sequence And the grammatical information sequence As the output result of the semantic channel decoder.

[0172] The verification data results of the encoding and decoding method of the semantic channel provided by the embodiment of the present invention are listed below.

[0173] Experimental conditions include:

[0174] The selected scenario is a semantic communication scenario in which the code elements are binary code elements, the modulation method is BPSK, and both the sender and the receiver know the channel state information under the condition of discrete memoryless channel.

[0175] Polar coding was selected as the syntax channel coding method. The syntax codeword length was 8 bits, and the syntax information sequence length was 4 bits. There were 8 types of semantic information and 8 types of synonym sets. Each synonym set contained 2 syntax codewords, and the minimum inter-synonymous set Hamming distance was 2.

[0176] The average symbol power of the syntax information sequence during transmission is set to 1, and the average noise power in the channel ranges from [0.1585, 2.512]. This means the received signal-to-noise ratio ranges from -4dB to 8dB. The performance metric for evaluating semantic channel coding is the semantic error information rate.

[0177] Figure 14 shows the performance of the semantic channel coding scheme in a semantic communication scenario over an additive white Gaussian noise (AWGN) channel under the semantic error information rate metric, as well as the upper bound of this performance. The semantic channel coding scheme uses the criterion of maximizing the minimum Hamming distance between synonymous sets to partition synonymous sets, and the semantic channel decoder uses the maximum likelihood group decoding criterion.

[0178] From the result data shown in Figure 14, we can see that:

[0179] The encoding and decoding method of the semantic channel provided in the present embodiment can recover the grammatical information sequence containing semantic information for each signal-to-noise ratio value within the test signal-to-noise ratio range under the condition of an additive white Gaussian noise channel (AWGN), and can obtain the semantic channel coding error rate. The semantic error rate decreases at an exponential rate as the signal-to-noise ratio increases and approaches 0.

[0180] It should be noted that, since the conditions are additive white Gaussian noise channel (AWGN) and signal energy normalization, the logarithmic maximum likelihood group decoding criterion, the minimum group Euclidean distance decoding criterion, and the maximum correlation metric decoding criterion are all equivalent to the maximum likelihood group decoding criterion. The performance of this embodiment also represents the performance of these three decoding criteria under the conditions of this embodiment.

[0181] It should be noted that the method of the embodiment of the present application can be performed by a single device, such as a computer or server. The method of this embodiment can also be applied in a distributed scenario and performed by multiple devices working together. In such a distributed scenario, one of the multiple devices may only perform one or more steps of the method of the embodiment of the present application, and the multiple devices will interact with each other to complete the method.

[0182] It should be noted that the above description is limited to some embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in an order different from that described in the above embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0183] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a coding device for a semantic channel.

[0184] Referring to FIG15a , the encoding apparatus of the semantic channel includes:

[0185] The relationship parameter acquisition module 1512 is used to obtain the pre-built semantic codebook and synonym mapping relationship;

[0186] Syntax information sequence determination module 1514, used to determine the syntax information sequence to be sent;

[0187] a syntax codeword determination module 1516, configured to determine a syntax codeword corresponding to the syntax information sequence in the semantic codebook according to the synonymous mapping relationship; and

[0188] The modulation module 1518 is configured to modulate the syntax codeword to obtain a signal vector.

[0189] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a decoding device for a semantic channel.

[0190] Referring to FIG15b , the decoding device of the semantic channel includes:

[0191] The relationship parameter acquisition module 1522 is used to obtain the pre-built semantic codebook and synonym mapping relationship;

[0192] A recovery module 1524 is configured to perform signal recovery processing on the signal vector according to the semantic codebook and a preset decoding criterion to obtain a target synonym set number for representing semantic information;

[0193] A grammar information sequence set determination module 1526 is configured to determine a target grammar information sequence set corresponding to a target synonym set sequence number according to the synonym mapping relationship; and

[0194] The output module 1528 is configured to select a grammar information sequence from the target grammar information sequence set as an output result.

[0195] For the convenience of description, the above devices are described as being divided into various modules according to their functions. Of course, when implementing this application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0196] The apparatus of the above embodiment is used to implement the encoding and decoding method of the corresponding semantic channel in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0197] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and runnable on the processor, wherein when the processor executes the program, the encoding and decoding method of the semantic channel described in any of the above embodiments is implemented.

[0198] FIG16 shows a more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment. The device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are communicatively connected to each other within the device via the bus 1050.

[0199] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0200] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage devices, dynamic storage devices, etc. The memory 1020 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0201] The input / output interface 1030 is used to connect input / output modules to implement information input and output. The input / output modules can be configured as components within the device (not shown in the figure) or can be externally connected to the device to provide corresponding functions. Input devices may include a keyboard, mouse, touch screen, microphone, various sensors, etc., and output devices may include a display, speaker, vibrator, indicator light, etc.

[0202] The communication interface 1040 is used to connect to a communication module (not shown) to enable communication between the device and other devices. The communication module can communicate via a wired method (such as USB, network cable, etc.) or a wireless method (such as mobile network, WiFi, Bluetooth, etc.).

[0203] The bus 1050 comprises a path for transmitting information between the various components of the device (eg, the processor 1010 , the memory 1020 , the input / output interface 1030 , and the communication interface 1040 ).

[0204] It should be noted that although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040, and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, it will be understood by those skilled in the art that the above device may only include the components necessary to implement the embodiments of this specification, and does not necessarily include all the components shown in the figure.

[0205] The electronic device of the above embodiment is used to implement the encoding and decoding method of the corresponding semantic channel in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0206] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the encoding and decoding method of the semantic channel as described in any of the above embodiments.

[0207] The computer-readable media of this embodiment include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0208] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the encoding and decoding method of the semantic channel as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0209] Based on the same concept, corresponding to any of the above-mentioned embodiments, the present application also provides a computer program product, including computer program instructions. When the computer program instructions are run on a computer, the computer executes the method described in any of the above embodiments, which has the beneficial effects of the corresponding method embodiments and will not be repeated here.

[0210] It is understandable that before using the technical solutions of each embodiment of the present disclosure, the type, scope of use, usage scenarios, etc. of the personal information involved will be informed to the user in an appropriate manner, and the user's authorization will be obtained.

[0211] For example, in response to a user's active request, a prompt message is sent to the user to clearly inform the user that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the electronic device, application, server, storage medium, or other software or hardware that performs the operation of the disclosed technical solution based on the prompt message.

[0212] As an optional but non-limiting implementation, in response to a user's active request, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. Furthermore, the pop-up window may also contain a selection control for the user to select "agree" or "disagree" to provide personal information to the electronic device.

[0213] It is understandable that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0214] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application (including the claims) is limited to these examples. Within the scope of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0215] In addition, for simplicity of description and discussion, and in order not to make the embodiment of the application difficult to understand, the known power supply / ground connection with integrated circuit (IC) chip and other components may or may not be shown in the accompanying drawings provided. In addition, the device can be shown in the form of a block diagram to avoid making the embodiment of the application difficult to understand, and this also takes into account the following fact, that is, the details of the embodiment of these block diagram devices are highly dependent on the platform to be implemented in the embodiment of the application (that is, these details should be fully within the scope of understanding of those skilled in the art). When specific details (for example, circuit) are set forth to describe exemplary embodiments of the application, it will be apparent to those skilled in the art that the embodiment of the application can be implemented without these specific details or when these specific details are changed. Therefore, these descriptions should be considered to be illustrative rather than restrictive.

[0216] Although the present invention has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those skilled in the art based on the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may utilize the embodiments discussed.

[0217] The embodiments of the present application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of this application.

Claims

1. A coding method for a semantic channel, comprising: Obtaining a pre - constructed semantic codebook and a synonym mapping relationship; Determining a syntax information sequence to be transmitted; Determining a syntax codeword corresponding to the syntax information sequence in the semantic codebook according to the synonym mapping relationship; And Modulating the syntax codeword to obtain a signal vector.

2. The method according to claim 1, further comprising: Determining a syntax codeword corresponding to each syntax information sequence in the input syntax information sequence set; Dividing the syntax codewords into the semantic number of synonym sets according to the number of semantics to obtain the semantic codebook; wherein, each codeword in the synonym set expresses the same semantics.

3. The method according to claim 2, further comprising: Determining the synonym set serial number of each synonym set in the semantic codebook; Determining the syntax information sequence corresponding to each synonym set; wherein, the synonym set serial number corresponds one - to - one with the semantic information; Establishing a correspondence relationship between the synonym set serial number and the syntax information sequence corresponding to the same synonym set to obtain the synonym mapping relationship.

4. The method according to claim 1, wherein The determining a syntax codeword corresponding to the syntax information sequence in the semantic codebook according to the synonym mapping relationship includes: Mapping the syntax information sequence to be transmitted to a synonym set serial number for representing semantic information according to the synonym mapping relationship; Determining a target synonym set corresponding to the synonym set serial number in the semantic codebook; Selecting a codeword from the target synonym set as the syntax codeword.

5. A decoding method for a semantic channel, comprising: Obtaining a pre - constructed semantic codebook and a synonym mapping relationship; Performing signal recovery processing on the signal vector according to the semantic codebook and a preset decoding criterion to obtain a target synonym set serial number for representing semantic information; Determining a target syntax information sequence set corresponding to the target synonym set serial number according to the synonym mapping relationship; And Selecting a syntax information sequence from the target syntax information sequence set as an output result.

6. The method according to claim 5, wherein The decoding criterion includes: maximum likelihood group decoding criterion; The performing signal recovery processing on the signal vector according to the pre - constructed semantic codebook and a preset decoding criterion to obtain a target synonym set serial number for representing semantic information includes: Determining the group likelihood probability between the signal vector and each synonym set in the semantic codebook; Determining the maximum group likelihood probability among the group likelihood probabilities; Determining a first synonym set corresponding to the maximum group likelihood probability and determining the synonym set serial number of the first synonym set as the target synonym set serial number.

7. The method according to claim 5, wherein The decoding criterion includes: logarithmic maximum likelihood group decoding criterion; The performing signal recovery processing on the signal vector according to the pre - constructed semantic codebook and a preset decoding criterion to obtain a target synonym set serial number for representing semantic information includes: Determining the logarithmic group likelihood probability between the signal vector and each synonym set in the semantic codebook; Determining the maximum logarithmic group likelihood probability among the logarithmic group likelihood probabilities; Determine a second synonymous set corresponding to the maximum logarithmic group likelihood probability, and determine the synonymous set number of the second synonymous set as the target synonymous set number.

8. The method according to claim 5, wherein The decoding criterion includes: minimum group Euclidean distance decoding criterion; The obtaining of the target synonymous set number for representing semantic information by performing signal recovery processing on the signal vector according to a pre-constructed semantic codebook and a preset decoding criterion includes: Determine the group Euclidean distance between the signal vector and each synonymous set in the semantic codebook; Determine the minimum group Euclidean distance among the group Euclidean distances; Determine a third synonymous set corresponding to the minimum group Euclidean distance, and determine the synonymous set number of the third synonymous set as the target synonymous set number.

9. The method according to claim 5, wherein The decoding criterion includes: maximum group correlation metric decoding criterion; The obtaining of the target synonymous set number for representing semantic information by performing signal recovery processing on the signal vector according to a pre-constructed semantic codebook and a preset decoding criterion includes: Determine the group correlation metric between the signal vector and each synonymous set in the semantic codebook; Determine the maximum group correlation metric among the group correlation metrics; Determine a fourth synonymous set corresponding to the maximum group correlation metric, and determine the synonymous set number of the fourth synonymous set as the target synonymous set number.

10. The method according to claim 5, wherein, The decoding criterion includes: minimum group Hamming distance decoding criterion; The obtaining of the target synonymous set number for representing semantic information by performing signal recovery processing on the signal vector according to a pre-constructed semantic codebook and a preset decoding criterion includes: Determine the group Hamming distance between the signal vector and each synonymous set in the semantic codebook; Determine the minimum group Hamming distance among the group Hamming distances; Determine a fifth synonymous set corresponding to the minimum group Hamming distance, and determine the synonymous set number of the fifth synonymous set as the target synonymous set number.

11. The method according to claim 1, wherein, The determining of the target grammar information sequence set corresponding to the target synonymous set number according to the synonymous mapping relationship includes: Determine the selected synonymous set corresponding to the target synonymous set number; Determine the target grammar information sequence corresponding to each grammar codeword in the selected synonymous set according to the synonymous mapping relationship; and Integrate the target grammar information sequences to obtain the target grammar information sequence set.

12. An encoding device for a semantic channel, comprising: A relationship parameter acquisition module, configured to acquire a pre-constructed semantic codebook and a synonymous mapping relationship; A grammar information sequence determination module, configured to determine a grammar information sequence to be transmitted; A grammar codeword determination module, configured to determine a grammar codeword corresponding to the grammar information sequence in the semantic codebook according to the synonymous mapping relationship; A modulation module, configured to modulate the grammar codeword to obtain a signal vector.

13. A decoding device for a semantic channel, comprising: A relationship parameter acquisition module, configured to acquire a pre-constructed semantic codebook and a synonymous mapping relationship; A recovery module, configured to perform signal recovery processing on the signal vector according to the semantic codebook and a preset decoding criterion to obtain a target synonymous set number for representing semantic information; A syntax information sequence set determination module, configured to determine a target syntax information sequence set corresponding to a target synonym set number according to the synonym mapping relationship; and An output module, configured to select a syntax information sequence from the target syntax information sequence set as an output result.

14. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, the method described in claim 1 or 5 is implemented.

15. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the method described in claim 1 or 5.

16. A computer program product, comprising: Computer program instructions, when run on a computer, cause the computer to execute the method described in claim 1 or 5.

Images