Information transmission method and device, electronic equipment and storage medium

Abstract

The present disclosure relates to the technical field of communication, and particularly relates to an information transmission method and device, electronic equipment and storage medium. Specifically, the sending end device performs similarity matching on the extracted semantic information and the semantic library of the first service node. If the matching is successful, the semantic information is mapped into second service information after synchronization and then sent. If the matching is unsuccessful, the semantic information is updated through a cloud server, then mapped into second service information after synchronization, and then sent. The receiving end device performs similarity matching on the second service information and the semantic library of the second service node. If the matching is successful, the second service information is inversely mapped into semantic information after synchronization. If the matching is unsuccessful, the semantic information is updated through the cloud server, then inversely mapped into semantic information after synchronization. The semantic information is recovered to obtain third service information. Thus, cross-modal semantic communication in a smart and simple network is realized, and the accuracy and stability of the semantic communication are ensured.

Description

Technical Field

[0001] This disclosure relates to the field of communication technology, and in particular to information transmission methods, apparatus, electronic devices and storage media. Background Technology

[0002] In the future Internet of Everything, network nodes are trending towards intelligence. This extreme intelligence leads to a rapid expansion of the information space, even resulting in a dimensional disaster. This exacerbates the difficulty of representing the information-carrying space, making it difficult for traditional network service capabilities to match the high-dimensional information space. The excessive amount of data transmitted in communication makes information service systems unable to continuously meet people's complex, diverse, and intelligent information transmission needs. With social development and the progress of the times, communication technology and artificial intelligence are constantly evolving. Intelligent simplified networks are a communication technology that combines artificial intelligence with communication technology to improve communication performance. By using artificial intelligence models to encode, propagate, and decode business information, the amount of data transmitted in communication services can be significantly reduced, greatly improving information transmission efficiency. These models are relatively stable and possess reusability and propagation capabilities. The propagation and reuse of these models will help enhance network intelligence while reducing overhead and resource waste, forming an intelligent simplified network with extremely intelligent nodes and an extremely simplified network structure.

[0003] Traditional syntax communication requires that the decoded information at the receiving end strictly match the encoded information at the sending end, achieving bit-level error-free transmission. However, in semantic communication within intelligent simplified networks, this requirement is not a strict match between the decoded and encoded sequences; it only necessitates that the semantic information recovered by the receiving end matches the semantic information of the sending end. Because semantic communication allows for further compression of transmitted content, it redefines the error requirements for information transmission in the system, increasing fault tolerance. This holds promise for overcoming the transmission bottleneck of classical syntax communication systems and providing new solutions for the evolution of next-generation wireless communication systems. Currently, semantic information research is gradually becoming a hot topic in academia, and data transmission based on semantic information will be a highly competitive key technology.

[0004] However, the successful establishment of a semantic communication system inevitably requires support for the simultaneous transmission of information of multiple data types (such as text, images, and video). This means that the semantic communication system needs to standardize the transmission standards for all types of data, but existing research has not provided an overall plan for the semantic communication transmission process. In this context, synchronizing transmitted data through a semantic library can standardize the semantic encoding and transmission process of various types of data, thereby meeting the needs of cross-modal communication of various data types between different devices. Summary of the Invention

[0005] This disclosure provides a method, apparatus, device, and storage medium for transmitting information in a smart and simplified network, enabling cross-modal communication of various data types in the smart and simplified network.

[0006] According to one aspect of this disclosure, an information transmission method is provided, comprising:

[0007] The sending device acquires the first service information;

[0008] Extract semantic information from the first business information;

[0009] The semantic information is matched with the semantic library in the first service node for similarity, and it is determined whether the similarity meets the preset standard.

[0010] If so, the semantic information is synchronized using the semantic library of the first service node, and then mapped and converted into second business information before being sent to the receiving device.

[0011] If not, the semantic library of the first service node is updated through the cloud server, and the semantic information is synchronized using the updated semantic library of the first service node, then mapped and converted into the second business information and sent to the receiving device.

[0012] The receiving device receives the second service information;

[0013] The second business information is matched with the semantic library in the second service node for similarity, and it is determined whether the similarity meets the preset standard.

[0014] If so, then the second business information is synchronized using the semantic library of the second service node, and then reverse-mapped into the semantic information;

[0015] If not, the semantic library of the second service node is updated through the cloud server, and the second business information is synchronized using the updated semantic library of the second service node, and then the semantic information is obtained by reverse mapping.

[0016] The semantic information is restored to obtain the third business information.

[0017] Optionally, the semantic information is a semantic primitive obtained by extracting the first business information, and the data type includes triples, multi-tuples, or graph structures; the second business information is a semantic mapping symbol obtained by encoding the semantic information.

[0018] Optionally, it also includes: in response to the first service node updating the semantic information through the cloud server, the cloud server determines whether the semantic information required for the first service node to update exists in the cloud semantic library. If so, it sends the standard semantic information and standard second business information to the first service node for updating. Otherwise, it adds the standard semantic information to the cloud semantic library according to the needs of the sending device, and encodes and generates the corresponding standard second business information in the cloud server before sending them together to the first service node.

[0019] In response to the second service node updating the second service information through the cloud server, the cloud server sends the standard semantic information and the standard second service information to the second service node for updating.

[0020] Optionally, the first service node includes multiple first edge servers, and the second service node includes multiple second edge servers; the cloud server divides the cloud semantic library into multiple sub-semantic libraries according to the application scenario, and assigns the multiple sub-semantic libraries to the corresponding first edge servers and second edge servers.

[0021] Optionally, before the second service information is matched with the semantic library in the second service node, the receiving device further includes: physically denoising the received second service information and performing error correction processing on the second service information.

[0022] Optionally, the data type of the first business information includes text and / or images and / or videos and / or audio.

[0023] According to another aspect of this disclosure, an information transmission apparatus is provided, including a transmitting end device and a receiving end device, wherein the transmitting end device includes:

[0024] The perception module is used to acquire the first business information;

[0025] The information extraction module is used to extract semantic information from the first business information;

[0026] The first matching module is used to perform similarity matching between the semantic information and the semantic library in the first service node, and to determine whether the similarity meets the preset standard.

[0027] The information mapping module synchronizes the semantic information using the semantic library of the first service node, then maps and converts it into second business information before sending it to the receiving device.

[0028] The first update module is used to update the semantic library of the first service node through the cloud server;

[0029] The receiving device includes:

[0030] The information receiving module is used to receive the second service information;

[0031] The second matching module is used to perform similarity matching between the second business information and the semantic library in the second service node, and to determine whether the similarity meets the preset standard.

[0032] The information inverse mapping module synchronizes the second business information using the semantic library of the second service node and then inverse maps the semantic information to obtain the semantic information.

[0033] The second update module is used to update the semantic library of the second service node through the cloud server;

[0034] The information processing module is used to recover the semantic information to obtain the third business information.

[0035] Optionally, the first service node includes multiple first edge servers, and the second service node includes multiple second edge servers; the cloud server divides the cloud semantic library into multiple sub-semantic libraries according to the application scenario, and assigns the multiple sub-semantic libraries to the corresponding first edge servers and second edge servers.

[0036] Optionally, the receiving device further includes a noise reduction and error correction module, used to physically denoise the received second service information and perform error correction processing on the second service information.

[0037] This disclosure also provides an electronic device, including:

[0038] At least one processor; and

[0039] A memory communicatively connected to the at least one processor; wherein,

[0040] The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the information transmission method described in any of the above technical solutions.

[0041] This disclosure also provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to perform the information transmission method according to any one of the above embodiments.

[0042] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the information transmission method according to any one of the above embodiments.

[0043] The information transmission method, apparatus, device, and storage medium disclosed herein can provide a unified transmission standard for various data types, realize cross-modal semantic communication in intelligent and simplified networks, and ensure the accuracy and stability of semantic communication by performing similarity matching with the semantic library of the semantic library.

[0044] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0045] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0046] Figure 1 This is a flowchart of the information transmission method in an embodiment of this disclosure;

[0047] Figure 2 This is a flowchart illustrating the initialization phase of a semantic communication system in an embodiment of this disclosure.

[0048] Figure 3 This is a flowchart illustrating the communication phase of the semantic communication system in this embodiment of the present disclosure.

[0049] Figure 4 This is a schematic block diagram of the information transmission device in the embodiments of this disclosure;

[0050] Figure 5 This is a semantic communication system architecture diagram in an embodiment of this disclosure. Detailed Implementation

[0051] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0052] This disclosure primarily provides a method and apparatus for multimodal information transmission in intelligent and simplified networks. In intelligent and simplified networks, service information is mainly transmitted through artificial intelligence models. By using the artificial intelligence model, the first service information to be transmitted is compressed into second service information related to the artificial intelligence model, significantly reducing the data communication volume in the network, with compression efficiency far exceeding traditional compression algorithms. Specifically, a sending device uses a pre-configured first model to extract the first service information and obtain the second service information to be transmitted; the sending device transmits the second service information to a receiving device. The receiving device receives the second service information and uses a pre-configured second model to recover the second service information to obtain third service information. The third service information recovered by the second model has slight quality differences compared to the original first service information, but the content is identical, providing a virtually indistinguishable user experience. Before the sending device transmits the second service information to the receiving device, the method further includes: an update module determining whether the receiving device needs to update the second model, and if an update is required, transmitting a pre-configured third model to the receiving device, which then uses the third model to update the second model. By processing business information using pre-trained artificial intelligence models, the amount of data transmitted in communication services can be significantly reduced, greatly improving information transmission efficiency. These models are relatively stable and possess reusability and propagation capabilities. The propagation and reuse of these models will help enhance network intelligence while reducing overhead and resource waste. The model can be divided into several model slices according to different segmentation rules. These model slices can also be transmitted between different network nodes, and the model slices can be assembled into a model. Model slices can be distributed and stored across multiple network nodes. When a network node discovers that it is missing or needs to update a certain model or model slice, it can request it from nearby nodes that may have that slice.

[0053] The transmission of the business information and the model both occur within a communication network, based on network protocols. The network nodes traversed along the paths for transmitting the business information and the model include intelligent routers. The functions of intelligent routers include, but are not limited to, business information transmission, model transmission, absorbing model self-updates, and security protection. The transmission function of intelligent routers involves transmitting business information or models from source nodes to destination nodes, with multiple paths existing between them. The model transmission function of intelligent routers can transmit model slices, improving the model transmission rate by rationally arranging model slices to travel along multiple paths and transmitting them via multiple paths.

[0054] This disclosure provides an information transmission method, such as Figure 1 As shown, it includes:

[0055] Step S101: The sending device obtains the first service information;

[0056] Step S102: Extract semantic information from the first business information;

[0057] Step S103: Perform similarity matching between the semantic information and the semantic library in the first service node, and determine whether the similarity meets the preset standard;

[0058] If so, proceed to step S104a, synchronize the semantic information using the semantic library of the first service node, and then reverse map it into semantic information.

[0059] If not, proceed to step S104b, update the semantic library of the first service node through the cloud server (update the semantic information and its second business information in the semantic library), and after synchronizing the semantic information using the updated semantic library of the first service node, map the updated semantic information into the second business information and send it to the receiving device.

[0060] Step S105: The receiving device receives the second service information;

[0061] Step S106: Perform similarity matching between the second business information and the semantic library in the second service node, and determine whether the similarity meets the preset standard;

[0062] If so, proceed to step S107a, whereby the second service node's semantic library is used to synchronize the second business information, and then the information is reverse-mapped into semantic information.

[0063] If not, proceed to step S107b, update the semantic library of the second service node through the cloud server (update the semantic information and its second business information in the semantic library), and use the updated semantic library of the second service node to synchronize the second business information, and then perform reverse mapping to obtain the semantic information.

[0064] Step S108: Perform semantic information recovery processing to obtain third business information.

[0065] Specifically, in this embodiment, the sending device can utilize a neural network model for information extraction. The extracted semantic information can be in various forms, such as triples, tuples, or graph structures. The semantic information in the semantic database can be stored in the form of a knowledge graph, and triples are the basic building blocks of a knowledge graph. The representation of a triple is entity-relationship-entity, which conforms to the vertex-edge-vertex relationship in a graph in mathematics, facilitating storage using a graph database. Therefore, triples are preferred in this embodiment. The purpose of semantic information extraction is to extract the key information. Semantic information extraction does not necessarily have to be in the form of triples. Depending on different needs, appropriate attributes can be added to extract N-tuples. For example, when extracting quadruples, a category attribute can be added to the entity-relationship-entity structure, becoming entity-relationship-entity-category, which facilitates the classification, recognition, and storage of triple knowledge in the knowledge graph.

[0066] Furthermore, before mapping and transmitting the extracted triples, they are sent to the first service node for semantic similarity matching. If the semantic information exists in the semantic library of the first service node, it meets the preset standard. After successful semantic matching, in order to further save transmission resources, the triples are compressed and mapped to the corresponding semantic symbols before transmission. This process can be regarded as semantic channel joint coding, and the output semantic mapping symbols are (A; B; C). If they do not exist, they do not meet the standard. The first service node needs to update the knowledge graph in the semantic library through the cloud server, add the required triples and semantic mapping symbols, and use the updated semantic information to map to obtain the corresponding semantic mapping symbols, i.e., the second business information. Similarly, after receiving the second service information, the receiving device also needs to send it to the second service node for semantic similarity matching. If the semantic library of the second service node contains the received semantic mapping symbol, it is determined that it meets the preset standard. The second service information is synchronized with the second service information matched in the semantic library and then reverse-mapped to obtain the corresponding semantic information. The semantic information is then restored to the third service information that is highly similar to the first service information. If it does not exist, it does not meet the standard. The second service node needs to update the knowledge graph in the semantic library through the cloud server, add the required triples and semantic mapping symbols, and use the updated semantic mapping symbols to perform reverse mapping to obtain the corresponding semantic information. The semantic information is then restored to the third service information that is highly similar to the first service information.

[0067] It should be noted that, in this embodiment, the synchronization processing of semantic information or second business information refers to replacing the transmitted semantic information or second business information with the standard semantic information or second business information stored in the semantic library after successful matching, in order to reduce semantic errors in the semantic recovery process of the receiving device. The purpose of the replacement is to synchronize the transmitted semantic information. Since the decoding module may have limited decoding capabilities, it is more suitable for standardized semantic information. For example, "teacher" and "teacher" are two different expressions of the same semantic meaning. Their semantic mapping symbols may be very different, and they can be synchronized with the semantic library to replace them with "teacher", retaining only one expression. Therefore, synchronizing semantic information or second business information (semantic mapping symbols) at the sending and receiving ends can reduce semantic errors that may occur during transmission or decoding. On the other hand, due to the randomness of data and the diversity of knowledge extraction models, different expressions of the same content will not generate completely identical semantic information at the sending end. Even if the input data is completely identical, the generated semantic information may have slight differences due to various reasons. If all semantic information with similar meanings but different expressions were stored in the semantic library, it would be a huge waste of storage resources and subsequent computing resources. Therefore, the semantic matching and synchronization in this embodiment can keep only one expression in the semantic library, which can also effectively save storage and computing resources.

[0068] The aforementioned cloud server comprises two parts: a cloud semantic library and a semantic mapping model. The cloud semantic library contains both semantic information and corresponding secondary business information. When adding standard semantic information, the semantic mapping model generates corresponding semantic mapping symbols based on that information. When updating the semantic library of the first or second service node, the cloud semantic library synchronously updates the semantic information and corresponding semantic mapping symbols, ensuring consistency between the semantic information and the semantic mapping symbols. For example, if the sending device needs to send the semantic information "metaverse," but the first service node's semantic library does not find a matching semantic information (i.e., no word expressing the same meaning), the cloud server adds the semantic information "metaverse." At this point, the cloud semantic library does not yet have the corresponding semantic mapping symbol for "metaverse." If the cloud does not update synchronously, the receiving device may not be able to recognize the semantic mapping symbol and therefore cannot reverse map the semantic information "metaverse." If the cloud synchronously updates the semantic mapping symbol corresponding to "metaverse," the receiving device can update the semantic information and secondary business information (i.e., semantic mapping symbols) in the second semantic library through the cloud server, ensuring consistency between the semantic information and semantic mapping symbols between the sending and receiving ends.

[0069] Through the above technical solution, in this embodiment, the first service node or the second service node is selected first for similarity matching. If the first service node or the second service node fails to match, the cloud server is used for matching. This reduces the latency of the matching process and the load on the cloud server.

[0070] As an optional implementation, the first service node includes multiple first edge servers, i.e., edge servers at the sending end, and the second service node includes multiple second edge servers, i.e., edge servers at the receiving end. For example... Figure 2 As shown, the cloud server can divide its semantic library (i.e., the cloud semantic library) into multiple sub-semantic libraries according to different application scenarios or other classification methods, and send them to the corresponding edge servers. In other words, the cloud semantic library stores the complete semantic library, while each edge server stores only a portion of the semantic library. The sending device can send the semantic information to the corresponding edge server for similarity matching based on the application scenario. In addition, multiple edge servers can communicate collaboratively to achieve load balancing, which can reduce matching time and improve communication efficiency.

[0071] For example, based on the different types of information transmission, message types can be roughly divided into four categories: text, image, voice, and video. Since data transmission uses a semantic library as the semantic standard, the sending device needs to extract semantic information from the initial business information to convert various data types into standard triplet types, thereby achieving information standardization. For example... Figure 3 As shown, in this embodiment, it is assumed that the text information sent by the sending device is "The power supply voltage of device X is too low". When using text information as input parameters, the sending device uses a neural network model to extract information, performs natural language processing on the text information, extracts its semantic entities, inter-entity relationships, and entity attributes, and then constructs triples as output, for example, the output triple is (X; voltage; low). Similar to the sending device, the receiving device also supports semantic information restoration for multiple data types. The receiving device still needs to perform semantic library similarity matching when receiving the second service information. After a successful match, the second service information is reverse-mapped and restored to text information. The artificial intelligence model is used to perform semantic restoration based on the semantic entities, inter-entity relationships, and entity attributes in the triples, that is, natural language generation based on the triples, and then restored to text information as the model output. Assuming that it can be restored correctly, the third service information can be "The power supply voltage of device X is low", which is slightly different from the text of the first service information, but the meaning of the information to be transmitted is almost the same, and the semantic information is successfully transmitted.

[0072] As an optional implementation, in response to the first service node updating semantic information through the cloud server, the cloud server determines whether the semantic information required for the first service node to update exists in the cloud semantic library. If so, the standard semantic information and the corresponding standard second business information are sent to the first service node for updating. Otherwise, the standard semantic information required by the sending device is added to the cloud semantic library, and the corresponding standard second business information is encoded in the cloud server and then sent to the first service node.

[0073] In response to the second service node updating the second business information through the cloud server, since the cloud will inevitably contain the required standard semantic information and standard second business information after the first service node requests it, the cloud server will directly send the required semantic information and its second business information to the second service node for updating.

[0074] Specifically, in this embodiment, there are two scenarios. When the edge server needs to update through the cloud server, the cloud server first matches the semantic information to be sent (e.g., the extracted triples) with the knowledge graph of the cloud semantic library. If the cloud semantic library contains the required triples, it directly sends the triples and their semantic mapping symbols to the edge server for updating. If the cloud semantic library does not contain the required triples, it adds the triples to the cloud semantic library, updates the knowledge graph of the cloud semantic library, and generates its corresponding semantic mapping symbols, thereby continuously improving the cloud semantic library and ensuring its completeness. After the semantic library is updated, the edge server will continue to execute the semantic symbol mapping process.

[0075] As an optional implementation, before the second service information is matched with the semantic library in the second service node, the receiving device further includes: physically denoising the received second service information and performing error correction processing on the second service information, thereby ensuring the accuracy of the second service information.

[0076] This disclosure also provides an information transmission device, such as Figure 4 As shown, it includes a transmitting device 1 and a receiving device 2, wherein the transmitting device includes:

[0077] The sensing module 101 is used to acquire the first business information;

[0078] Information extraction module 102 is used to extract semantic information from the first business information;

[0079] The first matching module 103 is used to perform similarity matching between semantic information and the semantic library in the first service node, and to determine whether the similarity meets the preset standard.

[0080] The information mapping module 104 synchronizes the semantic information using the semantic library of the first service node, and then maps and converts it into second business information before sending it to the receiving device.

[0081] The first update module 105 is used to update the semantic library of the first service node through the cloud server, that is, to update the semantic information and its second business information in the semantic library. The information mapping module 104 then uses the updated semantic library of the first service node to synchronize the semantic information, and maps and converts it into the second business information to be sent to the receiving device.

[0082] Receiver device 2 includes:

[0083] Information receiving module 201 is used to receive second service information;

[0084] The second matching module 202 is used to perform similarity matching between the second business information and the semantic library in the second service node, and to determine whether the similarity meets the preset standard.

[0085] The information inverse mapping module 203 synchronizes the second business information using the semantic library of the second service node and then inverse maps the semantic information to obtain semantic information.

[0086] The second update module 204 is used to update the semantic library of the second service node through the cloud server, that is, to update the semantic information and the second business information in the semantic library. The information inverse mapping module 203 uses the updated semantic library of the second service node to synchronize the second business information and obtain the semantic information through inverse mapping.

[0087] The information processing module 205 is used to recover semantic information to obtain third business information.

[0088] Specifically, such as Figure 5As shown, the semantic communication system in this embodiment comprises a sending device 1, a first service node 3 (which may be an edge server of the sending end), a receiving device 2, a second service node 4 (which may be an edge server of the receiving end), and a cloud server 5. The cloud server connects to both the edge servers of the sending and receiving ends. To reduce latency in the matching process, similarity matching is preferentially performed on the edge servers; matching is only performed through the cloud server when the edge servers cannot match. The perception module 101 is used to collect first business information. The data types of the first business information include, but are not limited to, one or more of text, images, videos, and audio. The information extraction module 102 can use a neural network model to extract information. The extracted semantic information can be in various forms such as triples, tuples, or graph structures. The semantic information in the semantic database can be stored in the form of a knowledge graph. Triples are the basic building blocks of a knowledge graph, and their representation is entity-relationship-entity, conforming to the point-edge-point relationship in a graph in mathematics, which is convenient for storage using a graph database. Therefore, the triple form is preferred in this embodiment. The purpose of semantic information extraction is to extract the key information. Semantic information extraction does not necessarily have to extract triples. Depending on different needs, appropriate attributes can be added to extract N-tuples. For example, when extracting quadruples, a category attribute can be added to the entity-relation-entity, becoming entity-relation-entity-category, which facilitates the classification, recognition and storage of triple knowledge in knowledge graphs.

[0089] Furthermore, before mapping and transmitting the extracted triples, the first matching module 103 performs similarity matching with the semantic library of the first service node. If the semantic information exists in the semantic library of the first service node, it meets the preset standard. After successful semantic matching, in order to further save transmission resources, the information mapping module 104 compresses the triples and maps them into corresponding semantic symbols before transmission. This process can be regarded as semantic channel joint coding, and the output semantic mapping symbols are (A; B; C). If they do not exist, they do not meet the standard. The first update module 105 needs to update the knowledge graph in the semantic library through the cloud server, add the required triples and semantic mapping symbols, and the information mapping module 104 uses the updated standard semantic information to map and obtain the corresponding semantic mapping symbols, i.e., the second business information. Similarly, after receiving the second service information, the information receiving module 201 of the receiving device also needs to use the second matching module 202 to perform similarity matching between the second service information and the semantic library of the second service node. If the semantic library of the second service node contains the received semantic mapping symbol, it is determined that it meets the preset standard. The information inverse mapping module 203 synchronizes the second service information with the second service information matched in the semantic library and then performs inverse mapping to obtain the corresponding semantic information. The information processing module 205 then restores the semantic information to the third service information that is highly similar to the first service information. If it does not exist, it does not meet the standard. The second update module 204 needs to update the knowledge graph in the semantic library through the cloud server, add the required triples and their semantic mapping symbols, perform inverse mapping using the updated semantic mapping symbols to obtain the corresponding semantic information, and then restore the semantic information to the third service information that is highly similar to the first service information.

[0090] It should be noted that, in this embodiment, synchronizing semantic information or second business information means replacing the transmitted semantic information or second business information with the standard semantic information or second business information stored in the semantic library after successful matching, in order to reduce semantic errors in the semantic recovery process of the receiving device. The purpose of replacement is to synchronize the transmitted semantic information. Since the decoding module may have limited decoding capabilities, it is more suitable for standardized semantic information. For example, "teacher" and "teacher" are two different expressions of the same semantic meaning. Their semantic mapping symbols may be very different, and they can be synchronized with the semantic library to replace them with "teacher", retaining only one expression. Therefore, synchronizing semantic information or second business information (semantic mapping symbols) at the sending and receiving ends can reduce semantic errors that may occur during transmission or decoding. On the other hand, due to the randomness of data and the diversity of knowledge extraction models, different expressions of the same content will not generate completely identical semantic information at the sending end. Even if the input data is completely identical, the generated semantic information may have slight differences due to various reasons. If all semantic information with similar meanings but different expressions were stored in the semantic library, it would be a huge waste of storage resources and subsequent computing resources. Therefore, the semantic matching and synchronization in this embodiment can keep only one expression in the semantic library, which can also effectively save storage and computing resources.

[0091] The aforementioned cloud server comprises two parts: a cloud semantic library and a semantic mapping model. The cloud semantic library contains both semantic information and corresponding secondary business information. When adding standard semantic information, the semantic mapping model generates corresponding semantic mapping symbols based on that information. When updating the semantic library of the first or second service node, the cloud semantic library synchronously updates the semantic information and corresponding semantic mapping symbols, ensuring consistency between the semantic information and the semantic mapping symbols. For example, if the sending device needs to send the semantic information "metaverse," but the first service node's semantic library does not find a matching semantic information (i.e., no word expressing the same meaning), the cloud server adds the semantic information "metaverse." At this point, the cloud semantic library does not yet have the corresponding semantic mapping symbol for "metaverse." If the cloud does not update synchronously, the receiving device may not be able to recognize the semantic mapping symbol and therefore cannot reverse map the semantic information "metaverse." If the cloud synchronously updates the semantic mapping symbol corresponding to "metaverse," the receiving device can update the semantic information and secondary business information (i.e., semantic mapping symbols) in the second semantic library through the cloud server, ensuring consistency between the semantic information and semantic mapping symbols between the sending and receiving ends.

[0092] As an optional implementation, the first service node may include multiple first edge servers, i.e., edge servers at the sending end, and the second service node may include multiple second edge servers, i.e., edge servers at the receiving end. For example... Figure 2 As shown, the cloud server can divide its semantic library (i.e., the cloud semantic library) into multiple sub-semantic libraries according to different application scenarios or other classification methods, and send them to the corresponding edge servers. In other words, the cloud semantic library stores the complete semantic library, while each edge server stores only a portion of the semantic library. The sending device can send the semantic information to the corresponding edge server for similarity matching based on the application scenario. In addition, multiple edge servers can communicate collaboratively to achieve load balancing, which can reduce matching time and improve communication efficiency.

[0093] For example, based on the different types of information transmission, message types can be roughly divided into four categories: text, image, voice, and video. Since data transmission uses a semantic library as the semantic standard, the sending device needs to extract semantic information from the initial business information to convert various data types into standard triplet types, thereby achieving information standardization. For example... Figure 3 As shown, in this embodiment, it is assumed that the text information sent by the sending device is "The power supply voltage of device X is too low". When using text information as input parameters, the sending device uses a neural network model to extract information, performs natural language processing on the text information, extracts its semantic entities, inter-entity relationships, and entity attributes, and then constructs triples as output, for example, the output triple is (X; voltage; low). Similar to the sending device, the receiving device also supports semantic information restoration for multiple data types. The receiving device still needs to perform semantic library similarity matching when receiving the second service information. After a successful match, the second service information is reverse-mapped and restored to text information. The artificial intelligence model is used to perform semantic restoration based on the semantic entities, inter-entity relationships, and entity attributes in the triples, that is, natural language generation based on the triples, and then restored to text information as the model output. Assuming that it can be restored correctly, the third service information can be "The power supply voltage of device X is low", which is slightly different from the text of the first service information, but the meaning of the information to be transmitted is almost the same, and the semantic information is successfully transmitted. To meet the management needs in industrial scenarios, third-party business information can also be converted into voice alarm information for broadcast, realizing text-to-speech modal conversion.

[0094] As an optional implementation, in response to the first service node updating semantic information through the cloud server, the cloud server determines whether the semantic information required for the first service node to update exists in the cloud semantic library. If so, the standard semantic information and the corresponding standard second business information (i.e., semantic mapping symbols) are sent to the first service node for updating. Otherwise, the required standard semantic information is added to the cloud semantic library, and the corresponding standard second business information is encoded in the cloud server and then sent to the first service node.

[0095] In response to the second service node updating the second business information through the cloud server, since the cloud will inevitably contain the required standard semantic information and standard second business information after the first service node requests it, the cloud server will directly send the required semantic information and its second business information to the second service node for updating.

[0096] Specifically, in this embodiment, there are two scenarios. When the edge server needs to update through the cloud server, the cloud server first matches the semantic information to be sent (e.g., the extracted triples) with the knowledge graph of the cloud semantic library. If the cloud semantic library contains the required triples, it directly sends the triples and their semantic mapping symbols to the edge server for updating. If the cloud semantic library does not contain the required triples, it adds the triples to the cloud semantic library, updates the knowledge graph of the cloud semantic library, and generates its corresponding semantic mapping symbols, thereby continuously improving the cloud semantic library and ensuring its completeness. After the semantic library is updated, the edge server will continue to execute the semantic symbol mapping process.

[0097] As an optional implementation, the receiving device further includes a denoising and error correction module 206, which performs physical denoising on the received second service information and error correction on the second service information before performing similarity matching between the second service information and the semantic library in the second service node, thereby ensuring the accuracy of the second service information.

[0098] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0099] Specifically, electronic devices are intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0100] The device includes a computing unit that can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) or loaded from a storage unit into random access memory (RAM). The RAM can also store various programs and data required for device operation. The computing unit, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0101] Multiple components in the device are connected to the I / O interface, including: input units such as keyboards and mice; output units such as various types of displays and speakers; storage units such as disks and optical discs; and communication units such as network interface cards (NICs), modems, and wireless transceivers. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0102] The computing unit can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of computing units include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit performs the various methods and processes described above, such as the information transmission methods in the above embodiments. For example, in some embodiments, the information transmission method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the computing unit, one or more steps of the information transmission method described above may be performed. Alternatively, in other embodiments, the computing unit may be configured to perform the information transmission method by any other suitable means (e.g., by means of firmware).

[0103] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0104] The program code used to implement the information transmission method of this disclosure can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0105] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on 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 of the foregoing.

[0106] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0107] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0108] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0109] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0110] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. An information transmission method, characterized in that, include: The sending device acquires the first service information; Extract semantic information from the first business information; The semantic information is matched with the semantic library in the first service node for similarity, and it is determined whether the similarity meets the preset standard. If so, the semantic information is synchronized using the semantic library of the first service node, and then mapped and converted into second business information before being sent to the receiving device. If not, the semantic library of the first service node is updated through the cloud server, and the semantic information is synchronized using the updated semantic library of the first service node, then mapped and converted into the second business information and sent to the receiving device. The receiving device receives the second service information; The second business information is matched with the semantic library in the second service node for similarity, and it is determined whether the similarity meets the preset standard. If so, then the second business information is synchronized using the semantic library of the second service node, and then reverse-mapped into the semantic information; If not, the semantic library of the second service node is updated through the cloud server, and the second business information is synchronized using the updated semantic library of the second service node, and then the semantic information is obtained by reverse mapping. The semantic information is restored to obtain the third business information.

2. The information transmission method according to claim 1, wherein, The semantic information is a semantic primitive obtained by extracting the first business information, and the data type includes triples, multi-tuples, or graph structures; the second business information is a semantic mapping symbol obtained by mapping and transforming the semantic information.

3. The information transmission method according to claim 1, characterized in that, Also includes: In response to the first service node updating the semantic information through the cloud server, the cloud server determines whether the semantic information required for the first service node to update exists in the cloud semantic library. If so, it sends the standard semantic information and standard second business information to the first service node for updating; otherwise, it adds the standard semantic information to the cloud semantic library according to the needs of the sending device, and encodes the corresponding standard second business information in the cloud server before sending it to the first service node. In response to the second service node updating the second service information through the cloud server, the cloud server sends the standard semantic information and the standard second service information to the second service node for updating.

4. The information transmission method according to claim 1, characterized in that, The first service node includes multiple first edge servers, and the second service node includes multiple second edge servers; the cloud server divides the cloud semantic library into multiple sub-semantic libraries according to the application scenario, and assigns the multiple sub-semantic libraries to the corresponding first edge servers and second edge servers.

5. The information transmission method according to claim 1, characterized in that, Before the second service information is matched with the semantic library in the second service node, the receiving device performs physical noise reduction on the received second service information and performs error correction processing on the second service information.

6. The information transmission method according to any one of claims 1-5, characterized in that, The data types of the first business information include text and / or images and / or videos and / or audio.

7. An information transmission device, characterized in that, It includes a transmitting device and a receiving device, wherein the transmitting device includes: The perception module is used to acquire the first business information; The information extraction module is used to extract semantic information from the first business information; The first matching module is used to perform similarity matching between the semantic information and the semantic library in the first service node, and to determine whether the similarity meets the preset standard. The information mapping module synchronizes the semantic information using the semantic library of the first service node, then maps and converts it into second business information before sending it to the receiving device. The first update module is used to update the semantic library of the first service node through the cloud server; The receiving device includes: The information receiving module is used to receive the second service information; The second matching module is used to perform similarity matching between the second business information and the semantic library in the second service node, and to determine whether the similarity meets the preset standard. The information inverse mapping module synchronizes the second business information using the semantic library of the second service node and then inverse maps the semantic information to obtain the semantic information. The second update module is used to update the semantic library of the second service node through the cloud server; The information processing module is used to recover the semantic information to obtain the third business information.

8. The information transmission device according to claim 7, characterized in that, The first service node includes multiple first edge servers, and the second service node includes multiple second edge servers; the cloud server divides the cloud semantic library into multiple sub-semantic libraries according to the application scenario, and assigns the multiple sub-semantic libraries to the corresponding first edge servers and second edge servers.

9. The information transmission device according to claim 7, characterized in that, The receiving device further includes a noise reduction and error correction module, used to physically denoise the received second service information and perform error correction processing on the second service information.

10. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the information transmission method according to any one of claims 1-6.

11. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the information transmission method according to any one of claims 1-6.

12. A computer program product comprising a computer program that, when executed by a processor, implements the information transmission method according to any one of claims 1-6.

Images