Data generation method and apparatus, and related device

Abstract

The present disclosure relates to the technical field of communications. Provided are a data generation method and apparatus, and a related device. The method comprises: receiving a data generation request sent by a second network function; and on the basis of the data generation request, sending first information to a third network function, and / or sending second information to a fourth network function, wherein the first information is used for triggering, by means of a Siamese network, generation of data required by the second network function, and the second information is used for retrieving, from data stored in the fourth network function, the data required by the second network function.

Description

Data generation methods, apparatus and related equipment

[0001] Cross-references to related applications

[0002] This disclosure claims priority to Chinese Patent Application No. 202410818262.2, filed in China on June 24, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This disclosure relates to the field of communication technology, and in particular to a data generation method, apparatus and related equipment. Background Technology

[0004] With the evolution of 5G and the in-depth research of 6G, network intelligence and artificial intelligence (AI) will become important trends in the future evolution of networks. However, the key factor determining and restricting the rapid development of AI is data. The training of AI models and the generation of intelligent strategies cannot be separated from accurate and multi-dimensional data. In particular, as large models gradually empower networks, massive amounts of network data are needed to build large network models capable of accurate reasoning.

[0005] However, on the one hand, due to reasons such as data privacy protection and the lack of openness of non-standard data of manufacturers' equipment, it is difficult to obtain data in certain dimensions; on the other hand, there is little network anomaly data, and it takes a long time to collect a lot of negative sample data, which is difficult to meet the needs of AI training. Therefore, if we want to promote the rapid development of endogenous AI and large network models, data is the first core issue that needs to be addressed, and data generation methods are necessary. Summary of the Invention

[0006] The purpose of this disclosure is to provide a data generation method, apparatus, and related equipment that solves data problems such as data privacy and imbalanced samples.

[0007] To achieve the above objectives, embodiments of this disclosure provide a data generation method, executed by a first network function, comprising:

[0008] Receive data generation requests sent by the second network function;

[0009] Based on the data generation request, send first information to the third network function, and / or send second information to the fourth network function;

[0010] The first information is used to trigger the generation of the data required for the second network function through the twin network, and the second information is used to retrieve the data required for the second network function from the data stored in the fourth network function.

[0011] Optionally, the data generation request includes information indicating at least one of the following:

[0012] Twin scope; AI scenario description; twin granularity; data volume; feature dimension; data threshold; generation duration.

[0013] Optionally, based on the data generation request, sending first information to the third network function includes:

[0014] The first information is obtained by parsing the data to generate the request;

[0015] Send the first information to the third network function;

[0016] The first information includes at least one of the following: AI scene description information; the identifier and topology of the twin network element.

[0017] Optionally, based on the data generation request, a second message is sent to the fourth network function, including:

[0018] The second information is obtained by parsing the data generation request;

[0019] The second information is sent to the fourth network function;

[0020] The second information includes data generation conditions.

[0021] Optionally, based on the data generation request, sending first information to the third network function includes:

[0022] Upon receiving feedback information from the fourth network function, and the feedback information indicating that the fourth network function has not stored the data required by the second network function, the first information is sent to the third network function according to the data generation request.

[0023] Optionally, before receiving the data generation request sent by the second network function, the method further includes:

[0024] Send a registration request to the fifth network function, the registration request including information indicating at least one of the following:

[0025] The identifier of the network element; the scope of the twin; the processing capacity; and the data generation identifier.

[0026] Optionally, after receiving the data generation request sent by the second network function, the method further includes:

[0027] Based on the data generation request, third information is sent to the third network function or the sixth network function, the third information including at least one of the following:

[0028] Data volume; feature dimensions; data threshold; generation duration.

[0029] Optionally, the data required by the second network function is generated by the sixth network function based on the third information to configure the twin network.

[0030] To achieve the above objectives, embodiments of this disclosure provide a data generation method, executed by a third network function, comprising:

[0031] Receive the first information sent by the first network function;

[0032] Generate a twin network based on the first information.

[0033] Optionally, the first information includes at least one of the following: AI scene description information; the identifier and topology of the twin network element.

[0034] Optionally, generating a twin network based on the first information includes:

[0035] If the first information includes AI scene description information, a query is performed based on the AI ​​scene description information to see if a matching general template exists. If a matching general template exists, a Siamese network is generated using the general template; or,

[0036] If the first information includes the identifier and topology of the twin network element, and there is no matching general template, or if the first information does not include AI scene description information, a twin network is generated based on the identifier and topology of the twin network element.

[0037] Optionally, it also includes:

[0038] Receive the third information sent by the first network function;

[0039] Send the third information to the sixth network function;

[0040] The third information includes at least one of the following:

[0041] Data volume; feature dimensions; data threshold; generation duration.

[0042] Optionally, generating a twin network based on the first information includes:

[0043] When multiple pieces of the first information corresponding to different data generation requests are received, a twin network for each piece of the first information is generated in parallel based on the multiple pieces of the first information.

[0044] To achieve the above objectives, embodiments of this disclosure provide a data generation method, executed by a sixth network function, comprising:

[0045] Receive third information sent by the first network function or the third network function;

[0046] Configure the twin network based on the third information to obtain the data required for the second network function;

[0047] The third information includes at least one of the following:

[0048] Data volume; feature dimensions; data threshold; generation duration.

[0049] Optionally, after configuring the twin network based on the third information to obtain the data required for the second network function, the method further includes:

[0050] The data is sent to the second network function.

[0051] Optionally, after configuring the twin network based on the third information to obtain the data required for the second network function, the method further includes:

[0052] The data and the third information are sent to the fourth network function.

[0053] To achieve the above objectives, embodiments of this disclosure provide a data generation method, executed by a fourth network function, comprising:

[0054] Receive data and third-party information sent by the fifth network function;

[0055] The data and the third information are associated and stored;

[0056] The third information includes at least one of the following:

[0057] Data volume; feature dimensions; data threshold; generation duration.

[0058] Optionally, the data and the third information are stored together, including:

[0059] If no data associated with the third information exists in the current stored data, the received data and the third information will be associated and stored together.

[0060] Optionally, it also includes:

[0061] If data associated with the third information exists in the current stored data, the received data will be deleted.

[0062] To achieve the above objectives, embodiments of this disclosure provide a data generation method, executed by a second network function, comprising:

[0063] Send a data generation request to the first network function, the data generation request being used to request data required by the second network function;

[0064] Receive data sent by the fourth or sixth network function.

[0065] Optionally, the data generation request includes information indicating at least one of the following:

[0066] Twin scope; AI scenario description; twin granularity; data volume; feature dimension; data threshold; generation duration.

[0067] Optionally, before sending the data generation request to the first network function, the method further includes:

[0068] Send a discovery request to the fifth network function;

[0069] Receive the target address information fed back by the fifth network function;

[0070] The discovery request includes information indicating at least one of the following:

[0071] Twin scope; processing capacity; data generation identifier.

[0072] To achieve the above objectives, embodiments of this disclosure provide a data generation method, executed by a fifth network function, comprising:

[0073] Receive a registration request sent by a first network function, the registration request including information indicating at least one of the following:

[0074] The identifier of the network element; the scope of the twin; the processing capacity; and the data generation identifier.

[0075] Optionally, it also includes:

[0076] Receive discovery requests sent by the second network function;

[0077] Based on the discovery request, determine the first matching network function;

[0078] Send the target address information of the first network function to the second network function;

[0079] The discovery request includes information indicating at least one of the following:

[0080] Twin scope; processing capacity; data generation identifier.

[0081] To achieve the above objectives, embodiments of this disclosure provide a data generation apparatus applied to a first network function, comprising:

[0082] The first receiving module is used to receive data generation requests sent by the second network function;

[0083] The first sending module is used to generate a request based on the data, send first information to the third network function, and / or send second information to the fourth network function;

[0084] The first information is used to trigger the generation of the data required for the second network function through the twin network, and the second information is used to retrieve the data required for the second network function from the data stored in the fourth network function.

[0085] Optionally, the data generation request includes information indicating at least one of the following:

[0086] Twin scope; AI scenario description; twin granularity; data volume; feature dimension; data threshold; generation duration.

[0087] Optionally, the first sending module is further configured to:

[0088] The first information is obtained by parsing the data to generate the request;

[0089] Send the first information to the third network function;

[0090] The first information includes at least one of the following: AI scene description information; the identifier and topology of the twin network element.

[0091] Optionally, the first sending module is further configured to:

[0092] The second information is obtained by parsing the data generation request;

[0093] The second information is sent to the fourth network function;

[0094] The second information includes data generation conditions.

[0095] Optionally, the first sending module is further configured to:

[0096] Upon receiving feedback information from the fourth network function, and the feedback information indicating that the fourth network function has not stored the data required by the second network function, the first information is sent to the third network function according to the data generation request.

[0097] Optionally, the device further includes:

[0098] The registration module is used to send a registration request to the fifth network function, the registration request including information indicating at least one of the following:

[0099] The identifier of the network element; the scope of the twin; the processing capacity; and the data generation identifier.

[0100] Optionally, the device further includes:

[0101] The third sending module is configured to generate a request based on the data and send third information to the third network function or the sixth network function, wherein the third information includes at least one of the following:

[0102] Data volume; feature dimensions; data threshold; generation duration.

[0103] Optionally, the data required by the second network function is generated by the sixth network function based on the third information to configure the twin network.

[0104] To achieve the above objectives, embodiments of this disclosure provide a data generation apparatus applied to a third network function, comprising:

[0105] The second receiving module is used to receive the first information sent by the first network function;

[0106] The first processing module is used to generate a twin network based on the first information.

[0107] Optionally, the first information includes at least one of the following: AI scene description information; the identifier and topology of the twin network element.

[0108] Optionally, the first processing module is further configured to:

[0109] If the first information includes AI scene description information, a query is performed based on the AI ​​scene description information to see if a matching general template exists. If a matching general template exists, a Siamese network is generated using the general template; or,

[0110] If the first information includes the identifier and topology of the twin network element, and there is no matching general template, or if the first information does not include AI scene description information, a twin network is generated based on the identifier and topology of the twin network element.

[0111] Optionally, the device further includes:

[0112] The seventh receiving module is used to receive the third information sent by the first network function;

[0113] The fourth sending module is used to send the third information to the sixth network function;

[0114] The third information includes at least one of the following:

[0115] Data volume; feature dimensions; data threshold; generation duration.

[0116] Optionally, the first processing module is further configured to:

[0117] When multiple pieces of the first information corresponding to different data generation requests are received, a twin network for each piece of the first information is generated in parallel based on the multiple pieces of the first information.

[0118] To achieve the above objectives, embodiments of this disclosure provide a data generation apparatus applied to a sixth network function, comprising:

[0119] The third receiving module is used to receive third information sent by the first network function or the third network function;

[0120] The second processing module is used to configure the twin network according to the third information and obtain the data required for the second network function.

[0121] The third information includes at least one of the following:

[0122] Data volume; feature dimensions; data threshold; generation duration.

[0123] Optionally, the device further includes:

[0124] The fifth sending module is used to send the data to the second network function.

[0125] Optionally, the device further includes:

[0126] The sixth sending module is used to send the data and the third information to the fourth network function.

[0127] To achieve the above objectives, embodiments of this disclosure provide a data generation apparatus applied to a fourth network function, including:

[0128] The fourth receiving module is used to receive data and third information sent by the fifth network function;

[0129] A storage module is used to associate and store the data and the third information.

[0130] The third information includes at least one of the following:

[0131] Data volume; feature dimensions; data threshold; generation duration.

[0132] Optionally, the storage module is further configured to:

[0133] If no data associated with the third information exists in the current stored data, the received data and the third information will be associated and stored together.

[0134] Optionally, the device further includes:

[0135] The third processing module is used to delete the received data if data associated with the third information exists in the current stored data.

[0136] To achieve the above objectives, embodiments of this disclosure provide a data generation apparatus applied to a second network function, comprising:

[0137] The second sending module is used to send a data generation request to the first network function, the data generation request being used to request data required by the second network function;

[0138] The fifth receiving module is used to receive data sent by the fourth or sixth network function.

[0139] Optionally, the data generation request includes information indicating at least one of the following:

[0140] Twin scope; AI scenario description; twin granularity; data volume; feature dimension; data threshold; generation duration.

[0141] Optionally, the device further includes:

[0142] The discovery module is used to send discovery requests to the fifth network function.

[0143] The eighth receiving module is used to receive the target address information fed back by the fifth network function;

[0144] The discovery request includes information indicating at least one of the following:

[0145] Twin scope; processing capacity; data generation identifier.

[0146] To achieve the above objectives, embodiments of this disclosure provide a data generation apparatus applied to a fifth network function, comprising:

[0147] The sixth receiving module is configured to receive a registration request sent by the first network function, the registration request including information indicating at least one of the following:

[0148] The identifier of the network element; the scope of the twin; the processing capacity; and the data generation identifier.

[0149] Optionally, the device further includes:

[0150] The ninth receiving module is used to receive discovery requests sent by the second network function;

[0151] The fourth processing module is used to determine the first matching network function based on the discovery request;

[0152] The sixth sending module is used to send the target address information of the first network function to the second network function;

[0153] The discovery request includes information indicating at least one of the following:

[0154] Twin scope; processing capacity; data generation identifier.

[0155] To achieve the above objectives, embodiments of this disclosure provide a network element, including a first network function that performs the data generation method described above, a third network function that performs the data generation method described above, a sixth network function that performs the data generation method described above, and a fourth network function that performs the data generation method described above.

[0156] Optionally, a first interface is provided between the first network function and the third network function, a second interface is provided between the third network function and the sixth network function, a third interface is provided between the first network function and the sixth network function, a fourth interface is provided between the first network function and the fourth network function, and a fifth interface is provided between the sixth network function and the fourth network function.

[0157] To achieve the above objectives, embodiments of this disclosure provide a data generation system, including: a network element as described above, a second network function executing the data generation method as described above, and a fifth network function executing the data generation method as described above.

[0158] To achieve the above objectives, embodiments of this disclosure provide a network device, including: a transceiver, a processor, a memory, and a program or instructions stored in the memory and executable on the processor; wherein, when the processor executes the program or instructions, it implements the data generation method described above.

[0159] To achieve the above objectives, embodiments of this disclosure provide a readable storage medium having a program or instructions stored thereon, which, when executed by a processor, implement the steps in the data generation method described above.

[0160] To achieve the above objectives, embodiments of this disclosure provide a computer program product, including computer instructions that, when executed by a processor, implement the steps of the data generation method described above.

[0161] The beneficial effects of the above-mentioned technical solution disclosed herein are as follows:

[0162] The method of this disclosure, by receiving a data generation request from a second network function, and after understanding the data requirements of the second network function, sends first information to a third network function according to the data generation request to trigger the twin network to generate the required data; and / or sends second information to a fourth network function to retrieve the required data from the data stored in the fourth network function, thus solving data problems such as data privacy and imbalanced samples. Attached Figure Description

[0163] Figure 1 is a flowchart of one of the data generation methods according to an embodiment of this disclosure;

[0164] Figure 2 is a schematic diagram of the structure of a network digital twin element according to an embodiment of this disclosure;

[0165] Figure 3 is a flowchart of the startup and orchestration of the third network function in an embodiment of this disclosure;

[0166] Figure 4 is a schematic diagram illustrating the application of the method according to an embodiment of this disclosure;

[0167] Figure 5 is a second flowchart of the data generation method according to an embodiment of this disclosure;

[0168] Figure 6 is a flowchart of the data generation method according to an embodiment of this disclosure;

[0169] Figure 7 is a flowchart of the data generation method according to an embodiment of this disclosure;

[0170] Figure 8 is a flowchart of the data generation method according to an embodiment of this disclosure;

[0171] Figure 9 is a flowchart of the data generation method according to an embodiment of this disclosure;

[0172] Figure 10 is a schematic diagram of the module structure in Figure 1;

[0173] Figure 11 is a schematic diagram of the module structure in Figure 5;

[0174] Figure 12 is a schematic diagram of the module structure in Figure 6;

[0175] Figure 13 is a schematic diagram of the module structure in Figure 7;

[0176] Figure 14 is a schematic diagram of the module structure in Figure 8;

[0177] Figure 15 is a schematic diagram of the module structure in Figure 9;

[0178] Figure 16 is a structural diagram of a network device according to an embodiment of this disclosure. Detailed Implementation

[0179] To make the technical problems, technical solutions and advantages to be solved by this disclosure clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0180] It should be understood that the phrase "an embodiment" or "one embodiment" throughout the specification means that a particular feature, structure, or characteristic relating to an embodiment is included in at least one embodiment of this disclosure. Therefore, "in one embodiment" or "one embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0181] In the various embodiments of this disclosure, it should be understood that the sequence number of each process described below does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this disclosure.

[0182] In addition, the terms "system" and "network" are often used interchangeably in this article.

[0183] In the embodiments provided in this disclosure, it should be understood that "B corresponding to A" means that B is associated with A and B can be determined based on A. However, it should also be understood that determining B based on A does not mean that B is determined solely based on A; B can also be determined based on A and / or other information.

[0184] For ease of understanding, the following describes some aspects related to the embodiments of this disclosure:

[0185] Currently, the Network Data Analytics Function (NWDAF) proposed for network intelligence is limited in its role in network intelligence due to restrictions such as data privacy.

[0186] Digital twin networks, through detailed modeling of various network entities and functions, obtain a digital model of the entire network in virtual space. This model can be virtually mapped to the physical network, replicating all or part of the physical network's functions to a certain extent. Therefore, digital twins can simulate and generate network data under different network conditions for AI model training, thus solving data problems such as data privacy and imbalanced samples.

[0187] As shown in Figure 1, a data generation method according to an embodiment of this disclosure, wherein the method is executed by a first network function, includes:

[0188] Step 11: Receive the data generation request sent by the second network function;

[0189] Step 12: Based on the data generation request, send first information to the third network function and / or send second information to the fourth network function;

[0190] The first information is used to trigger the generation of the data required for the second network function through the twin network, and the second information is used to retrieve the data required for the second network function from the data stored in the fourth network function.

[0191] That is, the first network function receives a data generation request from the second network function. After understanding the data requirements of the second network function, it sends a first message to the third network function to trigger the twin network to generate the required data; and / or sends a second message to the fourth network function to retrieve the required data from the data stored in the fourth network function. This solves data problems such as data privacy and imbalanced samples.

[0192] Optionally, the second network function can be NWDAF, which subsequently acquires this data and uses it for AI model training, thus solving problems such as poor generalization of AI models caused by missing data.

[0193] Of course, the second network function is not limited to NWDAF, but can also be other network functions that require data, which will not be listed here.

[0194] Optionally, in this embodiment, the first network function can be a network function of a Network Digital Twin (NDT) element, and the first network function can be called a Data Instruction Parsing Function (DIPF). The NDT element also includes a third network function, a fourth network function, and a sixth network function. The third network function can be called a Data Generation Orchestration Function (DGOF), the sixth network function can be called a Data Generation Function (DGF), and the fourth network function can be called a Data Storage Function (DSF).

[0195] It should be noted that in this embodiment, the NDT network element mainly constructs a twin of the physical network through digital twin technology. Depending on the level of modeling granularity, it can replicate the functions and performance of the physical network to varying degrees, thereby simulating and generating network data. Specifically, different types of data are generated under different threshold conditions to meet the multi-dimensional needs of AI model training. The structure of the NDT network element is shown in Figure 2. A first interface (Ig1 interface) is provided between the first network function (DIPF) and the third network function (DGOF); a second interface (Gg interface) is provided between the third network function (DGOF) and the sixth network function (DGF); a third interface (Ig2 interface) is provided between the first network function (DIPF) and the sixth network function (DGF); a fourth interface (Is interface) is provided between the first network function (DIPF) and the fourth network function (DSF); and a fifth interface (Gs interface) is provided between the sixth network function (DGF) and the fourth network function (DSF).

[0196] Optionally, the data generation request includes information indicating at least one of the following:

[0197] Twin scope; AI scenario description; twin granularity; data volume; feature dimension; data threshold; generation duration.

[0198] The Twin Domain refers to a twinned network domain, which can include one or more of the terminal, access network, and core network.

[0199] Among them, AI scene description is mainly used for scene description. Through this transmission message, the network element nodes and topology information required for twinning can be obtained, thereby realizing the accurate construction of twinning environment and generating real data that is closer to the physical network.

[0200] Twin granularity refers to the granularity of the twins possessed by NDT, and how fine it can be. This parameter can be used to determine the granularity of the data that NDT can simulate and generate, so as to reduce the waste of network, computing power and other resources.

[0201] Here, data volume refers to the amount of data required for the second network function.

[0202] Among them, the feature dimension represents the feature dimension of data generation. Defining the specific feature dimension required makes NDT more clear about the features to be generated, thereby reducing unnecessary data generation. On the one hand, it improves data generation efficiency, and on the other hand, it reduces storage space waste.

[0203] The data threshold setting represents the threshold conditions for data generation. This parameter allows for flexible definition of the conditions for the required data, thereby addressing issues such as the difficulty in obtaining negative sample data and poor model generalization caused by imbalanced samples.

[0204] Here, generation duration refers to the duration during which the data is generated.

[0205] As one implementation method, the data generation request includes request information, which can be: { Twin Domain: {=UE|=RAN|=CN / =UE&=RAN&=CN…}; AI Scene description: {Fault prediction; MIMO optimization…}; Twin granularity: {1:1, 1:2,…1:N}, Data volume, Feature dimension, Threshold setting, Generation duration…}.

[0206] Among them: Fault prediction refers to fault prediction; MIMO optimization refers to MIMO optimization.

[0207] Optionally, based on the data generation request, sending first information to the third network function includes:

[0208] The first information is obtained by parsing the data to generate the request;

[0209] Send the first information to the third network function;

[0210] The first information includes at least one of the following: AI scene description information; the identifier and topology of the twin network element.

[0211] That is, after receiving the data generation request, the first network function parses the data generation request to obtain AI scene description information, and / or the identifier and topology of the twin network element (Twin entity ID(s) & topology).

[0212] In one implementation, DIPF can retrieve the network domain from the Twin Domain, and then retrieve and characterize the network elements and topology relationships involved in the AI ​​scene based on the AI ​​Scene description to obtain Twin entity ID(s) & topology.

[0213] Of course, the first piece of information may also include one or more of the following: twin granularity, data volume, feature dimension, data threshold, and generation duration.

[0214] After receiving the first information, the third network function can generate a twin network based on the first information. Then, the sixth network function further configures the twin network to simulate and generate the data required by the second network function.

[0215] Optionally, to quickly respond to data request services, the third network function can set up a general template based on one or more AI scenarios. For example, one AI scenario corresponds to one general template, and this general template has a twinned process set up. Therefore, when the first information includes AI scenario description information, the system queries whether a matching general template exists based on the AI ​​scenario description information, and if a matching general template exists, it uses the general template to generate a twinned network.

[0216] The general template can be understood as a pre-made orchestration template. The third network function can retrieve a general template from the pre-made template library that matches the received AI scene description information, initiate the twinning process of the matching general template, and generate a twinning network.

[0217] Optionally, the third network function generates a twin network based on the twin network element's identifier and topology if the first information includes the twin network element's identifier and topology, and there is no matching general template, or if the first information does not include AI scene description information.

[0218] In other words, the third network function can directly initiate custom orchestration based on Twin entity ID(s) & topology to generate a Siamese network. Of course, the third network function can first search the pre-made template library of AI scene description information to see if there is a general template that matches the received AI scene description information. If not, it can then initiate custom orchestration based on Twin entity ID(s) & topology to generate a Siamese network.

[0219] Optionally, when the third network function receives multiple pieces of the first information corresponding to different data generation requests, it generates a twin network for each piece of the first information in parallel based on the multiple pieces of the first information.

[0220] That is, the second network function can send multiple data generation requests simultaneously, and after being parsed by the first network function, the third network function can process the multiple data generation requests in parallel.

[0221] Thus, as described above, in this embodiment, the third network function sets up a twin network orchestration method that combines a pre-built orchestration template adapted to multi-process services with flexible custom orchestration, for parallel data generation services of multiple instructions (data generation requests). As shown in Figure 3, when the first information is received, if the AI ​​scene description information of the first information matches the scene in the general template, then a twin network is built based on the pre-generated general template, and its twin parameters are instantiated. If the received instruction does not match the general template and there is a need for personalized customization, then the orchestration process is started according to the instruction to generate a twin network.

[0222] Optionally, in this embodiment, sending first information to the third network function according to the data generation request includes:

[0223] Upon receiving feedback information from the fourth network function, and the feedback information indicating that the fourth network function has not stored the data required by the second network function, the first information is sent to the third network function according to the data generation request.

[0224] In other words, the first network function will first send the second information to the fourth network function to query the data required by the second network function. Then, if the fourth network function does not store the data required by the second network function, it will send the first information to the third network function.

[0225] Optionally, in this embodiment, sending second information to the fourth network function according to the data generation request includes:

[0226] The second information is obtained by parsing the data generation request;

[0227] The second information is sent to the fourth network function;

[0228] The second information includes data generation conditions.

[0229] That is, after receiving the data generation request, the first network function parses the data generation request to obtain the data generation conditions, so that the fourth network function can use the data generation conditions to retrieve the data required by the second network function from the stored data.

[0230] Optionally, the data generation conditions include, but are not limited to, at least one of the following: data volume, feature dimension, data threshold, and generation duration. In this case, the content of the data generation conditions is the corresponding information of the data required by the second network function.

[0231] Optionally, in this embodiment, before receiving the data generation request sent by the second network function, the method further includes:

[0232] Send a registration request to the fifth network function, the registration request including information indicating at least one of the following:

[0233] The identifier of the network element; the scope of the twin; the processing capacity; and the data generation identifier.

[0234] In this way, NDT network elements can register with the fifth network function through the first network function, so that the fifth network function can match the second network function with an applicable NDT network element after receiving the discovery request from the second network function.

[0235] The registration request includes at least one of the following: registration information, the identifier of the network element (NDT IP), twin domain, processing capacity, and data generation identifier. The twin domain indicates the scale of the NDT's twin capabilities, encompassing which network elements and domains. The processing capacity indicates the NDT's multi-threaded processing capability, determining how many twin service processes can run simultaneously (i.e., how many twin networks can be generated in parallel). The data generation identifier identifies the data generation type.

[0236] After receiving the discovery request from the second network function, the fifth network function matches the target NDT network element based on the registration information of the registered NDT network elements and feeds back the target address information, such as the target NDT network element's Internet Protocol Address (IP). In this way, the second network function can send a data generation request to the first network function of the target NDT network element through this IP.

[0237] The discovery request includes information indicating at least one of the following: twin scope; processing capacity; data generation identifier. Here, the meanings of twin scope, processing capacity, and data generation identifier are consistent with those in the registration information.

[0238] Optionally, in this embodiment, after receiving the data generation request sent by the second network function, the method further includes:

[0239] Based on the data generation request, third information is sent to the third network function or the sixth network function, the third information including at least one of the following:

[0240] Data volume; feature dimensions; data threshold; generation duration.

[0241] That is, the first network function can parse the data generation request and directly send the third information to the sixth network function, or forward the third information to the sixth network function via the third network function. After receiving the third information, the sixth network function configures the twin network based on the third information to obtain the data required by the second network function.

[0242] Specifically, the sixth network function translates the third information into network configuration and configures it in the twin network. If it is necessary to generate data on network congestion failures caused by insufficient network bandwidth, then Threshold setting is required, and the required amount of data, feature dimensions, duration of occurrence, etc., are configured before starting the overall twin process to generate data.

[0243] Optionally, the data required by the second network function is generated by the sixth network function.

[0244] Optionally, the data required by the second network function is generated by the sixth network function based on the third information to configure the twin network.

[0245] Optionally, after obtaining the data required by the second network function, the sixth network function can send the data to the second network function. Alternatively, the sixth network function can also send the data and the third information to the fourth network function, storing them together to avoid repeatedly generating data when requesting the same data, thus reducing resource waste. Furthermore, if matching data is found, no generation is needed; the data is sent directly to the requester. If not, the data generation process is then initiated, simplifying the process. In this case, the third information can be referred to as generation conditions.

[0246] Optionally, after receiving the data and third information sent by the sixth network function, the fourth network function can first determine whether there is data associated with the third information in the currently stored data. If not, the data is associated and stored; if not, the received data is deleted to reduce space usage.

[0247] In this context, associative storage combines third-party information and data into a single entry: {generation conditions and corresponding data}. The storage format can be JavaScript Object Notation (JSON), Extensible Markup Language (XML), or similar formats.

[0248] The application of the method of this disclosure embodiment will be described below with reference to Figure 4, wherein the second network device is NWDAF as an example and the fifth network device is Network Repository Function (NRF) as an example.

[0249] Step 1: The NDT network element registers with the NRF through DIPF. The registration request includes at least {NDT IP, Data generation identifier, Twin Domain, processing capacity, etc.}.

[0250] Step 2: NWDAF sends a discovery request (NDT_Discovery_Request) to NRF, which may include {Data generation identifier, Twin Domain, processing capacity}.

[0251] Step 3: The NRF sends a discovery response (NDT_Response) to the NWDAF. The NRF selects the appropriate NDT based on the registration and discovery requests and returns the IP address to the NWDAF.

[0252] Step 4: NWDAF sends a data generation request to NDT's DIPF based on the IP, including but not limited to: { Twin Domain: {=UE|=RAN|=CN / =UE&=RAN&=CN…}; AI Scene description: {Fault prediction; MIMO optimization…}; Twin granularity: {1:1,1:2,…1:N}, Data volume, Feature dimension, Threshold setting, Generation duration…}.

[0253] Step 5: DIPF completes the data generation request parsing and obtains the first information, the second information, and the third information.

[0254] Step 6: DIPF sends the second information to DSF. Based on the second information, DIPF searches in DSF according to {generation conditions} to see if there is already data that can be satisfied in the stored data. If there is, there is no need to generate it and it is sent directly to the requester. If not, the data generation process is started and DIPF sends the first information to DGOF.

[0255] Step 7: If the output in Step 6 is "None", then DIPF Startup and Orchestration will begin: The pre-built template library will be searched to determine if a matching generic template exists. If so, the twinning process for that matching template will be initiated to generate a twin network. If not, a custom orchestration process will be initiated to generate a twin network based on Twin entity ID(s) and topology. If multiple data generation requests occur simultaneously, multiple twin service processes will be orchestrated and started.

[0256] Step 8: Building upon Step 7, DGF configures the twin network based on third-party information and simulates data generation. DGF sends the generated data to NWDAF. DGF can also send a task completion notification to DGOF, which terminates the data generation service process across multiple processes, releasing related network and computing resources.

[0257] Step 9: DGF sends the third information and the generated data ({Generation conditions, Corresponding data}) to DSF. DSF determines whether the data has been stored based on the generation condition. If so, it deletes the received data; otherwise, it stores {generation conditions and corresponding data}.

[0258] Furthermore, in this embodiment, within the NDT network element:

[0259] The first interface is the interface between DIPF and DGOF, mainly used to transmit the parsed first information;

[0260] The second interface is the interface between DGOF and DGF, mainly used to transmit conditional messages for data generation, i.e., the third information. Its specific format is as follows: {Data volume: M, Feature dimension: N, Threshold setting: =} <X%,Generation duration:T1-T2…};

[0261] The third interface: This is the interface between DIPF and DGF, mainly used to transmit conditional messages for data generation, i.e., third information. The specific format is as follows: {Data volume: M, Feature dimension: N, Threshold setting: = <X%,Generation duration:T1-T2…};

[0262] The fourth interface is the interface between DIPF and DSF. It is mainly used to transmit the data generation condition message parsed by DIPF, i.e., the second information, which is used to retrieve and determine whether there is corresponding data stored in DSF.

[0263] The fifth interface is the interface between DSF and DGF, mainly used to transmit data generation conditions and corresponding generation data. The specific format is as follows: {Generation Condition: Corresponding Data}.

[0264] In addition, in this embodiment, the interface (Nndt interface) between the NDT network element and the second network function is used to transmit the aforementioned data generation request, with the following specific format: { Twin Domain: {=UE|=RAN|=CN / =UE&=RAN&=CN…}; AI Scene description: {Fault prediction; MIMO optimization…}; Twin granularity: {1:1, 1:2,…1:N}, Data volume: N, Feature dimension: M, Threshold setting: = <X%, Generation duration:T1-T2, …}。

[0265] In summary, the method of this disclosure, through digital twins, can simulate and generate network data under different network conditions for AI model training, thereby solving data problems such as data privacy and imbalanced samples.

[0266] As shown in Figure 5, a data generation method according to an embodiment of this disclosure, executed by a third network function, includes:

[0267] Step 51: Receive the first information sent by the first network function;

[0268] Step 52: Generate a twin network based on the first information.

[0269] In this way, the third network function, according to steps 51 and 52, can generate a twin network based on the first information sent by the first network function, so that the required data can be obtained through the twin network in the future, thus solving data problems such as data privacy and imbalanced samples.

[0270] Optionally, the first information includes at least one of the following: AI scene description information; the identifier and topology of the twin network element.

[0271] Optionally, generating a twin network based on the first information includes:

[0272] If the first information includes AI scene description information, a query is performed based on the AI ​​scene description information to see if a matching general template exists. If a matching general template exists, a Siamese network is generated using the general template; or,

[0273] If the first information includes the identifier and topology of the twin network element, and there is no matching general template, or if the first information does not include AI scene description information, a twin network is generated based on the identifier and topology of the twin network element.

[0274] Optionally, it also includes:

[0275] Receive the third information sent by the first network function;

[0276] Send the third information to the sixth network function;

[0277] The third information includes at least one of the following:

[0278] Data volume; feature dimensions; data threshold; generation duration.

[0279] Optionally, generating a twin network based on the first information includes:

[0280] When multiple pieces of the first information corresponding to different data generation requests are received, a twin network for each piece of the first information is generated in parallel based on the multiple pieces of the first information.

[0281] It should be noted that this method is implemented in conjunction with the data generation method of the above embodiments. The implementation of the above method embodiments is applicable to this method and can achieve the same technical effect.

[0282] As shown in Figure 6, an embodiment of this disclosure provides a data generation method, executed by a sixth network function, including:

[0283] Step 61: Receive third information sent by the first network function or the third network function;

[0284] Step 62: Configure the twin network according to the third information to obtain the data required for the second network function; wherein the third information includes at least one of the following:

[0285] Data volume; feature dimensions; data threshold; generation duration.

[0286] In this way, after receiving the third information, the sixth network function uses the third information to configure the twin network information, thereby obtaining the required data and solving data problems such as data privacy and imbalanced samples.

[0287] Among them, the twin network is generated by the third network function based on the first information sent by the first network function.

[0288] Optionally, after configuring the twin network based on the third information to obtain the data required for the second network function, the method further includes:

[0289] The data is sent to the second network function.

[0290] Optionally, after configuring the twin network based on the third information to obtain the data required for the second network function, the method further includes:

[0291] The data and the third information are sent to the fourth network function.

[0292] It should be noted that this method is implemented in conjunction with the data generation method of the above embodiments. The implementation of the above method embodiments is applicable to this method and can achieve the same technical effect.

[0293] As shown in Figure 7, an embodiment of this disclosure provides a data generation method, executed by a fourth network function, including:

[0294] Step 71: Receive data and third information sent by the fifth network function;

[0295] Step 72: Associate and store the data and the third information.

[0296] The third information includes at least one of the following:

[0297] Data volume; feature dimensions; data threshold; generation duration.

[0298] Thus, after the fourth network function associates and stores the data received from the fifth network function with the third information according to steps 71 and 72, the first network function can easily query the required data based on the storage.

[0299] Optionally, the data and the third information are stored together, including:

[0300] If no data associated with the third information exists in the current stored data, the received data and the third information will be associated and stored together.

[0301] Optionally, it also includes:

[0302] If data associated with the third information exists in the current stored data, the received data will be deleted.

[0303] It should be noted that this method is implemented in conjunction with the data generation method of the above embodiments. The implementation of the above method embodiments is applicable to this method and can achieve the same technical effect.

[0304] As shown in Figure 8, an embodiment of this disclosure provides a data generation method, executed by a second network function, including:

[0305] Step 81: Send a data generation request to the first network function, the data generation request being used to request the data required by the second network function;

[0306] Step 82: Receive data sent by the fourth or sixth network function.

[0307] That is, according to steps 81 and 82, the second network function will request the required data from the first network function through a data generation request, thereby receiving the required data sent by the fourth or sixth network function.

[0308] The first network function will send first information to the third network function and / or send second information to the fourth network function after receiving a data generation request from the second network function. The first information is used to trigger the generation of data required by the second network function through the twin network, and the second information is used to retrieve the data required by the second network function from the data stored in the fourth network function.

[0309] Optionally, the data generation request includes information indicating at least one of the following:

[0310] Twin scope; AI scenario description; twin granularity; data volume; feature dimension; data threshold; generation duration.

[0311] Optionally, before sending the data generation request to the first network function, the method further includes:

[0312] Send a discovery request to the fifth network function;

[0313] Receive the target address information fed back by the fifth network function;

[0314] The discovery request includes information indicating at least one of the following:

[0315] Twin scope; processing capacity; data generation identifier.

[0316] It should be noted that this method is implemented in conjunction with the data generation method of the above embodiments. The implementation of the above method embodiments is applicable to this method and can achieve the same technical effect.

[0317] As shown in Figure 9, an embodiment of this disclosure provides a data generation method, executed by a fifth network function, including:

[0318] Step 91: Receive a registration request sent by the first network function, the registration request including information indicating at least one of the following:

[0319] The identifier of the network element; the scope of the twin; the processing capacity; and the data generation identifier.

[0320] That is, the fifth network function will receive the registration request sent by the first network function, and the first network function will register so that it can be matched with the appropriate first network function in the future.

[0321] Optionally, it also includes:

[0322] Receive discovery requests sent by the second network function;

[0323] Based on the discovery request, determine the first matching network function;

[0324] Send the target address information of the first network function to the second network function;

[0325] The discovery request includes information indicating at least one of the following:

[0326] Twin scope; processing capacity; data generation identifier.

[0327] It should be noted that this method is implemented in conjunction with the data generation method of the above embodiments. The implementation of the above method embodiments is applicable to this method and can achieve the same technical effect.

[0328] As shown in Figure 10, an embodiment of this disclosure provides a data generation apparatus applied to a first network function, including:

[0329] The first receiving module 1010 is used to receive a data generation request sent by the second network function;

[0330] The first sending module 1020 is used to generate a request based on the data, send first information to the third network function, and / or send second information to the fourth network function;

[0331] The first information is used to trigger the generation of the data required for the second network function through the twin network, and the second information is used to retrieve the data required for the second network function from the data stored in the fourth network function.

[0332] The device receives a data generation request from a second network function. After understanding the data requirements of the second network function, it sends a first message to a third network function to trigger the twin network to generate the required data; and / or sends a second message to a fourth network function to retrieve the required data from the data stored in the fourth network function. This solves data problems such as data privacy and imbalanced samples.

[0333] Optionally, the data generation request includes information indicating at least one of the following:

[0334] Twin scope; AI scenario description; twin granularity; data volume; feature dimension; data threshold; generation duration.

[0335] Optionally, the first sending module is further configured to:

[0336] The first information is obtained by parsing the data to generate the request;

[0337] Send the first information to the third network function;

[0338] The first information includes at least one of the following: AI scene description information; the identifier and topology of the twin network element.

[0339] Optionally, the first sending module is further configured to:

[0340] The second information is obtained by parsing the data generation request;

[0341] The second information is sent to the fourth network function;

[0342] The second information includes data generation conditions.

[0343] Optionally, the first sending module is further configured to:

[0344] Upon receiving feedback information from the fourth network function, and the feedback information indicating that the fourth network function has not stored the data required by the second network function, the first information is sent to the third network function according to the data generation request.

[0345] Optionally, the device further includes:

[0346] The registration module is used to send a registration request to the fifth network function, the registration request including information indicating at least one of the following:

[0347] The identifier of the network element; the scope of the twin; the processing capacity; and the data generation identifier.

[0348] Optionally, the device further includes:

[0349] The third sending module is configured to generate a request based on the data and send third information to the third network function or the sixth network function, wherein the third information includes at least one of the following:

[0350] Data volume; feature dimensions; data threshold; generation duration.

[0351] Optionally, the data required by the second network function is generated by the sixth network function based on the third information to configure the twin network.

[0352] It should be noted that the device uses the data generation method executed by the first network function described above. The implementation of the above method embodiment is applicable to this device and can achieve the same technical effect.

[0353] As shown in Figure 11, an embodiment of this disclosure provides a data generation apparatus applied to a third network function, including:

[0354] The second receiving module 1110 is used to receive the first information sent by the first network function;

[0355] The first processing module 1120 is used to generate a twin network based on the first information.

[0356] This device can generate a twin network based on the first information sent by the first network function, so that the required data can be obtained through the twin network in the future, thus solving data problems such as data privacy and imbalanced samples.

[0357] Optionally, the first information includes at least one of the following: AI scene description information; the identifier and topology of the twin network element.

[0358] Optionally, the first processing module is further configured to:

[0359] If the first information includes AI scene description information, a query is performed based on the AI ​​scene description information to see if a matching general template exists. If a matching general template exists, a Siamese network is generated using the general template; or,

[0360] If the first information includes the identifier and topology of the twin network element, and there is no matching general template, or if the first information does not include AI scene description information, a twin network is generated based on the identifier and topology of the twin network element.

[0361] Optionally, the device further includes:

[0362] The seventh receiving module is used to receive the third information sent by the first network function;

[0363] The fourth sending module is used to send the third information to the sixth network function;

[0364] The third information includes at least one of the following:

[0365] Data volume; feature dimensions; data threshold; generation duration.

[0366] Optionally, the first processing module is further configured to:

[0367] When multiple pieces of the first information corresponding to different data generation requests are received, a twin network for each piece of the first information is generated in parallel based on the multiple pieces of the first information.

[0368] It should be noted that the device uses the data generation method executed by the third network function described above. The implementation of the above method embodiment is applicable to this device and can achieve the same technical effect.

[0369] As shown in Figure 12, an embodiment of this disclosure provides a data generation apparatus applied to a sixth network function, including:

[0370] The third receiving module 1210 is used to receive third information sent by the first network function or the third network function;

[0371] The second processing module 1220 is used to configure the twin network according to the third information and obtain the data required for the second network function.

[0372] The third information includes at least one of the following:

[0373] Data volume; feature dimensions; data threshold; generation duration.

[0374] After receiving the third information, the device uses the third information to configure the twin network information, thereby obtaining the required data and solving data problems such as data privacy and imbalanced samples.

[0375] Optionally, the device further includes:

[0376] The fifth sending module is used to send the data to the second network function.

[0377] Optionally, the device further includes:

[0378] The sixth sending module is used to send the data and the third information to the fourth network function.

[0379] It should be noted that the device uses the data generation method executed by the sixth network function described above. The implementation of the above method embodiment is applicable to this device and can achieve the same technical effect.

[0380] As shown in Figure 13, an embodiment of this disclosure provides a data generation apparatus applied to a fourth network function, including:

[0381] The fourth receiving module 1310 is used to receive data and third information sent by the fifth network function;

[0382] Storage module 1320 is used to associate and store the data and the third information;

[0383] The third information includes at least one of the following:

[0384] Data volume; feature dimensions; data threshold; generation duration.

[0385] After the device associates and stores the data received from the fifth network function with the third information, it can facilitate the first network function to query the required data based on the stored data.

[0386] Optionally, the storage module is further configured to:

[0387] If no data associated with the third information exists in the current stored data, the received data and the third information will be associated and stored together.

[0388] Optionally, the device further includes:

[0389] The third processing module is used to delete the received data if data associated with the third information exists in the current stored data.

[0390] It should be noted that the device uses the data generation method executed by the fourth network function described above. The implementation of the above method embodiment is applicable to this device and can achieve the same technical effect.

[0391] As shown in Figure 14, an embodiment of this disclosure provides a data generation apparatus applied to a second network function, including:

[0392] The second sending module 1410 is used to send a data generation request to the first network function, the data generation request being used to request data required by the second network function;

[0393] The fifth receiving module 1420 is used to receive data sent by the fourth network function or the sixth network function.

[0394] The device requests the required data from the first network function through a data generation request, thereby receiving the required data sent by the fourth or sixth network function.

[0395] Optionally, the data generation request includes information indicating at least one of the following:

[0396] Twin scope; AI scenario description; twin granularity; data volume; feature dimension; data threshold; generation duration.

[0397] Optionally, the device further includes:

[0398] The discovery module is used to send discovery requests to the fifth network function.

[0399] The eighth receiving module is used to receive the target address information fed back by the fifth network function;

[0400] The discovery request includes information indicating at least one of the following:

[0401] Twin scope; processing capacity; data generation identifier.

[0402] It should be noted that the device uses the data generation method executed by the second network function described above. The implementation of the above method embodiment is applicable to this device and can achieve the same technical effect.

[0403] As shown in Figure 15, an embodiment of this disclosure provides a data generation apparatus applied to a fifth network function, including:

[0404] The sixth receiving module 1510 is configured to receive a registration request sent by the first network function, the registration request including information indicating at least one of the following:

[0405] The identifier of the network element; the scope of the twin; the processing capacity; and the data generation identifier.

[0406] The device receives a registration request from the first network function, which then registers itself so that it can be matched with suitable first network functions in the future.

[0407] Optionally, the device further includes:

[0408] The ninth receiving module is used to receive discovery requests sent by the second network function;

[0409] The fourth processing module is used to determine the first matching network function based on the discovery request;

[0410] The sixth sending module is used to send the target address information of the first network function to the second network function;

[0411] The discovery request includes information indicating at least one of the following:

[0412] Twin scope; processing capacity; data generation identifier.

[0413] It should be noted that the device uses the data generation method executed by the fifth network function described above. The implementation of the above method embodiment is applicable to this device and can achieve the same technical effect.

[0414] As shown in Figure 2, an embodiment of this disclosure provides a network element, including a first network function that performs the data generation method described above, a third network function that performs the data generation method described above, a sixth network function that performs the data generation method described above, and a fourth network function that performs the data generation method described above.

[0415] Optionally, a first interface is provided between the first network function and the third network function, a second interface is provided between the third network function and the sixth network function, a third interface is provided between the first network function and the sixth network function, a fourth interface is provided between the first network function and the fourth network function, and a fifth interface is provided between the sixth network function and the fourth network function.

[0416] Embodiments of this disclosure provide a data generation system, including: a network element as described above, a second network function performing the data generation method as described above, and a fifth network function performing the data generation method as described above.

[0417] As shown in Figure 16, an embodiment of this disclosure provides a network device, including a transceiver 1610, a processor 1600, a memory 1620, and a program or instructions stored in the memory 1620 and executable on the processor 1600; when the processor 1600 executes the program or instructions, it implements the above-described data generation method.

[0418] The transceiver 1610 is used to receive and send data under the control of the processor 1600.

[0419] In Figure 16, the bus architecture may include any number of interconnected buses and bridges, specifically linking various circuits of one or more processors represented by processor 1600 and memory represented by memory 1620. The bus architecture may also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. The bus interface provides an interface. Transceiver 1610 may be multiple elements, including transmitters and receivers, providing units for communicating with various other devices over a transmission medium. Processor 1600 is responsible for managing the bus architecture and general processing, and memory 1620 may store data used by processor 1600 during operation.

[0420] Embodiments of this disclosure provide a readable storage medium having a program or instructions stored thereon, which, when executed by a processor, implement the steps in the data generation method described above.

[0421] The embodiments of this disclosure provide a computer program product, including computer instructions. When the computer instructions are executed by a processor, they implement the various processes of the method embodiments shown in Figures 1, 5, 6, 7, 8, or 9 above, and can achieve the same technical effects. To avoid repetition, they will not be described again here.

[0422] It should be further noted that the terminals described in this specification include, but are not limited to, smartphones, tablets, etc., and many of the functional components described are referred to as modules in order to emphasize the independence of their implementation.

[0423] In this disclosure, the module can be implemented in software so that it can be executed by various types of processors. For example, an identified executable code module may include one or more physical or logical blocks of computer instructions, which may be constructed as objects, procedures, or functions. Nevertheless, the executable code of the identified module does not need to be physically located together, but may include different instructions stored in different bits, which, when logically combined, constitute the module and achieve the module's intended purpose.

[0424] In practice, an executable code module can be a single instruction or many instructions, and can even be distributed across multiple different code segments, different programs, and across multiple memory devices. Similarly, operational data can be identified within the module and can be implemented in any suitable form and organized within any suitable type of data structure. This operational data can be collected as a single dataset or distributed across different locations (including different storage devices), and can exist, at least in part, solely as electronic signals within the system or network.

[0425] When a module can be implemented using software, considering the current level of hardware technology, modules that can be implemented in software can be implemented using hardware circuits by those skilled in the art to achieve the corresponding functions, without considering cost. These hardware circuits include conventional Very Large Scale Integration (VLSI) circuits or gate arrays, as well as existing semiconductors such as logic chips and transistors, or other discrete components. Modules can also be implemented using programmable hardware devices, such as field-programmable gate arrays, programmable array logic, and programmable logic devices.

[0426] The exemplary embodiments described above are with reference to the accompanying drawings. Many different forms and embodiments are feasible without departing from the spirit and teaching of this disclosure. Therefore, this disclosure should not be construed as limiting the exemplary embodiments set forth herein. Rather, these exemplary embodiments are provided to make this disclosure complete and convey the scope of this disclosure to those skilled in the art. In these drawings, component dimensions and relative dimensions may be exaggerated for clarity. The terminology used herein is for the purpose of describing particular exemplary embodiments only and is not intended to be limiting. As used herein, unless clearly indicated otherwise, the singular forms “a,” “an,” and “the” are intended to include all such forms. It will be further understood that the terms “comprising” and / or “including”, when used in this specification, indicate the presence of the stated features, integers, steps, operations, components, and / or elements, but do not exclude the presence or addition of one or more other features, integers, steps, operations, components, and / or groups thereof. Unless otherwise indicated, when stated, a range of values ​​includes the upper and lower limits of the range and any subranges in between.

[0427] The above description represents the preferred embodiments of this disclosure. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles described herein, and these improvements and modifications should also be considered within the scope of protection of this disclosure.

Claims

1. A data generation method, executed by a first network function, the method comprising: Receive data generation requests sent by the second network function; Based on the data generation request, send first information to the third network function, and / or send second information to the fourth network function; The first information is used to trigger the generation of the data required for the second network function through the twin network, and the second information is used to retrieve the data required for the second network function from the data stored in the fourth network function.

2. The method according to claim 1, wherein, The data generation request includes information indicating at least one of the following: Twin scope; AI scenario description; twin granularity; data volume; feature dimension; data threshold; generation duration.

3. The method according to claim 1 or 2, wherein, Based on the data generation request, send first information to the third network function, including: The first information is obtained by parsing the data to generate the request; Send the first information to the third network function; The first information includes at least one of the following: AI scene description information; the identifier and topology of the twin network element.

4. The method according to claim 1 or 2, wherein, Based on the data generation request, send second information to the fourth network function, including: The second information is obtained by parsing the data generation request; The second information is sent to the fourth network function; The second information includes data generation conditions.

5. The method according to claim 1, wherein, Based on the data generation request, send first information to the third network function, including: Upon receiving feedback information from the fourth network function, and the feedback information indicating that the fourth network function has not stored the data required by the second network function, the first information is sent to the third network function according to the data generation request.

6. The method according to claim 1, wherein, Before receiving the data generation request sent by the second network function, it also includes: Send a registration request to the fifth network function, the registration request including information indicating at least one of the following: The identifier of the network element; the scope of the twin; the processing capacity; and the data generation identifier.

7. The method according to claim 1, wherein, After receiving the data generation request sent by the second network function, it also includes: Based on the data generation request, third information is sent to the third network function or the sixth network function, the third information including at least one of the following: Data volume; feature dimensions; data threshold; generation duration.

8. The method according to claim 7, wherein, The data required for the second network function is generated by the sixth network function based on the third information configured in the twin network.

9. A data generation method, executed by a third network function, the method comprising: Receive the first information sent by the first network function; Generate a twin network based on the first information.

10. The method according to claim 9, wherein, The first information includes at least one of the following: AI scene description information; the identifier and topology of the twin network element.

11. The method according to claim 10, wherein, Generate a twin network based on the first information, including: If the first information includes AI scene description information, a query is performed based on the AI ​​scene description information to see if a matching general template exists. If a matching general template exists, a Siamese network is generated using the general template; or, If the first information includes the identifier and topology of the twin network element, and there is no matching general template, or if the first information does not include AI scene description information, a twin network is generated based on the identifier and topology of the twin network element.

12. The method according to claim 9, further comprising: Receive the third information sent by the first network function; Send the third information to the sixth network function; The third information includes at least one of the following: Data volume; feature dimensions; data threshold; generation duration.

13. The method according to claim 9, wherein, Generate a twin network based on the first information, including: When multiple pieces of the first information corresponding to different data generation requests are received, a twin network for each piece of the first information is generated in parallel based on the multiple pieces of the first information.

14. A data generation method, executed by a sixth network function, comprising: Receive third information sent by the first network function or the third network function; Configure the twin network based on the third information to obtain the data required for the second network function; The third information includes at least one of the following: Data volume; feature dimensions; data threshold; generation duration.

15. The method according to claim 14, wherein, After configuring the twin network based on the third information and obtaining the data required for the second network function, the process further includes: The data is sent to the second network function.

16. The method of claim 14, wherein, After configuring the twin network based on the third information and obtaining the data required for the second network function, the process further includes: The data and the third information are sent to the fourth network function.

17. A data generation method, performed by a fourth network function, comprising: Receive data and third-party information sent by the fifth network function; The data and the third information are associated and stored; The third information includes at least one of the following: Data volume; feature dimensions; data threshold; generation duration.

18. The method according to claim 17, wherein, The data and the third information are associated and stored, including: If no data associated with the third information exists in the current stored data, the received data and the third information will be associated and stored together.

19. The method of claim 17, further comprising: If data associated with the third information exists in the current stored data, the received data will be deleted.

20. A data generation method, executed by a second network function, comprising: Send a data generation request to the first network function, the data generation request being used to request data required by the second network function; Receive data sent by the fourth or sixth network function.

21. The method according to claim 20, wherein, The data generation request includes information indicating at least one of the following: Twin scope; AI scenario description; twin granularity; data volume; feature dimension; data threshold; generation duration.

22. The method according to claim 20, wherein, Before sending the data generation request to the first network function, it also includes: Send a discovery request to the fifth network function; Receive the target address information fed back by the fifth network function; The discovery request includes information indicating at least one of the following: Twin scope; processing capacity; data generation identifier.

23. A data generation method, executed by a fifth network function, comprising: Receive a registration request sent by a first network function, the registration request including information indicating at least one of the following: The identifier of the network element to which it belongs; Twin scope; processing capacity; data generation identifier.

24. The method according to claim 23, further comprising: Receive discovery requests sent by the second network function; Based on the discovery request, determine the first matching network function; Send the target address information of the first network function to the second network function; The discovery request includes information indicating at least one of the following: Twin scope; processing capacity; data generation identifier.

25. A data generation apparatus for a first network function, the apparatus comprising: The first receiving module is used to receive data generation requests sent by the second network function; The first sending module is used to generate a request based on the data, send first information to the third network function, and / or send second information to the fourth network function; The first information is used to trigger the generation of the data required for the second network function through the twin network, and the second information is used to retrieve the data required for the second network function from the data stored in the fourth network function.

26. A data generation apparatus for use in a third network function, the apparatus comprising: The second receiving module is used to receive the first information sent by the first network function; The first processing module is used to generate a twin network based on the first information.

27. A data generation apparatus for use in a sixth network function, the apparatus comprising: The third receiving module is used to receive third information sent by the first network function or the third network function; The second processing module is used to configure the twin network according to the third information and obtain the data required for the second network function. The third information includes at least one of the following: Data volume; feature dimensions; data threshold; generation duration.

28. A data generation apparatus for a fourth network function, the apparatus comprising: The fourth receiving module is used to receive data and third information sent by the fifth network function; A storage module is used to associate and store the data and the third information. The third information includes at least one of the following: Data volume; feature dimensions; data threshold; generation duration.

29. A data generation apparatus for a second network function, the apparatus comprising: The second sending module is used to send a data generation request to the first network function, the data generation request being used to request data required by the second network function; The fifth receiving module is used to receive data sent by the fourth or sixth network function.

30. A data generation apparatus for use in a fifth network function, the apparatus comprising: The sixth receiving module is configured to receive a registration request sent by the first network function, the registration request including information indicating at least one of the following: The identifier of the network element; the scope of the twin; the processing capacity; and the data generation identifier.

31. A network element comprising performing a first network function of the data generation method as described in any one of claims 1-8, a third network function of the data generation method as described in any one of claims 9-13, a sixth network function of the data generation method as described in any one of claims 14-16, and a fourth network function of the data generation method as described in any one of claims 17-19.

32. The network element according to claim 31, wherein a first interface is provided between the first network function and the third network function, a second interface is provided between the third network function and the sixth network function, a third interface is provided between the first network function and the sixth network function, a fourth interface is provided between the first network function and the fourth network function, and a fifth interface is provided between the sixth network function and the fourth network function.

33. A data generation system, comprising: The network element as described in claim 31 or 32 performs a second network function of the data generation method as described in any one of claims 20-22, and performs a fifth network function of the data generation method as described in claim 23 or 24.

34. A network device, comprising: Transceiver, processor, memory, and programs or instructions stored in the memory and executable on the processor; When the processor executes the program or instructions, it implements the data generation method as described in any one of claims 1-8, or the data generation method as described in any one of claims 9-13, or the data generation method as described in any one of claims 14-16, or the data generation method as described in any one of claims 17-19, or the data generation method as described in any one of claims 20-22, or the data generation method as described in claim 23 or 24.

35. A readable storage medium having a program or instructions stored thereon, which, when executed by a processor, implement the data generation method as claimed in any one of claims 1-8, or the data generation method as claimed in any one of claims 9-13, or the data generation method as claimed in any one of claims 14-16, or the data generation method as claimed in any one of claims 17-19, or the data generation method as claimed in any one of claims 20-22, or the steps of the data generation method as claimed in claim 23 or 24.

36. A computer program product comprising computer instructions that, when executed by a processor, implement the data generation method as described in any one of claims 1-8, or the data generation method as described in any one of claims 9-13, or the data generation method as described in any one of claims 14-16, or the data generation method as described in any one of claims 17-19, or the data generation method as described in any one of claims 20-22, or the steps of the data generation method as described in claim 23 or 24.

Images