Data transmission method, device and storage medium

Abstract

The present application proposes a data transmission method, device and storage medium. The data transmission method applied to a digital twin system includes: generating network prediction parameters based on network data received from a physical device; feeding back the network prediction parameters to the physical device; wherein the digital twin system and the physical device are co-deployed.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a data transmission method, device and storage medium. Background Art

[0002] The wireless communication network is a complex and ever-changing system. In order to provide users with high-quality network services, network equipment needs to use network environment status information, such as channel characteristics, user business models, and user movement trajectories, to assist in internal algorithm decision-making (such as adaptive coding and modulation schemes, space division multiplexing stream number selection, etc.). In addition, network operators and managers often need to change the parameter configuration of network equipment or perform transformation / upgrades to cope with a variety of application scenarios and network conditions. Before these operations are implemented, their effects need to be effectively pre-verified, and they can only be officially put into use if they meet expectations.

[0003] The network environment status information can generally be obtained by measuring and estimating the reference signal, measuring and reporting by the user equipment (UE), extracting from historical records, etc. These methods require occupying time and frequency domain resources and consuming system computing power, and the information obtained has a lag.

[0004] For operations such as wireless network upgrades / renovations and parameter configuration changes, performance tests are generally performed in a laboratory environment or simulation platform before the operation is put into use. These test environments are different from the actual field, so the wireless system performance test indicators obtained are not particularly consistent with the actual results. Summary of the invention

[0005] In view of this, the embodiments of the present application provide a data transmission method, device and storage medium, which realize a digital twin system architecture in which the digital twin system and physical devices are deployed at the same site, effectively improving the test accuracy of network prediction parameters.

[0006] The embodiment of the present application provides a data transmission method, which is applied to a digital twin system, including:

[0007] generating network prediction parameters based on network data received from the physical device;

[0008] The network prediction parameters are fed back to the physical device; wherein the digital twin system and the physical device are deployed at the same site.

[0009] The present application provides a data transmission method, which is applied to a physical device and includes:

[0010] Send the collected network data to the digital twin system;

[0011] Receive network prediction parameters generated by the digital twin system; wherein the digital twin system is deployed at the same site as the physical device.

[0012] The embodiment of the present application provides a data transmission device, which is applied to a digital twin system, including:

[0013] a generating module configured to generate network prediction parameters based on network data received from the physical device;

[0014] A sending module is configured to feed back the network prediction parameters to the physical device; wherein the digital twin system and the physical device are deployed at the same site.

[0015] The present application provides a data transmission device, which is applied to a physical device, including:

[0016] A sending module, configured to send the collected network data to the digital twin system;

[0017] A receiving module is configured to receive network prediction parameters generated by the digital twin system; wherein the digital twin system and the physical device are deployed at the same site.

[0018] An embodiment of the present application provides a communication device, including: a memory, and one or more processors;

[0019] The memory is configured to store one or more programs;

[0020] When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any of the above embodiments.

[0021] An embodiment of the present application provides a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any of the above embodiments is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a flow chart of a data transmission method provided in an embodiment of the present application;

[0023] Figure 2 is a flow chart of another data transmission method provided by an embodiment of the present application;

[0024] Figure 3 It is a schematic diagram of the principle of a digital twin system provided in an embodiment of the present application;

[0025] Figure 4 It is a schematic diagram of the implementation principle of predicting millimeter wave smart beams based on a digital twin system provided in an embodiment of the present application;

[0026] Figure 5 This is a schematic diagram of the implementation principle of online training and verification of the gNB AI MAC scheduling function based on a digital twin system provided in an embodiment of the present application;

[0027] Figure 6 is a structural block diagram of a data transmission device provided in an embodiment of the present application;

[0028] Figure 7 is a structural block diagram of another data transmission device provided in an embodiment of the present application;

[0029] Figure 8 It is a structural diagram of a communication device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0030] The following will describe the embodiments of the present application in conjunction with the accompanying drawings. The following will describe the present application in conjunction with the accompanying drawings of the embodiments, and the examples cited are only used to explain the present application and are not used to limit the scope of the present application.

[0031] In one embodiment, Figure 1 is a flowchart of a data transmission method provided by an embodiment of the present application. This embodiment is applied to the case of real-time prediction and evaluation of network status and performance. This embodiment can be executed by a digital twin system. In one example, a digital twin system refers to a digital software system that provides twin application services for physical network elements. Figure 1 As shown, this embodiment includes: S110-S120.

[0032] S110. Generate network prediction parameters based on network data received from physical devices.

[0033] In one example, a physical device may also be referred to as a physical network element, which refers to a network element device that actually exists in a physical network; physical devices include but are not limited to cellular networks, campus enterprise networks, industrial Internet of Things, and campus LANs; network element devices include but are not limited to base stations, core network equipment, transmission network equipment, and WLAN access points.

[0034] In one example, the network prediction parameter refers to prediction and simulation data of network status and / or prediction and simulation data of network performance. In one example, network data is used to characterize the current status data of the network where the physical device is located. Exemplarily, the network data may include but is not limited to one of the following: data measured by the physical layer and high layers for characterizing the quality of wireless signals (for example, reference signal received power (RSRP), reference signal received quality (RSRQ), received signal strength indicator (RSSI) and signal to interference plus noise ratio (SINR), etc.), data for characterizing network status and performance (for example, resource utilization, resource scheduling information, event records and parameter changes, etc.), user measurements and event reporting (for example, channel quality indication (CQI), rank indication (RI), precoding matrix indication (PMI) and neighboring RSRP, etc.), user service quality of service (QoS) metrics (for example, delay, packet loss rate, jitter and throughput, etc.).

[0035] In one example, the physical device can send the collected network data to the digital twin system in real time or periodically. The digital twin system analyzes and processes the network data to generate corresponding prediction and simulation data of network status and performance.

[0036] S120. Feedback the network prediction parameters to the physical device; wherein the digital twin system and the physical device are deployed at the same site.

[0037] In one example, the digital twin system sends network prediction parameters that characterize network status and performance to the physical device, so that the physical device uses internal algorithms and makes internal decisions based on the network prediction parameters to determine subsequent execution operations (for example, determining the number of spatial multiplexing streams to be used, coding and modulation formats, etc.), thereby effectively pre-verifying the effects of the execution operations before they are implemented; at the same time, co-deploying the digital twin system and the physical device at the same site can avoid the differences between the test environment and the actual field, thereby ensuring the consistency and fit between the network prediction parameters and the actual results.

[0038] In one embodiment, the digital twin system and the physical device are co-located, including at least one of the following: the digital twin system and the physical device are located in the same physical location; the digital twin system and the physical device share the same physical hardware; the digital twin system is integrated in the hardware control board inserted into the physical device. In one example, the digital twin system and the physical device are located in the same physical location, which can be understood as the digital twin system being deployed locally on the physical device. Deploying the digital twin system and the physical device locally can effectively improve the real-time and security of the digital twin system. On the one hand, the digital twin system and the physical device can interact in real time. For example, the digital twin system can synchronize the network data collected by the physical device in real time or periodically. Driven by the collected network data, the digital twin system generates the corresponding prediction and simulation data of the network status and performance as the network prediction parameters, and feeds the network prediction parameters back to the network device; on the other hand, the local secure connection between the digital twin system and the physical device can reduce the risk of network attacks and privacy leakage.

[0039] In one embodiment, the digital twin system includes at least two digital twin models; the data transmission method applied to the digital twin system also includes: generating associated digital twin entities based on the digital twin model; the digital twin entity builds a digital sandbox environment. In one example, a digital twin model library may be included in the digital twin system, and a plurality of digital twin models are included in the digital twin model library. The digital twin model library refers to a computer code library containing various digital twin models; the digital twin model is a digital representation of a real-world entity or system, which carries knowledge such as natural science, professional technology, industry standards and specifications, domain expert experience, mathematics or statistics, and specifically can be a computer program code that implements a mathematical model, an AI / ML model, an algorithm flow, a protocol flow or a specific function. Exemplarily, the digital twin model may include but is not limited to at least one of the following: a twin channel model, a UE mobile trajectory twin model, a UE service twin model, a UE twin model, a core network twin model, and a base station behavior model. Several digital twin entities can be generated on demand based on these digital twin models, and a digital sandbox environment can be jointly constructed through these digital twin entities. In one example, the digital twin entity may include but is not limited to one of the following: twin channel, twin interference, twin service data, twin UE and twin core network, etc.

[0040] In one embodiment, the physical device runs in the digital sandbox environment and performs parameter and function verification, and uses the network prediction parameters provided by the digital sandbox environment to perform network operations. The physical device runs in the digital sandbox environment and can complete its own parameter and function verification, and use the network prediction parameters provided by the digital sandbox environment to assist the algorithm in making decisions on subsequent execution operations.

[0041] In one embodiment, the digital twin system includes at least two digital twin service applications; each digital twin service application uses a different digital twin entity to provide a specific function service to the physical device. In one example, the digital twin service application can be a digital twin service APP. A digital twin service application pool can be included in the digital twin system, and a variety of digital twin service applications are included in the digital twin service application pool. In one example, the digital twin service application refers to an application that can provide specific functions and services to the physical device; the digital twin service application calls a specific digital twin model to generate a number of digital twin entities, and combines them to implement the specified function. The number of digital twin service applications in the digital twin system can change dynamically, and can be added or deleted as needed. The digital twin service application can be deployed in containerization (such as docker, Podman), virtual machines, and executable files. Each digital twin service application can use different digital twin entities to provide specific functional services to the physical device; for example, predicting the channel, analyzing the root cause of the poor performance of the physical device, verifying the effect of the network data corresponding to the physical device, and online verification of the AI ​​function of the physical device.

[0042] In one embodiment, the digital twin system includes: a model and application orchestration management subsystem for orchestrating and managing the lifecycle of digital twin models and digital twin service applications. The model and application orchestration management subsystem orchestrates and manages the lifecycle of digital twin models and digital twin service applications, such as registering, activating, and deactivating service applications, and registering, deploying, training, generating, testing, monitoring, and updating digital twin models.

[0043] In one embodiment, the model and application orchestration management subsystem deploys and activates the digital twin service application on demand. In one example, the digital twin service application in the digital twin system can be deployed and activated on demand by the model and application orchestration management subsystem.

[0044] In one embodiment, generating network prediction parameters based on network data collected from physical devices includes: periodically collecting network data from physical devices; generating prediction and simulation data based on the network data as network prediction parameters. The digital twin entity in the digital twin system can generate corresponding prediction and simulation data of network status and performance as network prediction parameters based on the network data collected in real time by the physical device and the digital twin model it complies with.

[0045] In one embodiment, generating prediction and simulation data based on network data as network prediction parameters includes: calling a digital twin model based on the network input parameters of the digital twin system to generate associated digital twin entities; sending the network data to a digital twin service application composed of digital twin entities to generate prediction and simulation data as network prediction parameters. In one example, the network input parameters refer to the relevant parameters of the network in which the digital twin system is located. For example, the network input parameters may include but are not limited to at least one of the following: network configuration, network topology, and map parameters. System administrators or system developers can deploy and activate digital twin service applications on demand in the digital twin system through the model and application orchestration management subsystem; these digital twin service applications can call digital twin models according to the network input parameters of the digital twin system in the business process to generate associated digital twin entities, and these network input parameters can be passed to the digital twin service application in the form of configuration files or instruction messages. The digital twin service application generates prediction and simulation data of network status and performance based on the collected network data as network prediction parameters.

[0046] In one embodiment, Figure 2 1 is a flow chart of another data transmission method provided by an embodiment of the present application. This embodiment is applied to the case of real-time prediction and evaluation of network status and performance. This embodiment can be executed by a physical device. Figure 2 As shown, this embodiment includes: S210-S220.

[0047] S210: Send the collected network data to the digital twin system.

[0048] S220. Receive network prediction parameters generated by the digital twin system; wherein the digital twin system is deployed at the same site as the physical device.

[0049] In one embodiment, the digital twin system and the physical device are deployed at the same site, including at least one of the following: the digital twin system and the physical device are located at the same physical location; the digital twin system and the physical device share the same physical hardware; the digital twin system is integrated into a hardware control panel inserted into the physical device.

[0050] In one embodiment, the digital twin system includes at least two digital twin models; the process of creating a digital sandbox environment includes:

[0051] Generate associated digital twin entities based on the digital twin model;

[0052] The digital twin entity builds a digital sandbox environment.

[0053] In one embodiment, the physical device runs in a digital sandbox environment and performs parameter and function verification, and performs network operations using network prediction parameters provided by the digital sandbox environment.

[0054] In one embodiment, the digital twin system includes at least two digital twin service applications; each digital twin service application uses a different digital twin entity to provide a service with specific functions to a physical device.

[0055] In one embodiment, the digital twin system includes: a model and application orchestration management subsystem for orchestrating and managing the lifecycle of digital twin models and digital twin service applications.

[0056] In one embodiment, the model and application orchestration management subsystem deploys and activates digital twin service applications on demand.

[0057] In one embodiment, sending the collected network data to the digital twin system includes:

[0058] The collected network data is periodically sent to the digital twin system.

[0059] In one embodiment, the process of generating network prediction parameters includes:

[0060] Calling the digital twin model to generate associated digital twin entities based on the network input parameters of the digital twin system;

[0061] The network data is sent to the digital twin service application composed of digital twin entities to generate prediction and simulation data as network prediction parameters.

[0062] It should be noted that for the explanations of parameters such as network prediction, network prediction parameters, digital twin system, physical device, digital twin model, digital twin service application in the data transmission method applied to physical devices, please refer to the explanations of the corresponding parameters in the above-mentioned data transmission method applied to the digital twin system, and will not be repeated here.

[0063] In one embodiment, Figure 3 Schematic diagram of the principle of a digital twin system provided in the embodiment of the present application. Figure 3 As shown, the digital twin system includes: a model and application orchestration management subsystem, a digital twin model library, and a digital twin service application pool; the digital twin model library includes multiple digital twin models (for example, digital twin model 1, digital twin model 2, digital twin model 3...); the digital twin service application pool includes multiple digital twin service applications (for example, digital twin service application 1, digital twin service application 2, digital twin service application 3...). Figure 3 The functions of each module in the digital twin system shown are as follows:

[0064] Physical equipment: refers to the network element equipment that actually exists in the physical network. Physical networks include but are not limited to cellular networks, campus enterprise networks, industrial Internet of Things, and campus LANs; network element equipment includes but is not limited to base stations, core network equipment, transmission network equipment, WLAN access points, etc.

[0065] Digital twin system: refers to a digital software system that provides twin application services to physical devices. In this application, this system can be deployed locally on the physical device, that is, deployed on the physical device hardware or in the hardware control board inserted into the physical device.

[0066] Digital twin model library: refers to a computer code library containing various digital twin models. The digital twin model in this application is a digital representation of a real-world entity or system. It carries knowledge of natural science, professional technology, industry standards and specifications, domain expert experience, mathematics or statistics, etc. Specifically, it can be a computer program code that implements a mathematical model, AI / ML model, algorithm flow, protocol flow or specific function.

[0067] Digital twin service application: refers to an application that provides specific functions and services to physical devices. Digital twin service applications can call specific digital twin models to generate several digital twin entities and combine them to achieve specified functions. The number of digital twin service applications in a digital twin system can change dynamically and be added or deleted as needed. Digital twin service applications can be deployed in containerized (such as docker, Podman), virtual machines, and executable files.

[0068] Model and application orchestration management subsystem: used to orchestrate and manage the entire life cycle of digital twin models and digital twin service applications, such as twin model registration, twin model training, testing, monitoring and updating, as well as twin service application registration, activation and deactivation.

[0069] like Figure 3 As shown in the figure, the interaction process between the digital twin system and the physical network elements includes the following steps:

[0070] Step 1. System management or developers deploy computer codes of various digital twin models in the digital twin model library through the model and application orchestration management subsystem. For example, the digital twin model may include but is not limited to one of the following: twin channel model, UE mobile trajectory twin model, UE service twin model, UE twin model, core network twin model and base station behavior model, etc.

[0071] Step 2: System management or developers deploy and activate digital twin service applications in the digital twin system through the model and application orchestration management subsystem. These digital twin service applications call digital twin models to generate digital twin entities based on network input parameters (such as network configuration, network topology, and maps) in business processes. These network input parameters are passed to digital twin service applications in the form of configuration files or command messages.

[0072] Step 3: The physical network element synchronously sends the network data collected in real time to the digital twin system. The collected network data may include but is not limited to one of the following: physical layer and high-layer measurements (such as RSRP, RSRQ, RSSI and SINR, etc.), network status and performance (such as resource utilization, scheduling information, event records, parameter changes, etc.), user measurements and event reporting (such as CQI, RI, PMI, neighboring area RSRP, etc.), user service QoS metrics (such as latency, packet loss rate, jitter, throughput, etc.).

[0073] In step 4, the digital twin entity can generate the output results defined by the model based on the network data collected in real time by the physical network elements and in accordance with the digital twin model process it complies with.

[0074] Step 5: The digital twin service application interacts with its associated digital twin entity, obtains the digital twin data required for its business process, analyzes and processes it, generates prediction and simulation data of the network status and performance defined by the digital twin service application as network prediction parameters, and sends the network prediction parameters to the physical network elements.

[0075] Step 6: The internal algorithm of the physical network element uses the prediction and simulation data returned by the digital twin service application to make internal decisions and determine subsequent execution actions (such as determining the number of spatial multiplexing streams used, coding and modulation formats, etc.).

[0076] Example 1

[0077] In this embodiment, the prediction process of the millimeter wave smart beam is described. Figure 4 This is a schematic diagram of the implementation principle of the prediction of millimeter wave smart beams based on a digital twin system provided in an embodiment of the present application. In this embodiment, the prediction is explained by taking the physical device as a gNB as an example.

[0078] There are many beams in the millimeter wave communication system, and the width of each beam is narrow. When the UE moves, beam mismatch is likely to occur, resulting in a stuck user experience or even dropped calls. The goal of this embodiment is to use twin symbiosis technology to make advance real-time predictions of the optimal beam based on the UE's mobile trajectory prediction and ray tracing model, so as to achieve a stable and smooth user experience on the UE side. In this embodiment, the intelligent beam prediction service application deployed by the digital twin system has three modules, namely UE trajectory prediction, ray tracing model and future beam prediction. Among them, the UE trajectory prediction and ray tracing model respectively perform twin modeling on the UE mobile behavior and wireless signal propagation in the physical world, which belong to the digital twin model, while the future beam prediction module uses mobile behavior and wireless channel models to decide the optimal beam.

[0079] like Figure 4 As shown, the workflow of millimeter wave intelligent beam prediction in this embodiment includes the following steps:

[0080] Step 1: gNB periodically sends a beam prediction request message, which carries the following contents: Beam ID, RSRP, CQI, SINR, user uplink beam horizontal and elevation arrival angle estimation. The periods of these parameters may be inconsistent, so the transmission period T_REQUEST of the beam prediction request message takes the minimum period that can be obtained from these parameters. When the beam prediction request message is sent, if the value of a parameter is not updated, the previous value is used.

[0081] Step 2: The UE trajectory prediction module predicts the UE position in the future (T_PREDICT) based on historical and current measurements such as Beam ID, RSRP, CQI, SINR and angle estimation. The message format sent by the UE trajectory prediction module to the ray tracing model is {TTI number: UE position}, with a total of results for the T_PREDICT time period.

[0082] Step 3, ray tracing model: Generate the channel characteristics of the corresponding position based on the input 3D map and engineering parameters (such as base station latitude and longitude, altitude and antenna orientation, etc.), as well as the predicted UE position. The message format sent by the ray tracing model to the future beam prediction is {TTI number: channel characteristics, angle information}, with a total of results within the T_PREDICT time period. The channel characteristics can be multipath, angle and power, or only the path with the strongest power. The angle information is the horizontal and vertical downlink transmission angles of the strongest path.

[0083] Step 4: Future beam prediction: The optimal beam is determined based on the predicted wireless channel at the UE location. This module is based on beam measurement data and is implemented using mathematical models or AI / ML models (such as KNN, Transformer, etc.).

[0084] The optional beams of the digital twin system in this embodiment come from the beam set table, which is related to the specific antennas used by the base station. The beams in the beam set table are divided into two different data sets, Set A and Set B.

[0085] In one example, Set A includes a coarser-grained beam, such as an SSB beam; Set B includes a finer-grained beam, such as a service beam.

[0086] In an example, Set A and Set B may be subsets belonging to the same beam set table.

[0087] In one example, Set A and Set B may be the same Beam set, different Beam sets, or partially identical Beam sets; in one example, Set A may be a subset of Set B.

[0088] The content format of the beam set table may include: a beam ID and a corresponding beam pattern (such as the azimuth and elevation angle of the beam).

[0089] The message format output by the future beam prediction module to the physical network element may be {TTI number: beam, angle information}; the angle information is the horizontal and vertical downlink transmission angles of the strongest path, and there are a total of results within the T_PREDICT time period.

[0090] The beam weight selection algorithm within the gNB refers to the optimal beam prediction results sent by the digital twin system and sends the final decision to the physical layer.

[0091] Example 2

[0092] In this embodiment, the process of online training and verification of the gNB AI MAC scheduling function is used to illustrate. Figure 5 This is a schematic diagram of the implementation principle of online training and verification of the gNB AI MAC scheduling function based on a digital twin system provided in an embodiment of the present application. In this embodiment, the physical device is a gNB as an example for prediction.

[0093] The gNB MAC scheduling function is used to dynamically select which terminals can send and receive data through the air interface, and allocate wireless air interface resources to these terminals based on various information it has, such as user channel quality, user service traffic and demand, system resource usage, historical scheduling information, and user transmission result feedback. Because there are many factors to consider, traditional solutions generally use suboptimal algorithms or combinatorial traversal solutions, but these solutions are very complex and have always been a pain point in communication systems. AI can map the relationship between input information (including status, conditions, historical results, etc.) and potential solutions in a data-driven or model-driven manner, and directly provide solutions or approximate solutions to the corresponding problems, thereby reducing the complexity of the communication system.

[0094] The performance of an AI model depends on the degree of commonality between the data distribution used for training and the data distribution used for inference. Over time, the distribution of input data used for inference may change compared to the data distribution used for training. In this case, the performance of the model will degrade over time. Therefore, AI models need to be continuously iterated and updated.

[0095] The goal of this embodiment is to use twin symbiosis technology to build a digital twin training environment used by the gNB MAC scheduling reinforcement learning algorithm, and use the network status and rewards fed back by the environment to complete the training and verification of the AI ​​MAC Agent. The AI ​​MAC scheduling training and verification service application deployed by the digital twin system in this embodiment includes two modules: model training and verification digital twin environment and AI-MAC Agent (MAC reinforcement learning).

[0096] Among them, the model training and verification digital twin environment collects data based on the physical network to generate network status and performance prediction and evaluation-related digital twin entities (such as UE trajectory generation, channel twins, service twins, twin PHY, user / network performance evaluation, etc.).

[0097] The AI ​​MAC Agent generates MAC scheduling actions based on the network status and rewards fed back by the environment. In principle, its algorithm structure should be consistent with that of the AI ​​MAC Agent in the physical network element. If a new algorithm structure is required for training, it can also be different.

[0098] like Figure 5 As shown, the workflow of online training and verification of the gNB AI MAC scheduling function in this embodiment includes the following steps:

[0099] Step 1: The physical network element sends the collected network data to the digital twin system. The collected network data may include but is not limited to at least one of the following: physical layer and high-layer measurements (such as RSRP, RSRQ, RSSI and SINR, etc.), network status and performance (such as resource utilization, scheduling information, event records, parameter changes, etc.), user measurements and event reporting (such as CQI, RI, PMI, neighboring area RSRP, etc.), user service QoS metrics (such as latency, packet loss rate, jitter, throughput, etc.).

[0100] Step 2, model training and verification The digital twin environment uses the network data collected by the physical network elements to generate the following digital twin entities: UE trajectory generation, channel twin, service twin, twin PHY and user / network performance evaluation. Among them, the twin PHY is used to simulate the user's layer 1 link performance, such as RSRP, SINR, Bler, ack / nack feedback, etc. The user / network performance evaluation is used to generate performance metrics: such as throughput, PRB utilization, user QoS indicators, energy consumption, etc.

[0101] In step 3, the AI ​​MAC Agent generates MAC scheduling actions based on the state and rewards sent by the environment in step 2, and updates its strategy or value function based on experience (i.e., the sequence of state, action, and reward) until the training process ends (such as training reaches the target accuracy, stable state, fixed number of rounds, or time limit). Specific states include but are not limited to: CQI, RI, PMI, RSRP, SINR, RSRQ, ACK / NACK, packet status (size, delay, jitter), SR, BSR, historical scheduling information (UEID, PRB, MCS, etc.). Rewards are related to optimization goals, such as throughput, Bler, or weighted values ​​of multiple goals. Actions include but are not limited to: MCS, PRB, power, precoding, number of spatial division multiplexing layers, beam information, etc.

[0102] Step 4: The digital twin system transfers the trained AI MAC model to the physical network element. The model information transferred includes but is not limited to the following:

[0103] ID: The model’s unique ID in the system. New models and old models with the same function have different IDs.

[0104] Model characteristics include: model type, model format, model size, model generation time, model KPI information and model adoption number; in one example, the model type is used to identify the model function; the model format is used to determine how to interpret and use the model file, such as json, onnx, etc.; the model size is used to describe the model file size; the model generation time is used to indicate the end time of model training; the model KPI information includes reasoning performance (such as model accuracy, etc.) and execution performance (such as model reasoning execution time); model parameters may include parameter files of neural network models or other machine learning models.

[0105] Step 5: After the AI ​​model management module in the physical network element receives the model transmitted by the digital twin system, it decides whether to update the response AI MAC model deployed in the physical network element based on its information, or the AI ​​model management module in the physical network element sends the model update request to the network management system, which decides whether to update.

[0106] In one embodiment, Figure 6 is a structural block diagram of a data transmission device provided in an embodiment of the present application. This embodiment is applied to a digital twin system. Figure 6 As shown, the data transmission device in this embodiment includes: a generating module 610 and a sending module 620.

[0107] A generating module 610 configured to generate network prediction parameters based on network data received from a physical device;

[0108] The sending module 620 is configured to feed back the network prediction parameters to the physical device; wherein the digital twin system and the physical device are deployed at the same site.

[0109] In one embodiment, the digital twin system and the physical device are deployed at the same site, including at least one of the following: the digital twin system and the physical device are located at the same physical location; the digital twin system and the physical device share the same physical hardware; the digital twin system is integrated into a hardware control panel inserted into the physical device.

[0110] In one embodiment, the digital twin system includes at least two digital twin models; the data transmission device applied to the digital twin system also includes:

[0111] A generation module is further configured to generate an associated digital twin entity based on the digital twin model;

[0112] The building module is configured to build a digital sandbox environment for the digital twin entity.

[0113] In one embodiment, the physical device runs in a digital sandbox environment and performs parameter and function verification, and performs network operations using network prediction parameters provided by the digital sandbox environment.

[0114] In one embodiment, the digital twin system includes at least two digital twin service applications; each digital twin service application uses a different digital twin entity to provide a service with specific functions to a physical device.

[0115] In one embodiment, the digital twin system includes: a model and application orchestration management subsystem for orchestrating and managing the lifecycle of digital twin models and digital twin service applications.

[0116] In one embodiment, the model and application orchestration management subsystem deploys and activates digital twin service applications on demand.

[0117] In one embodiment, the generating module 610 includes:

[0118] A collection unit configured to periodically collect network data from a physical device;

[0119] The generating unit is configured to generate prediction and simulation data based on the network data as network prediction parameters.

[0120] In one embodiment, the generating unit includes:

[0121] A first generating subunit is configured to call the digital twin model to generate an associated digital twin entity based on a network input parameter of the digital twin system;

[0122] The second generation subunit is configured to send the network data to the digital twin service application composed of digital twin entities to generate prediction and simulation data as network prediction parameters.

[0123] The data transmission device provided in this embodiment is configured to implement Figure 1 The data transmission method applied to the digital twin system in the illustrated embodiment and the data transmission device provided in this embodiment are similar in implementation principle and technical effect and will not be described in detail here.

[0124] In one embodiment, Figure 7 is a structural block diagram of another data transmission device provided in an embodiment of the present application. This embodiment is applied to a physical device. Figure 7 As shown, the data transmission device in this embodiment includes: a sending module 710 and a receiving module 720.

[0125] A sending module 710 is configured to send the collected network data to the digital twin system;

[0126] The receiving module 720 is configured to receive network prediction parameters generated by the digital twin system; wherein the digital twin system is deployed at the same site as the physical device.

[0127] In one embodiment, the digital twin system and the physical device are deployed at the same site, including at least one of the following: the digital twin system and the physical device are located at the same physical location; the digital twin system and the physical device share the same physical hardware; the digital twin system is integrated into a hardware control panel inserted into the physical device.

[0128] In one embodiment, the digital twin system includes at least two digital twin models; the process of creating a digital sandbox environment includes:

[0129] Generate associated digital twin entities based on the digital twin model;

[0130] The digital twin entity builds a digital sandbox environment.

[0131] In one embodiment, the physical device runs in a digital sandbox environment and performs parameter and function verification, and performs network operations using network prediction parameters provided by the digital sandbox environment.

[0132] In one embodiment, the digital twin system includes at least two digital twin service applications; each digital twin service application uses a different digital twin entity to provide a service with specific functions to a physical device.

[0133] In one embodiment, the digital twin system includes: a model and application orchestration management subsystem for orchestrating and managing the lifecycle of digital twin models and digital twin service applications.

[0134] In one embodiment, the model and application orchestration management subsystem deploys and activates digital twin service applications on demand.

[0135] In one embodiment, sending the collected network data to the digital twin system includes:

[0136] The collected network data is periodically sent to the digital twin system.

[0137] In one embodiment, the process of generating network prediction parameters includes:

[0138] Calling the digital twin model to generate associated digital twin entities based on the network input parameters of the digital twin system;

[0139] The network data is sent to the digital twin service application composed of digital twin entities to generate prediction and simulation data as network prediction parameters.

[0140] The data transmission device provided in this embodiment is configured to implement Figure 2 The data transmission method applied to the physical device in the illustrated embodiment is similar to the implementation principle and technical effect of the data transmission device provided in this embodiment, and will not be described in detail here.

[0141] In one embodiment, Figure 8 Schematic diagram of a communication device provided in an embodiment of the present application. Figure 8 As shown, the device provided by the present application includes: a processor 810, a memory 820 and a communication module 830. The number of processors 810 in the device can be one or more. Figure 8 In the example, a processor 810 is used. The number of memories 820 in the device may be one or more. Figure 8 A memory 820 is taken as an example. The processor 810, the memory 820 and the communication module 830 of the device can be connected via a bus or other means. Figure 8 In the example, a bus connection is used. In this embodiment, the device may be a digital twin system or a physical device.

[0142] The memory 820, as a computer-readable storage medium, can be configured to store software programs, computer executable programs, and modules, such as program instructions / modules corresponding to the device of any embodiment of the present application (for example, the generation module 610 and the sending module 620 in the data transmission device applied to the digital twin system, or the sending module 710 and the receiving module 720 in the data transmission device applied to the physical device). The memory 820 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the device, etc. In addition, the memory 820 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 820 may further include a memory remotely arranged relative to the processor 810, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0143] In the case where the communication device is a digital twin system, the above-provided device can be configured to execute the data transmission method applied to the digital twin system provided in any of the above-mentioned embodiments, and have corresponding functions and effects.

[0144] In the case where the communication device is a physical device, the device provided above can be configured to execute the data transmission method applied to the physical device provided in any of the above embodiments, and have corresponding functions and effects.

[0145] An embodiment of the present application also provides a storage medium containing computer executable instructions, which, when executed by a computer processor, are used to execute a data transmission method applied to a digital twin system, the method comprising: generating network prediction parameters based on network data received from a physical device; feeding back the network prediction parameters to the physical device; wherein the digital twin system and the physical device are deployed at the same site.

[0146] An embodiment of the present application also provides a storage medium containing computer executable instructions, which, when executed by a computer processor, are used to execute a data transmission method applied to a physical device, the method comprising: sending collected network data to a digital twin system; receiving network prediction parameters generated by the digital twin system; wherein the digital twin system and the physical device are deployed at the same site.

[0147] It will be appreciated by those skilled in the art that the term user equipment encompasses any suitable type of wireless user equipment, such as a mobile phone, a portable data processing device, a portable web browser or a car-mounted mobile station.

[0148] In general, various embodiments of the present application can be implemented in hardware or dedicated circuits, software, logic or any combination thereof. For example, some aspects can be implemented in hardware, while other aspects can be implemented in firmware or software that can be executed by a controller, microprocessor or other computing device, although the present application is not limited thereto.

[0149] Embodiments of the present application may be implemented by executing computer program instructions by a data processor of a mobile device, for example in a processor entity, or by hardware, or by a combination of software and hardware. The computer program instructions may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages.

[0150] The block diagram of any logic flow in the drawings of the present application may represent program steps, or may represent interconnected logic circuits, modules and functions, or may represent a combination of program steps and logic circuits, modules and functions. A computer program may be stored on a memory. The memory may be of any type suitable for the local technical environment and may be implemented using any suitable data storage technology, such as but not limited to read-only memory (ROM), random access memory (RAM), optical memory devices and systems (digital versatile discs (DVD) or compact disks (CD)), etc. Computer-readable media may include non-transient storage media. A data processor may be of any type suitable for the local technical environment, such as but not limited to a general-purpose computer, a special-purpose computer, a microprocessor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FGPA) and a processor based on a multi-core processor architecture.

[0151] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed by a processor, can implement the data transmission method provided in any embodiment of the present application.

[0152] In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the present application, and the programming language includes an object-oriented programming language, such as Java, Smalltalk, C++, and also includes a conventional procedural programming language, such as "C" language or similar programming language. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0153] The above are only preferred embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A data transmission method, characterized in that: Applied to digital twin systems, including: generating network prediction parameters based on network data received from the physical device; Feeding back the network prediction parameters to the physical device; wherein the digital twin system and the physical device are co-located; wherein a local secure connection is used between the digital twin system and the physical device; The digital twin system includes at least two digital twin service applications; each of the digital twin service applications uses a different digital twin entity to provide a service with a specific function to the physical device; Among them, the digital twin system includes: a model and application orchestration management subsystem, which is used to orchestrate and manage the life cycle of digital twin models and digital twin service applications.

2. The method according to claim 1, characterized in that The digital twin system and the physical device are deployed at the same site, including at least one of the following: the digital twin system and the physical device are located at the same physical location; the digital twin system and the physical device share the same physical hardware; the digital twin system is integrated into a hardware control panel inserted into the physical device.

3. The method according to claim 1, characterized in that The digital twin system includes at least two digital twin models; the method further includes: generating an associated digital twin entity based on the digital twin model; The digital twin entity builds a digital sandbox environment.

4. The method according to claim 3, characterized in that The physical device runs in the digital sandbox environment and performs parameter and function verification, and performs network operations using network prediction parameters provided by the digital sandbox environment.

5. The method according to claim 1, characterized in that The model and application orchestration management subsystem deploys and activates the digital twin service application on demand.

6. The method according to any one of claims 1 to 5, characterized in that: The generating of network prediction parameters based on network data collected from physical devices includes: Periodically collect network data from physical devices; Generate prediction and simulation data based on the network data as network prediction parameters.

7. The method according to claim 6, characterized in that The generating prediction and simulation data based on the network data as network prediction parameters includes: Calling the digital twin model to generate associated digital twin entities based on network input parameters of the digital twin system; The network data is sent to a digital twin service application composed of the digital twin entity to generate prediction and simulation data as network prediction parameters.

8. A data transmission method, characterized in that: Applied to physical devices, including: Send the collected network data to the digital twin system; Receiving network prediction parameters generated by the digital twin system; wherein the digital twin system and the physical device are co-deployed at the same site; wherein a local secure connection is adopted between the digital twin system and the physical device; The digital twin system includes at least two digital twin service applications; each of the digital twin service applications uses a different digital twin entity to provide a service with a specific function to the physical device; Among them, the digital twin system includes: a model and application orchestration management subsystem, which is used to orchestrate and manage the life cycle of digital twin models and digital twin service applications.

9. A communication device, characterized in that: include: memory, and one or more processors; The memory is configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1 to 7 or 8 above.

10. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 or 8 is implemented.

Images