A virtual-real fusion registration satellite internet digital twin deduction evaluation method

Abstract

The application discloses a kind of virtual-real fusion registration satellite internet digital twin deduction evaluation methods, comprising: S1.Satellite internet global network data is acquired, network data is collected and time synchronization interpolation is carried out, and network equipment data matrix is constructed;S2.based on network equipment data matrix, establish satellite network basic routing topology graph;S3.based on satellite network basic routing topology graph, construct by user satellite network dynamic service demand vector, establish satellite network dynamic routing model;S4.based on network dynamic routing model, establish satellite internet digital twin deduction model;S5.based on internet digital twin deduction model, introduce real data and carry out pseudo-real correction, realize the virtual-real fusion digital twin dynamic deduction of all stages;S6.multiple dimensions index and evaluation function are established, and score is calculated.The application can overcome virtual-real mapping distortion, evaluation dimension single and dynamic adaptability insufficient and other problems, complete accurate deduction evaluation to satellite internet.

Description

Technical Field

[0001] This invention relates to the field of digital twin network technology, specifically to a method, device, and storage medium for satellite internet digital twin simulation and evaluation with virtual-real fusion registration. Background Technology

[0002] With the rapid development of IoT, big data, and AI technologies, digital twin technology, as an emerging virtual modeling method, has been widely applied in fields such as industrial manufacturing, smart cities, and traffic management. Digital twins construct digital copies of physical entities, enabling real-time monitoring and prediction of their operational status, performance, and behavior, thereby optimizing decision-making and improving system efficiency.

[0003] As a new generation of integrated space-air-ground information infrastructure, satellite internet is gradually becoming a core carrier for global communications, the Internet of Things, and emergency communications.

[0004] However, its network architecture is characterized by highly dynamic topology, coexistence of multiple devices, and complex propagation environment. Currently, traditional network simulation and optimization methods are mainly used, but these methods have the following shortcomings:

[0005] Existing digital twin models mostly adopt static topology assumptions, which are difficult to adapt to the drastic topology changes caused by the high-speed movement of satellite nodes, resulting in the accumulation of deviations between virtual simulation and physical network state;

[0006] Data from heterogeneous devices such as satellites, ground stations, and terminals are asynchronous in time and space, and the lack of a unified time and space calibration mechanism restricts the accuracy of global situational awareness;

[0007] Existing simulation models are mostly based on fixed evolution rules and cannot adapt to sudden disturbances, resulting in significant differences between the simulation results and the actual network response.

[0008] To address these technical issues, this application proposes a satellite internet digital twin simulation and evaluation method based on virtual-real fusion registration. Summary of the Invention

[0009] The main objective of this invention is to provide a method for the simulation and evaluation of satellite internet digital twins through virtual-real fusion registration. By combining a virtual-real fusion framework and a multi-factor dynamic evolution simulation model, and establishing a multi-dimensional index evaluation model, this invention aims to overcome the problems of virtual-real mapping distortion, single evaluation dimension, and insufficient dynamic adaptability in existing technologies, thereby achieving accurate simulation and evaluation of satellite internet and solving the technical problems mentioned in the background.

[0010] The present invention solves the above-mentioned technical problems by adopting the following technical solutions:

[0011] A method for evaluating and simulating the digital twin of a satellite internet through virtual-real fusion registration includes:

[0012] S1. Real-time sensing and acquisition of global satellite internet network data, dividing the entire satellite network equipment into three types of network equipment: communication satellites, ground base stations, and user terminals, collecting network data of each network equipment at the current time node, performing time synchronization interpolation, and constructing a data matrix of the three types of network equipment;

[0013] S2. Based on the network device data matrix, obtain the set of device nodes, the set of link connections, and the set of link weights, and combine them with the specified characteristics of satellite internet to establish a basic routing topology map of the satellite network;

[0014] S3. Based on the basic routing topology of the satellite network, construct a dynamic routing model for the satellite network by combining the dynamic service demand vector of users on the satellite network with the dynamic service demand vector of user terminals.

[0015] S4. Based on the network dynamic routing model, establish a multi-factor evolution model, and then dynamically fuse the various factor models to establish a satellite internet digital twin simulation model;

[0016] S5. Based on the Internet digital twin simulation model, real data is introduced for simulation correction during the virtual simulation process. An adaptive fusion coefficient adjustment and correction method based on residuals is adopted to realize the dynamic simulation of virtual and real fusion digital twins throughout the entire process.

[0017] S6. Establish multi-dimensional indicators and evaluation functions to calculate the score of the current satellite internet routing mechanism in order to evaluate the simulation results.

[0018] Preferably, the set of satellite network devices set in step S1 is as follows: , The total number of devices and the data matrix of the three types of network devices are represented by a block matrix, with null values ​​used to fill in fields not present in the device type, thus forming a vector collection set of satellite internet information. ,have:

[0019]

[0020] in:

[0021] It is a 0 matrix of the corresponding dimension;

[0022] The aforementioned satellite information matrix includes the status of each satellite at a specified point in time, including parameters such as position, velocity, attitude, power, satellite signal strength, and data transmission rate.

[0023]

[0024] For the number of communication satellites, the median filter function is... Here are the 3D position coordinates of satellite i. Let i be the velocity component of satellite i. These are the Euler angles for the three attitudes of satellite i. The power value of satellite i. The signal strength of satellite i, Let i be the data transmission rate of satellite i. Each row of the matrix represents a satellite device instance, and each column corresponds to a state parameter.

[0025] This is a ground base station information matrix, including the status of each ground base station, its ground location's longitude and latitude, satellite-to-ground link communication status, ground base station signal strength, and data traffic.

[0026]

[0027] The number of ground base stations, Let i be the latitude and longitude coordinates of the ground base station. This represents the communication status between ground base station i and the satellite. It is the signal strength of ground base station i. This refers to the data traffic of ground base station i;

[0028] This is a user terminal information matrix, including information for each user terminal, such as the latitude and longitude of the user's location, the user's connection status, the user's signal strength, and the user's data usage.

[0029]

[0030] For the number of user terminals, Let be the latitude and longitude of the location of user terminal i. The connection status of user terminal i. For the signal strength of user terminal i, This refers to the data usage of user terminal i.

[0031] Preferably, for vector acquisition sets For equipment Data completion using time interpolation is available as follows:

[0032]

[0033] in, For newly inserted data at sampling time point t, The sampling time point of device k at the previous time point, Let k+1 be the sampling time point for the device at the next time step. The interval to be interpolated for the device. The sampling time point of the device at the previous moment Data values, The next sampling time point for the device The data value.

[0034] Preferably, the specific operation procedure for establishing the basic routing topology map of the satellite network in step S2 includes:

[0035] S21. Based on the collected satellite network data feature matrix, establish a basic satellite network topology map and define a set of network nodes. ,have:

[0036]

[0037] in , , These represent the sets of satellites, ground base stations, and user terminals, respectively. , , These respectively represent satellite equipment, ground base station equipment, and user terminal equipment;

[0038] There is a set of links It includes links to all connected nodes at the current point in time;

[0039] S22. Introducing satellite orbital dynamics characteristics, and using orbital parameterization of satellite motion, we have:

[0040]

[0041]

[0042]

[0043] in, This indicates the link between the satellite and the ground base station. This indicates the link between the satellite and the user terminal. This indicates the link between the ground base station and the user terminal; Minimum elevation angle, For the satellite's position, For the location of the user terminal or ground base station, For the Earth's radius, For the semi-major axis of the track, For eccentricity, , As the perturbation force component, For the near-point angle, It is the average speed of motion;

[0044] S23. To address the satellite's motion characteristics, Doppler effect compensation is introduced, and a Doppler frequency shift correction is established:

[0045]

[0046]

[0047] in, This refers to the dynamic distance between stars. For the satellite's orbital radius, Let be the relative velocity between the two nodes. For the direction of motion, The center frequency of the carrier. For Doppler frequency shift, The speed of light;

[0048] S24. For links between N devices, construct links at any time... Link weight influence matrix ,have:

[0049]

[0050]

[0051] in, Let i be the link weight between devices i and j. For satellite equipment collection, For ground base station set, For user terminal set, It is any distinct device belonging to these three sets. It is the square of the distance between the two devices. and It is a specified constant used to adjust the impact of distance on the link. , , The distance influence constant between different sets, It is a constant representing the impact of weather on signal quality. This is the signal attenuation value due to weather conditions. It is the user terminal antenna gain. For link load, For the maximum allowable load, It refers to the degree to which the environment obstructs the signal. Maximum allowable blocking;

[0052] S25. Construct a basic network topology diagram with satellite network characteristics. ,have:

[0053]

[0054] in, It is an edge set. It is a weight set.

[0055] Preferably, the specific operation process of step S3 includes:

[0056] S31. Define device nodes in a satellite network To other device nodes Baseline business requirements Following a Markov process, we have:

[0057]

[0058] in, From node arrive In time bandwidth requirements, From node arrive In time latency requirements, From node arrive In time Link quality requirements;

[0059] S32. Define the dynamic business requirement vector matrix as follows:

[0060]

[0061]

[0062] The index portion indicates that, in terms of latitude... A Gaussian distribution centered on the core describes the spatial distribution of business requirements. The Earth's rotation period is used to reflect the diurnal rhythm of demand. Used to capture the tidal effect of demand. This is the vector sum of the movement velocities of the nodes, reflecting the movement characteristics of the terminal nodes. This represents the dynamic service demand matrix for the entire satellite network.

[0063] S33. Construct a routing model that integrates the dynamic nature of inter-satellite links with the spatiotemporal characteristics of dynamic service requirements. ,have:

[0064]

[0065] in, The distance sensitivity coefficient is a time-varying factor, dynamically adjusted by the link signal-to-noise ratio. When SNR decreases, increase the weight of the distance factor; This refers to the dynamic inter-satellite movement distance. Effective bandwidth is affected by the user terminal antenna elevation angle and load. As a weighting factor, Satellite speed The higher the elevation, the stronger the Doppler sensitivity. Including the second-order effects of satellite acceleration, accurately representing frequency shift transient characteristics, SLA is a service level quality standard. A dynamic business demand matrix; The remaining connection / disconnection time of the link is calculated based on the orbital forecast. The rate of change of relative azimuth angle between satellites. The maximum azimuth angle is:

[0066] .

[0067] Preferably, the specific operation process of step S4 includes:

[0068] S41. Construct a satellite network link evolution model with multi-factor model fusion. Through a hierarchical factor fusion framework, the physical communication channel status of satellite nodes and user terminal nodes in the satellite network, node failures in the satellite network, and user demand change factors are uniformly modeled. The link evolution model includes physical layer factor model, network layer factor model, and application layer factor model.

[0069] S42. Based on the established three factor models, a weighted fusion is performed to form a satellite network link evolution model, which is:

[0070]

[0071] in, For physical factor model, For network layer factor model, For the application of layer factor model, The factor weights are used to dynamically adjust the influence of each factor model on the evolutionary model.

[0072] Preferably, the process of establishing the factor model includes:

[0073] S421. Divide the transmission channel into a space propagation process from satellite to ground receiver and a ground-based propagation process, and establish a satellite physical layer communication model, which includes:

[0074]

[0075] This represents the signal power attenuation and signal dispersion experienced by the signal as it passes through the vacuum layer and ionosphere to reach the Earth's surface. This represents the large-scale fading portion of the channel. This indicates small-scale fading and shadowing fading affecting the signal;

[0076] S422. When equipment failure occurs in a satellite network, a network failure factor model is established, and failure indicator variables are defined, including:

[0077]

[0078] Among them, when the relay satellite node or the ground base station relay node fails, the following is set: The faulty node is defined as follows: The number of nodes connected to this faulty node is... Then update the network connection set as follows:

[0079]

[0080]

[0081] and It is the node coordinate position. It is the visible path between two nodes. This is a fault indicator variable; when a node fails, the device connection function... =0, Link weight;

[0082] When a user terminal node fails, the original demand matrix will be updated, and the demand vector of the failed node will be transferred to the nearest idle node.

[0083]

[0084] For nodes and exist The demand vector at any given moment The faulty node;

[0085] S423. Establish an application-layer user demand change factor model and define a time-varying function for user demand:

[0086]

[0087] in For nodes and exist The demand vector at any given moment To meet the requirements of a fixed baseline, For event-driven items, The disturbance is random and follows a normal distribution.

[0088] S424. Establish a time series model of user demand factors:

[0089]

[0090] in, These are forward projection predictions. … These are autoregressive coefficients, obtained by fitting historical data. … It is the moving average coefficient. It is a random time-of-flight error.

[0091] Preferably, the specific operation process of step S5 includes:

[0092] S51. Perform timestamp alignment by calibrating the timestamps of real data and virtual simulation using satellite synchronization signals, establishing a mapping relationship between real network nodes and digital twins, and defining the simulation mapping relationship between real state data and virtual model state, including:

[0093]

[0094] in, This is real-state data. This is a virtual model state. For time synchronization error, For time-varying digital twin model coefficients, For the derivation model function;

[0095] S52. Define the difference between the virtual and real states in the initial state as:

[0096]

[0097]

[0098] in, This represents the difference between the virtual and real states. For parameter weights, This is real-state data. This is virtual state data. The deviation threshold, Based on the threshold, The variance over a short period of time. and As an adjustable parameter, when > At that time, the simulation calibration mechanism is triggered;

[0099] S53. Calculate the normalized residuals between the virtual model's derived state vector and the real system's state vector, and we have:

[0100]

[0101] in, For the actual system state vector, This represents the virtual inference state vector;

[0102] S54. Construct dynamic fusion coefficients, namely:

[0103]

[0104] in, The upper limit threshold of the residual. The lower limit threshold;

[0105] S55. Based on the fusion coefficient, a virtual state correction function is set to optimize the parameters of the digital twin inference model, which includes:

[0106]

[0107] in, This represents the amount of change in virtual data over a short period of time.

[0108] Preferably, the specific operation process of step S6 includes:

[0109] Define a time-varying weight function over the entire simulation period T to represent the dominant evaluation objective at different simulation stages, and construct a fusion evaluation function corresponding to the evaluation objective, as follows:

[0110]

[0111] in, The score for routing scheme P, This indicates different stages in the simulation and evaluation cycle. For stage transition rate, For the synchronization error evaluation function, This is the steady-state performance evaluation function for routing. This is a function for evaluating route failure recovery.

[0112] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0113] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0114] As can be seen from the above technical solution, this invention provides a method for digital twin simulation and evaluation of satellite internet based on virtual-real fusion registration. Compared with the prior art, this invention has the following advantages:

[0115] 1. This invention addresses the multi-source data characteristics of satellite internet by constructing different data information matrices based on satellites, ground stations, and user terminals. This enables more effective capture of the spatiotemporal characteristics of satellite network nodes, providing a rich and accurate data foundation for subsequent network analysis and modeling.

[0116] 2. This invention establishes a multi-factor fusion link evolution model, utilizing the integration of physical layer factors, network layer factors, and application layer factors to create a fusion link extrapolation model. This facilitates accurate twin extrapolation of the entire satellite network, providing strong support for satellite network management and optimization. It also helps to detect satellite network faults in advance, optimize satellite network resource allocation, and ensure stable network operation and service quality.

[0117] 3. This invention establishes a multi-dimensional index evaluation model, utilizes a multi-dimensional, multi-stage twin inference evaluation model, and performs inference evaluation on the digital twin inference model. This enables early warning and fault analysis of network nodes with abnormal operating status, optimizes satellite network structure design and operating parameters, and improves the performance and efficiency of the entire satellite network.

[0118] It should be understood that the descriptions in this section are not intended to identify key or essential features of embodiments of the invention, nor are they intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Of course, implementing any product of the invention does not necessarily require achieving all of the advantages described above simultaneously. Attached Figure Description

[0119] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0120] Figure 1 This is a schematic diagram of the overall process of the present invention;

[0121] Figure 2 This is a schematic diagram of the simulation platform interface of the present invention;

[0122] Figure 3 This is a schematic diagram of the link network topology of the present invention. Detailed Implementation

[0123] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0124] For details in the embodiments, please refer to Figures 1 to 3 .

[0125] like Figure 1 As shown. The satellite internet digital twin simulation and evaluation method based on virtual-real fusion registration proposed in this embodiment of the invention includes the following steps:

[0126] Step 1: Real-time sensing and acquisition of global satellite internet network data. The entire satellite network equipment is divided into three types of network equipment: communication satellites, ground base stations, and user terminals. Network data at the current time node is collected for each network equipment, including location data, signal strength, power, data transmission rate, and network behavior data. Time synchronization interpolation is performed to construct a data matrix for the three types of network equipment.

[0127] The collection of satellite internet devices is represented as , The total number of devices in a satellite internet system includes communication satellites, ground base stations, and user terminals; assuming the number of communication satellites is... The status of each satellite at a given point in time includes parameters such as position, velocity, attitude, power, satellite signal strength, and data transmission rate, forming a communication satellite matrix. :

[0128]

[0129] Among them, the median filtering function The 3D position coordinates of the satellite. For the satellite's velocity components, Euler angles for three different orientations, This is the power value. For signal strength, This represents the data transmission rate; each row of the matrix represents a satellite device instance, and each column corresponds to a status parameter.

[0130] Assume the number of ground base stations is The status of each ground base station includes its ground location's longitude and latitude, the communication status of the satellite-to-ground link, the ground base station's signal strength, and data traffic, thus constructing a ground base station information matrix. :

[0131]

[0132] in, The latitude and longitude coordinates of the ground base station, and Communication status between the ground station and the satellite. It refers to the signal strength of the ground base station. It is ground station data traffic;

[0133] Assume the number of user terminals is Each user terminal's information includes the user's location (latitude and longitude), connection status, signal strength, and data usage, forming a user terminal information matrix. :

[0134]

[0135] in, The latitude and longitude of the user terminal's location. For the user's connection status, For the signal strength of the user terminal, User data usage;

[0136] Furthermore, the feature matrices of the three types of devices are unified into a large matrix, using a block matrix format, and null values ​​are used to fill in fields that are not present in the device type, thus forming a satellite internet information vector collection set. :

[0137]

[0138] in, For the aforementioned satellite information matrix, For ground base station information matrix, For user terminal information matrix, It is a 0 matrix of the corresponding dimension;

[0139] Step 2: Based on the collected satellite network device data matrix, obtain the device node set, link connection set, and link weight set. Taking into account the unique physical layer characteristics, network characteristics, and service requirements of satellite Internet, model each characteristic in detail and establish the basic routing topology of satellite network.

[0140] Establish a basic satellite network topology map, which is constructed from links between satellites and between satellites and the ground. This map integrates the inter-satellite and satellite-to-ground topology connections within the network, and defines a set of network nodes. :

[0141]

[0142] in , , These represent the sets of satellites, ground base stations, and user terminals, respectively. , , These respectively represent satellite equipment, ground base station equipment, and user terminal equipment;

[0143] Link set This includes links connecting all nodes at the current point in time. These links can be inter-satellite links, satellite-to-ground links, or links between any devices in the network. Due to the unique physical layer characteristics of satellite internet, satellite orbital dynamics are introduced, and satellite motion is parameterized using six orbital root numbers.

[0144]

[0145]

[0146]

[0147] in, This indicates the link between the satellite and the ground base station. This indicates the link between the satellite and the user terminal. This indicates the link between the ground base station and the user terminal; Indicates satellite With base station In time Visible, and the elevation angle is greater than , Indicates satellite With user terminal In time Visible, and the elevation angle is greater than , The coverage radius of the base station. For base station coordinates, For user coordinates, Minimum elevation angle, For the satellite's position, For the location of the user terminal or ground base station, For the Earth's radius,

[0148] For the semi-major axis of the track, For eccentricity, , As the perturbation force component, It is a near-point angle;

[0149] To address the satellite's motion characteristics, Doppler effect compensation is introduced to establish a Doppler frequency shift correction:

[0150]

[0151]

[0152] in, This refers to the dynamic distance between stars. For the satellite's orbital radius, Let be the relative velocity between the two nodes. For the direction of motion, The center frequency of the carrier.

[0153] Next, considering the link connection characteristics between different satellite links, a link weight matrix is ​​constructed for the three main elements in satellite internet: ground base stations, satellite equipment, and user terminals. This considers the environmental conditions of the satellite-to-ground links, obstacles between the satellite and ground stations, and the heterogeneity of user terminal equipment, thus constructing a link weight matrix for any given time period. Link weight influence matrix :

[0154]

[0155]

[0156] in, For satellite equipment collection, For ground base station set, For user terminal set, It is any distinct device belonging to these three sets. It is the square of the distance between the two devices. and It is a constant used to adjust the impact of distance on the link. , , The distance influence constant between different sets, It is a constant representing the impact of weather on signal quality. This is the signal attenuation value due to weather conditions. It is the user terminal antenna gain. For link load, For the maximum allowable load, It refers to the degree to which the environment obstructs the signal. Maximum allowable blocking;

[0157] By combining the node set, edge set, and weight set described above, a basic network topology diagram with satellite network characteristics is constructed. :

[0158]

[0159] in, It is an edge set. It is a weight set.

[0160] Step 3: Construct a dynamic service demand vector that combines users' bandwidth requirements, latency requirements, and link quality requirements for the satellite network; combine the dynamic service demand vector of user terminals in the satellite Internet, and fully consider the high-speed motion characteristics of low-orbit satellites, the intermittent on / off characteristics of inter-satellite links, and the mobility characteristics of user terminals to establish a dynamic routing model for the satellite network during the modeling process.

[0161] A dynamic routing model for satellite networks was established, and a routing model that integrates the dynamic nature of inter-satellite links with the spatiotemporal characteristics of dynamic service requirements was constructed.

[0162]

[0163] The first polynomial represents the dynamic distance term of the satellite nodes. The distance sensitivity coefficient is a time-varying factor, dynamically adjusted by the link signal-to-noise ratio. When SNR decreases, increase the weight of the distance factor; This refers to the dynamic inter-satellite movement distance. The effective bandwidth is affected by the user terminal antenna elevation angle and load; the second polynomial is the Doppler frequency shift rate term. As a weighting factor, Satellite speed The higher the elevation, the stronger the Doppler sensitivity. The second-order effect, incorporating satellite acceleration, accurately represents the transient characteristics of frequency shift. The third polynomial represents the dynamic service requirement term, and SLA stands for Service Level Quality Standard. A dynamic business demand matrix; The remaining connection / disconnection time of the link is calculated based on the orbital forecast. The rate of change of relative azimuth angle between satellites;

[0164] Step 4: Based on the network dynamic routing model, establish a multi-factor evolution model, and then dynamically fuse the various factor models to establish a satellite internet digital twin simulation model.

[0165] Based on the established three factor models, a weighted fusion is performed to form a satellite network link evolution model:

[0166]

[0167] in, For physical factor model, For network layer factor model, For the application of layer factor model, The factor weights are used to dynamically adjust the influence of each factor model on the evolutionary model.

[0168] Step 5: Combining virtual and real fusion, real data is introduced for simulation correction during the virtual simulation process. An adaptive fusion coefficient adjustment and correction method based on residuals is adopted to realize the dynamic simulation of virtual and real fusion digital twins throughout the entire process, which maintains both the characteristic inertia of virtual simulation and the authenticity.

[0169] A virtual-real fusion parameter adjustment method based on residual adaptive fusion coefficient adjustment is proposed. First, the normalized residuals of the virtual model's inferred state vector and the real system's state vector are calculated. When the residuals between the real model state and the inferred model result are large, the proportion of real data introduced is increased, and the coefficients are also increased, thereby achieving real data simulation correction. When the residuals between the real model state and the inferred model result are small, the proportion of real data introduced is decreased, and the coefficients are also decreased, thereby maintaining the characteristic inertia of virtual inference.

[0170]

[0171] in, For the actual system state vector, This represents the virtual inference state vector;

[0172] Further design of dynamic fusion coefficients:

[0173]

[0174] in, The upper limit threshold of the residual. To set the lower threshold, a Sigmoid response function is used to smoothly adjust the fusion coefficients. Simultaneously, to suppress coefficient oscillations, a historical inertia term is added.

[0175]

[0176] When the residual between the actual model state and the inferred model result is large, the proportion of real data introduced increases, and the coefficient also increases, thereby achieving real data simulation correction. When the residual between the actual model state and the inferred model result is small, the proportion of real data introduced decreases, and the coefficient also decreases, satisfying the monotonically increasing relationship between the fusion coefficient and the residual.

[0177]

[0178] Furthermore, based on the fusion coefficient, a virtual state correction function is proposed, and the parameters of the digital twin inference model are optimized to achieve fusion calibration.

[0179]

[0180] in, This refers to the amount of change in virtual data over a short period of time. These are the fusion correction coefficients mentioned above;

[0181] A parameter update and optimization function for digital twin inference models is proposed:

[0182]

[0183] in, The coefficients of the twin model at the current moment, These are the coefficients of the historical extrapolation model. The confidence weight matrix is ​​the real data. For residuals, The regularization coefficient is used to prevent overfitting. The coefficients are the optimal twin inference model coefficients;

[0184] Step 6: Based on the simulation model, establish a multi-dimensional index and a multi-stage integrated evaluation function to comprehensively evaluate the satellite internet routing simulation results and calculate the current satellite internet routing mechanism score.

[0185] Design a multi-stage, multi-dimensional fusion evaluation function, and define a time-varying weight function over the entire simulation period T to represent the dominant evaluation objectives of different simulation stages. The fusion evaluation function is defined as follows:

[0186]

[0187] in, The score for routing scheme P, This indicates different stages in the simulation and evaluation cycle. For stage transition rate, For the synchronization error evaluation function, This is the steady-state performance evaluation function for routing. This is the routing failure recovery evaluation function; thus, the evaluation score of satellite internet under this routing mechanism is obtained:

[0188]

[0189] in, It refers to the score under steady state, and the goal is to find the maximum value. It is a dynamic routing model;

[0190] Therefore, the fusion evaluation function serves as a multi-dimensional comprehensive score for routing scheme P. By quantifying the performance of the current path in key indicators such as latency, reliability, synchronization accuracy, and route recovery time, it directly reflects the merits of the routing scheme and obtains the final inference score of the routing scheme. The relationship between the score and the routing model can be summarized as follows: dynamic business requirements generate routing decisions, twin inference is performed through a digital twin model, and the fusion evaluation function calculates the routing score, transforming abstract indicators such as latency, reliability, and synchronization accuracy into calculable score values, avoiding single-dimensional bias. At the same time, the stage weights are automatically adjusted with the inference cycle to match the characteristics of the network life cycle.

[0191] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0192] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0193] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the virtual-real fusion registration satellite internet digital twin inference and evaluation methods described in the above embodiments.

[0194] It is understood that the system provided in the embodiments of the present invention corresponds to the method provided in the embodiments of the present invention, and the explanation, examples and beneficial effects of the relevant content can be referred to the corresponding parts of the above method.

[0195] This application also provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, communication interface, and memory communicate with each other via the communication bus.

[0196] Memory, used to store computer programs;

[0197] When the processor executes the program stored in the memory, it implements the above-mentioned satellite internet digital twin simulation and evaluation method of virtual-real fusion registration.

[0198] The communication bus mentioned in the above-mentioned electronic devices can be a standard bus for interconnecting peripheral components or an extended industrial standard structure bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc.

[0199] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0200] The memory may include random access memory or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0201] The processors mentioned above can be general-purpose processors, including central processing units, network processors, etc.; they can also be digital signal processors, application-specific integrated circuits, field-programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0202] It should also be noted that electronic devices include terminal devices, which can also be called terminals, user equipment, mobile stations, mobile terminals, etc. Terminal devices can be mobile phones, smart TVs, wearable devices, tablets, computers with wireless transceiver capabilities, virtual reality terminal devices, augmented reality terminal devices, wireless terminals in industrial control, wireless terminals in autonomous driving, wireless terminals in remote surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, and so on. The embodiments of this application do not limit the specific technologies or device forms used in the terminal devices.

[0203] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium, or a semiconductor medium (e.g., a solid-state drive).

[0204] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0205] Furthermore, it should be noted that if any directional indication (such as up, down, left, right, front, back, etc.) is involved in the embodiments of the present invention, the directional indication is only used to explain the relative positional relationship and movement of each component in a specific posture. If the specific posture changes, the directional indication will also change accordingly.

[0206] Furthermore, the meaning of "and / or" throughout the text includes three parallel solutions. Taking "A and / or B" as an example, it includes solution A, solution B, or a solution that simultaneously satisfies A and B. Additionally, in the embodiments of this invention, "multiple" refers to two or more. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

Claims

1. A satellite internet digital twin inference evaluation method for virtual-real fusion registration, characterized in that, The method comprises the following steps: S1. Real-time sensing and acquiring satellite Internet global network data, dividing the entire satellite network equipment into three network devices of communication satellite, ground base station and user terminal, collecting network data of each network device at the current time node, and performing time synchronization interpolation to construct three network device data matrices; S2. Based on the network device data matrix, obtain the device node set, link connection set and link weight set, and establish a satellite network basic routing topology graph combined with the specified characteristics of the satellite Internet; The specified characteristics of the satellite Internet include satellite orbit dynamics characteristics and Doppler effect characteristics; S3. Based on the satellite network basic routing topology graph, a dynamic service demand vector of the satellite network by the user is constructed, and a satellite network dynamic routing model is established combined with the dynamic service demand vector of the user terminal; S4. Based on the network dynamic routing model, a multi-factor evolution model is established, and each factor model is dynamically fused to establish a satellite Internet digital twin deduction model; S5. Based on the Internet digital twin deduction model, in the process of virtual deduction, real data is introduced for pseudo-correction, an adaptive fusion coefficient adjustment correction method based on residual error is adopted, and full-stage virtual-real fusion digital twin dynamic deduction is realized; S6. A multi-dimensional index and evaluation function are established to calculate the score of the current satellite Internet routing mechanism for deduction result evaluation; The specific operation process of the step S5 comprises: S51. Time stamp alignment is performed, the time stamps of the real data and the virtual deduction are calibrated through the satellite synchronization signal, the mapping relationship between the real network node and the digital twin is established, the deduction mapping relationship between the real state data and the virtual model state is defined, and the real state data and the virtual model state are defined as follows: wherein, is real state data, is virtual model state, is time synchronization error, is time-varying digital twin model coefficients, is a deduction model function; S52. The initial state difference value of the virtual state is defined as follows: wherein, is a virtual-real state difference value, is a parameter weight, is a real state data, is a virtual state data, is a bias threshold value, is a base threshold value, is a short-term time difference variance, and is an adjustable parameter, when > a pseudo-real calibration mechanism is triggered. S53. The normalized residual error of the virtual model deduction state vector and the real system state vector is calculated, and the normalized residual error is calculated as follows: wherein, is the real system state vector, is the virtual state vector; S54. A dynamic fusion coefficient is constructed, and the dynamic fusion coefficient is calculated as follows: wherein is a residual upper threshold value, is a lower threshold value, ; S55. According to the fusion coefficient, a virtual state correction function is set to optimize the parameters of the digital twin deduction model, and the virtual state correction function is calculated as follows: wherein, is the amount of change in virtual data in a short time.

2. The virtual-real fusion registered satellite internet digital twin deductive evaluation method of claim 1, wherein, The three types of network device data matrix in the S1 step is filled with empty values in the fields that do not have the device types by adopting the form of a block matrix to constitute a vector collection set of satellite Internet information , wherein, is a 0 matrix of the corresponding dimension, is a satellite information matrix, is a ground base station information matrix, is a user terminal information matrix, including information of each user terminal; For vector collection set For device With time interpolation data, there are: in, For newly inserted data at sampling time point t, The sampling time point of device k at the previous time point, Let k+1 be the sampling time point for the device at the next time step. The interval to be interpolated for the device. The sampling time point of the device at the previous moment Data values, The next sampling time point for the device The data value.

3. The virtual-real fusion registered satellite internet digital twin deductive evaluation method of claim 1, wherein, The specific operation process of establishing the satellite network basic routing topology graph in the step S2 comprises: S21. According to the satellite network data feature matrix collected, a satellite network basic topology graph is established, and a node set of a network is set , there are: wherein 、 、 represent a set of satellites, a set of ground base stations and a set of user terminals, respectively, 、 、 represent a satellite device, a ground base station device and a user terminal device, respectively; There is a set of links containing the links of all connected nodes at the current point in time; S22. The satellite orbit dynamics characteristics are introduced, and the satellite motion is parameterized, and the satellite motion is parameterized as follows: wherein, denotes a link of a satellite with a ground base station, denotes a link of a satellite with a user terminal, denotes a link between a ground base station and a user terminal; S23. According to the motion characteristics of the satellite, the Doppler effect compensation is introduced, and the Doppler frequency shift correction is established, and the Doppler frequency shift correction is calculated as follows: wherein, is the inter-satellite dynamic distance, is the satellite motion radius, is the relative motion velocity of two nodes, is the motion direction, is the carrier center frequency, is the Doppler frequency offset, is the light speed; S24. Constructing the link weight influence matrix at any time S25. Constructing the link weight influence matrix at any time , we have: wherein, is the link weight between devices i and j; S25. Constructing a base network topology map with satellite network characteristics , there are: wherein, is a set of edges, is a set of weights.

4. The virtual-real fusion registered satellite internet digital twin deductive evaluation method of claim 1, wherein, The specific operation process of the step S3 comprises: S31. Defining a device node in a satellite network to other device nodes of baseline service requirements subject to a Markov process, have: wherein, is a bandwidth requirement from node to at time , is a latency requirement from node to at time , is a link quality requirement from node to at time ; S32. The dynamic service demand vector matrix is defined as follows: wherein, is the earth rotation period, used to reflect the circadian rhythm of demand, is the tidal effect used to capture the demand, is the vector sum of the moving speed of nodes, embodying the moving characteristics of terminal nodes, is the dynamic traffic demand matrix of the entire satellite network, exponential represents the Gaussian distribution centered at the latitude is used to describe the spatial distribution of traffic demand; S33. Constructing a routing model that fuses the dynamicity of interstellar links with the spatiotemporal characteristics of dynamic traffic demands There are: where, is the time-varying distance sensitivity coefficient, dynamically adjusted by the link SNR, When SNR decreases, the distance factor weight increases; is the inter-satellite dynamic moving distance, is the effective bandwidth, affected by the user terminal antenna elevation angle and load, is the weight factor, The higher the satellite speed , the stronger the Doppler sensitivity, contains the second-order effect of satellite acceleration, accurately represents the frequency shift transient characteristics, and SLA is a service level quality standard, is the dynamic service demand matrix; is the remaining on-off time of the link calculated according to the orbit prediction, is the relative azimuth angle change rate between satellites, is the maximum azimuth angle.

5. The virtual-real fusion registered satellite internet digital twin deductive evaluation method of claim 1, wherein, The specific operation process of the step S4 comprises: S41. A satellite network link evolution model based on multi-factor model fusion is constructed, the physical communication channel state of the satellite node and the user terminal node in the satellite network, the node fault in the satellite network and the user demand change factor are uniformly modeled through a hierarchical factor fusion framework, and the link evolution model comprises a physical layer factor model, a network layer factor model and an application layer factor model; S42. According to the three factor models, the satellite network link evolution model is obtained by weighted fusion, and the satellite network link evolution model is calculated as follows: wherein, is a physical factor model, is a network layer factor model, is an application layer factor model, is a factor weight, dynamically adjusting the influence of each factor model on the evolution model.

6. The virtual-real fusion registered satellite internet digital twin inference evaluation method of claim 5, wherein, The establishment process of the factor model comprises: S421. The transmission channel is divided into a space propagation process of satellite to ground receiving end and a ground mobile propagation process, a satellite physical layer communication model is established, and there are: denotes the signal power attenuation and the signal dispersion the signal experiences when passing through the vacuum layer, the ionosphere to the earth surface, wherein denotes the large scale fading part of the channel, denotes the small scale fading and shadowing the signal experiences; S422. When a device in the satellite network fails, a network failure factor model is established, and a failure indication variable is defined, and there are: Wherein, when the relay forwarding satellite node or the ground base station forwarding node fails, set The number of nodes connected to the fault node is Then update the network connection set to: and is the node coordinate position, is the visual path between two nodes, is a fault indicator variable, when a node fails, the device connection function is 0, is the link weight; When a user terminal node fails, the original demand matrix is updated, and the demand vector of the failed node is transferred to the nearest idle node: is a node and is a demand vector at time instant, is a faulty node; S423. An application layer user demand change factor model is established, and a user demand time-varying function is defined: where is the node and at the demand vector at time instant is the fixed baseline demand, is the event-driven term, is the random disturbance, following a normal distribution; S424. A user demand factor time series model is established: wherein, is a forward extrapolation prediction, … is an autoregressive coefficient, fitted from historical data, … is a moving average coefficient, is a random time error.

7. The virtual-real fusion registered satellite internet digital twin inference evaluation method of claim 1, wherein, The specific operation process of the S6 step includes: A time-varying weight function is defined to represent the dominant evaluation target in different deduction stages within the entire deduction period T, and a fusion evaluation function corresponding to the evaluation target is constructed, and there are: wherein, is a score for the routing scheme P, denotes different phases in the deduction evaluation period, is a phase transition rate, is a synchronization error evaluation function, is a routing steady-state performance evaluation function, is a routing failure recovery evaluation function.

8. A computer-readable storage medium, characterized in that, The computer program is stored, and the computer program is executed by the processor to make the processor execute the steps of the method in any one of claims 1 to 7.

9. A computer device, comprising: The computer program is stored, and the computer program is executed by the processor to make the processor execute the steps of the method in any one of claims 1 to 7.

Images