A method and system for QoS calculation of low-Earth orbit satellite communication networks based on time-varying topology awareness

By using a time-varying topology sensing method to acquire the orbital position information of low-Earth orbit satellites, construct dynamic network connectivity relationships, and use a multilayer perceptron model to predict QoS indicators, the accuracy and dynamic adaptability issues of multidimensional QoS indicator calculation in low-Earth orbit satellite communication networks are solved, achieving higher-precision service quality assessment.

CN123074559APending Publication Date: 2026-11-10STATE GRID FUJIAN ELECTRIC POWER CO LTD +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611183282.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-05
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing technologies struggle to accurately calculate multidimensional QoS metrics in low-Earth orbit satellite communication networks, especially under the influence of dynamic topology changes and satellite motion, resulting in insufficient accuracy and dynamic adaptability in QoS calculations.

Method used

A time-varying topology sensing method is adopted to acquire the orbital position information of low-Earth orbit satellites, construct dynamic network connection relationships, extract time-varying topology features, and use a multilayer perceptron model to predict QoS indicators. Combined with the orbital propagation model and spatial geometric relationships, dynamic calculation of low-Earth orbit satellite communication networks is realized.

Benefits of technology

It significantly improves the QoS calculation capability under dynamic network conditions, enhances the dynamic adaptability of satellite service quality, and improves the calculation accuracy of QoS indicators.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123074559A_ABST
    Figure CN123074559A_ABST
Patent Text Reader

Abstract

This invention proposes a method and system for calculating QoS of low-Earth orbit (LEO) satellite communication networks based on time-varying topology awareness, comprising: Step S1: acquiring two-line LEO satellite orbital position information, and using the SGP4 orbital propagation model to calculate the spatial position of each LEO satellite within a continuous time window, constructing a dynamic time-series sequence of LEO satellites for subsequent QoS calculation; Step S2: setting the ground terminal position, proposing a candidate service satellite mechanism, and constructing a dynamic network connection relationship based on the spatial geometric relationship between this mechanism and the ground terminal; Step S3: using the dynamic network connection relationship within each time window as a time-varying topology structure, and extracting parameter features from the continuous time-varying topology structure; Step S4: performing simulation based on the time-varying topology awareness features described in S3 to obtain a series of QoS tags, and merging the tags to generate a LEO satellite dataset; Step S5: dividing the LEO satellite dataset and inputting it into a multilayer perceptron model to calculate QoS indicators.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention proposes a method and system for calculating QoS of low-Earth orbit satellite communication networks based on time-varying topology awareness, which relates to the technical field of low-Earth orbit satellite communication networks and intelligent communication. Background Technology

[0002] Low Earth Orbit (LEO) satellite communication networks refer to communication systems composed of satellites orbiting in low Earth orbit and related network control equipment. Compared with traditional terrestrial networks, LEO satellite communication networks have advantages such as wide coverage and suitability for communication in remote areas, making them of significant application value in communication scenarios.

[0003] Because low-Earth orbit satellites move at high speeds, their positions relative to ground terminals change continuously over time, resulting in variations in the visible satellite array and the communication capacity of each link at different times. Therefore,

[0004] Low Earth orbit (LEO) satellite communication networks are inherently time-varying topologies. If static topology or single-link metrics are used...

[0005] QoS calculations are difficult to accurately describe the dynamic communication network of low-Earth orbit satellites, thus affecting the accuracy of QoS calculations.

[0006] Quality of Service (QoS) is commonly used to measure the service capability of a communication network for services. Common metrics include end-to-end latency, jitter, throughput, packet loss rate, and link availability. In low-Earth orbit (LEO) satellite communication networks, QoS is affected not only by propagation distance but also by a combination of factors such as satellite motion and link capacity. Existing methods typically focus on estimating static, single metrics, making it difficult to integrate the dynamic geometry of LEO satellite orbits, topology changes, and multi-dimensional QoS metrics into a unified computational framework.

[0007] Therefore, there is a need for a method that can accurately describe the dynamic changes of low-Earth orbit satellite communication networks and calculate multi-dimensional QoS indicators, so as to improve the accuracy and dynamic adaptability of low-Earth orbit satellite communication network service quality assessment and prediction. Summary of the Invention

[0008] In view of this, and to fill the gaps and deficiencies in existing technologies, this invention proposes a QoS calculation method and system for low-Earth orbit satellite communication networks based on time-varying topology awareness, to address the problems encountered in the background technology. This invention proposes a technical solution capable of accurately calculating satellite service quality indicators under dynamically changing satellite topology conditions.

[0009] This invention proposes a method and system for calculating QoS in low-Earth orbit satellite communication networks based on time-varying topology awareness, including the following:

[0010] According to a first aspect of the present invention, the present invention proposes a QoS calculation method for low-Earth orbit satellite communication networks based on time-varying topology awareness, characterized in that it includes the following:

[0011] Step S1: Obtain the orbital position information of two-line low-Earth orbit satellites, and use the SGP4 orbital propagation model to calculate the spatial position of each low-Earth orbit satellite within a continuous time window, and construct the dynamic time series sequence of low-Earth orbit satellites for subsequent QoS calculation;

[0012] Step S2: Set the location of the ground terminal, propose a candidate service satellite mechanism, and construct a dynamic network connection relationship based on the spatial geometric relationship between this mechanism and the ground terminal;

[0013] Step S3: Use the dynamic network connection relationships within each time window as a time-varying topology, and extract parameter features from the continuous time-varying topology.

[0014] Step S4: Combine the time-varying topology sensing features described in S3 to perform simulation, obtain a series of QoS tags, and combine the tags to generate a low-Earth orbit satellite dataset;

[0015] Step S5: Divide the low-Earth orbit satellite dataset and input it into the final multilayer perceptron model to calculate the QoS index.

[0016] Further, step S1 includes the following:

[0017] Step S11: Obtain the low-Earth orbit satellite orbital position information, which includes two-line orbital element (TLE) data, and the TLE data includes satellite number, launch time, and security classification.

[0018] Step S12: Construct a time window sequence according to a preset time interval. Let the time window sequence be:

[0019] ;

[0020] in, Let represent the i-th time window, and n represent the total number of time windows;

[0021] Step S13: Calculate the position and motion of the low-Earth orbit satellite under different time windows using the SGP4 orbital propagation model. The calculation formula is as follows:

[0022] ;

[0023] in, This represents the two-row orbital elements corresponding to satellite s. This indicates the spatial position of satellite s within time window t. This represents the velocity of satellite s within time window t. Through the above steps, we obtain the spatial position data of low-orbit satellites at different times, which is used to construct the subsequent network topology.

[0024] Further, step S2 includes the following:

[0025] Step S21: Set the ground terminal location. The spatial location of the ground terminal is set as follows:

[0026] ;

[0027] in, Indicates the latitude of the ground terminal. Indicates the longitude of the ground terminal. Indicates the height of the ground terminal;

[0028] Step S22: Calculate the relative spatial geometric vector between the low-Earth orbit satellite and the ground terminal. The calculation formula is as follows:

[0029] ;

[0030] in, This represents the spatial geometric vector of satellite s relative to the ground terminal within the time window t;

[0031] Step S23: Calculate the slant distance between the satellite and the ground terminal. The calculation formula is as follows:

[0032] ;

[0033] in, This represents the slant distance between satellite s and the ground terminal;

[0034] Step S24: Calculate the elevation angle of the satellite relative to the ground terminal. In the local northeast-northeast coordinate system, the formula for calculating the elevation angle is:

[0035] ;

[0036] in, This represents the elevation angle of satellite s relative to the ground terminal within the time window t. Represents the component of a spatial geometric vector in the zenith direction;

[0037] Step S25: Establish a candidate service satellite scoring mechanism. Within each time window, candidate service satellites are determined based on the aforementioned elevation angle, slant range, and link capacity. The scoring formula is as follows:

[0038] ;

[0039] ;

[0040] in, Indicates the satellite score for candidate services. This indicates the link capacity. Through this scoring mechanism, candidate service satellites within each time window are obtained, providing a basis for subsequent satellite handover indicators and QoS calculations.

[0041] Step S26: Determine satellite visibility based on elevation angle threshold and construct dynamic network connectivity; the determination formula is:

[0042] ;

[0043] in, This indicates whether satellite s is visible to the ground terminal within the time window t. Indicates the preset elevation angle threshold;

[0044] when At that time, it is determined that there is a communicable access link between the ground terminal and satellite s; when At that time, it was determined that there was no communicable access link between the ground terminal and the satellite s, and finally the dynamic network connection relationship within each time window was obtained.

[0045] Further, step S3 includes the following:

[0046] Step S31: Represent the dynamic network connectivity within each time window as a time-varying topology, with the following formula:

[0047] ;

[0048] in, This represents the time-varying topology within a time window t. Represents a set of nodes. Represents a set of links.

[0049] Indicates node or link characteristics, Indicates link weight;

[0050] The time-varying topology corresponding to multiple consecutive time windows is represented as follows:

[0051] ;

[0052] Where g represents a continuous time-varying topology sequence of a low-Earth orbit satellite communication network;

[0053] Step S32: Filter out invisible satellites and retain visible satellites; extract the number of visible satellites, calculate the highest elevation angle and nearest slant range among them, then determine the candidate service satellites, calculate the candidate service satellite switching flag, and the candidate service satellites can be represented as:

[0054] ;

[0055] in, This represents the candidate service satellites within the time window t. Indicates the score for selecting candidate service satellites;

[0056] The switching indicator for candidate service satellites is:

[0057] ;

[0058] in, Indicates the switching indicator for candidate service satellites. Indicates the candidate service satellites for the previous time window;

[0059] After confirming the candidate service satellites, the slant range change rate and elevation angle change rate are calculated, and a time-varying topology sensing feature vector is constructed. The feature vector is represented as follows:

[0060] ;

[0061] in, This represents the average elevation angle of the first k candidate satellites. This represents the average slant range of the top k candidate satellites; ultimately, time-varying topology sensing features describing the changes of the low-Earth orbit satellite communication network over time are obtained.

[0062] Further, step S4 includes the following:

[0063] Step S41: Perform simulation based on the time-varying topology sensing features to obtain a series of QoS tags, and calculate the packet loss rate tag using the following formula:

[0064] ;

[0065] in, Indicates packet loss rate, Indicates the base packet loss rate. This represents the packet loss rate scaling factor. This represents the Sigmoid function. Indicates link utilization. Indicates the congestion threshold. Indicates the satellite's elevation angle. Indicates the low elevation angle threshold. This indicates the candidate service satellite switching flag, and a, b, and c represent weight parameters;

[0066] The formula for calculating throughput labels is:

[0067] ;

[0068] in, Indicates throughput;

[0069] The formula for calculating jitter tags is:

[0070] ;

[0071] in, This indicates jitter, and k represents the length of the sliding window.

[0072] Further, step S5 includes the following:

[0073] Step S51: Divide the low-Earth orbit satellite dataset into a classification dataset and a regression dataset. The classification dataset retains the entire time window and is used to predict link availability and QoS anomalies. The regression dataset retains only visible satellites and is used to predict end-to-end latency, jitter, throughput, and packet loss rate. Input the divided datasets into a multilayer perceptron, where the formula for calculating the hidden layer of the multilayer perceptron is:

[0074] ;

[0075] in, This represents the output of the Lth hidden layer. Let L represent the weight matrix of the Lth layer. This represents the bias vector of the Lth layer. Represents a nonlinear activation function;

[0076] Step S52: Calculate the link availability and QoS anomaly status predicted by the classification model, and predict the final QoS index using the regression model. The calculation formula is as follows:

[0077] ;

[0078] in, Represents a regression model. Indicates the parameters of the regression model. Indicates end-to-end delay. Indicates predicted jitter. Indicates the predicted throughput. This represents the predicted packet loss rate; furthermore, the classification model uses the binary cross-entropy loss function:

[0079] ;

[0080] The regression model uses the mean squared error loss function:

[0081] ;

[0082] Where N represents the number of samples, and m represents the number of classification tasks. This represents the label of the i-th sample on the j-th task. This represents the corresponding predicted probability. This represents the predicted QoS metric vector for the i-th sample. This represents the QoS label vector of the i-th sample; ultimately, the final QoS index prediction result of the low-Earth orbit satellite communication network is obtained.

[0083] According to a second aspect of the present invention, the present invention proposes a QoS calculation system for low-Earth orbit satellite communication networks based on time-varying topology awareness, for performing a QoS calculation method for low-Earth orbit satellite communication networks based on time-varying topology awareness as described in any one of the present invention, characterized in that it includes the following:

[0084] The information acquisition module is used to acquire the orbital position information of two-line low-Earth orbit satellites, and to calculate the spatial position of each low-Earth orbit satellite within a continuous time window using the SGP4 orbital propagation model, thereby constructing a dynamic time sequence of low-Earth orbit satellites for subsequent QoS calculations.

[0085] The dynamic network connection module is used to set the location of the ground terminal, propose a candidate service satellite mechanism, and construct a dynamic network connection based on the spatial geometric relationship between this mechanism and the ground terminal.

[0086] The time-varying module is used to treat the dynamic network connection relationships within each time window as a time-varying topology and extract parameter features from the continuous time-varying topology.

[0087] The QoS tag module is used to perform simulations by combining the time-varying topology sensing features described in S3, obtain a series of QoS tags, and combine the tags to generate a low-orbit satellite dataset.

[0088] The QoS calculation module divides the low-Earth orbit satellite dataset and inputs it into the final multilayer perceptron model to calculate the QoS index.

[0089] According to a third aspect of the present invention, the present invention proposes a QoS calculation system for a low-Earth orbit satellite communication network based on time-varying topology awareness, comprising an electronic device, wherein the electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the computer program, it implements a QoS calculation method for a low-Earth orbit satellite communication network based on time-varying topology awareness as described in any one of the present invention.

[0090] According to a fourth aspect of the present invention, the present invention proposes a QoS calculation system for a low-Earth orbit satellite communication network based on time-varying topology awareness, comprising a computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, it implements a QoS calculation method for a low-Earth orbit satellite communication network based on time-varying topology awareness as described in any one of the present invention.

[0091] The present invention has the following advantages:

[0092] 1. By adopting an orbital propagation model to construct a topology from orbital data and fully considering the spatial position changes caused by satellite motion, the QoS calculation capability under dynamic network conditions is significantly improved.

[0093] 2. A joint modeling mechanism for satellites and multiple QoS tags is proposed, which integrates satellite spatial geometric features with topological structure to improve the dynamic adaptability of satellites in their quality of service calculation;

[0094] 3. Input the dataset into the multilayer perceptron model for classification and regression prediction, so that the model can learn more features and improve the accuracy of the final QoS index calculation. Attached Figure Description

[0095] Figure 1 This is a schematic diagram of the overall process for measuring the service quality indicators of low-orbit satellites according to the present invention.

[0096] Figure 2 This is a schematic diagram illustrating the construction of dynamic network connection relationships according to the present invention.

[0097] Figure 3 This is a schematic diagram of the time-varying topology of the present invention.

[0098] Figure 4 This is a schematic diagram of the QoS simulation tag generated by the present invention.

[0099] Figure 5 This is a schematic diagram of the multilayer perceptron model of the present invention.

[0100] Figure 6 This is a schematic diagram comparing the performance of different models of the present invention on classification tasks.

[0101] Figure 7 This is a schematic diagram comparing the effects of different models of the present invention on regression tasks. Detailed Implementation

[0102] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings.

[0103] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0104] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention; as used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise; furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0105] like Figures 1 to 7 As shown, this invention proposes a QoS calculation method and system for low-Earth orbit satellite communication networks based on time-varying topology awareness, including the following:

[0106] This invention employs an orbital propagation model to construct a topology from orbital data, establishes a joint modeling mechanism for satellites and multiple QoS tags, inputs the dataset into a multilayer perceptron model for classification and regression prediction, and fully considers the spatial position changes caused by satellite motion, thereby realizing dynamic calculation of the service quality of low-Earth orbit satellite communication networks.

[0107] like Figure 1 As shown, according to a first aspect of the present invention, the present invention proposes a QoS calculation method for low-Earth orbit satellite communication networks based on time-varying topology awareness, characterized in that it includes the following:

[0108] Step S1: Obtain the orbital position information of two-line low-Earth orbit satellites, and use the SGP4 orbital propagation model to calculate the spatial position of each low-Earth orbit satellite within a continuous time window, and construct the dynamic time series sequence of low-Earth orbit satellites for subsequent QoS calculation;

[0109] Step S2: Set the location of the ground terminal, propose a candidate service satellite mechanism, and construct a dynamic network connection relationship based on the spatial geometric relationship between this mechanism and the ground terminal;

[0110] Step S3: Use the dynamic network connection relationships within each time window as a time-varying topology, and extract parameter features from the continuous time-varying topology.

[0111] Step S4: Combine the time-varying topology sensing features described in S3 to perform simulation, obtain a series of QoS tags, and combine the tags to generate a low-Earth orbit satellite dataset;

[0112] Step S5: Divide the low-Earth orbit satellite dataset and input it into the final multilayer perceptron model to calculate the QoS index.

[0113] Further, step S1 includes the following:

[0114] Step S11: Obtain the low-Earth orbit satellite orbital position information, which includes two-line orbital element (TLE) data, including satellite number, launch time, and secrecy classification.

[0115] Step S12: Construct a time window sequence according to a preset time interval. Let the time window sequence be:

[0116] ;

[0117] in, Let represent the i-th time window, and n represent the total number of time windows;

[0118] Step S13: Calculate the position and motion of the low-Earth orbit satellite under different time windows using the SGP4 orbital propagation model. The calculation formula is as follows:

[0119] ;

[0120] in, This represents the two-row orbital elements corresponding to satellite s. This indicates the spatial position of satellite s within time window t. This represents the velocity of satellite s within time window t. Through the above steps, we obtain the spatial position data of low-orbit satellites at different times, which is used to construct the subsequent network topology.

[0121] like Figure 2 The diagram illustrates the construction of dynamic network connections. In one embodiment of the present invention, step S2 includes the following:

[0122] Step S21: Set the ground terminal location. The spatial location of the ground terminal is set as follows:

[0123] ;

[0124] in, Indicates the latitude of the ground terminal. Indicates the longitude of the ground terminal. Indicates the height of the ground terminal;

[0125] Step S22: Calculate the relative spatial geometric vector between the low-Earth orbit satellite and the ground terminal. The calculation formula is as follows:

[0126] ;

[0127] in, This represents the spatial geometric vector of satellite s relative to the ground terminal within the time window t;

[0128] Step S23: Calculate the slant distance between the satellite and the ground terminal. The calculation formula is as follows:

[0129] ;

[0130] in, This represents the slant distance between satellite s and the ground terminal;

[0131] Step S24: Calculate the elevation angle of the satellite relative to the ground terminal. In the local northeast-northeast coordinate system, the formula for calculating the elevation angle is:

[0132] ;

[0133] in, This represents the elevation angle of satellite s relative to the ground terminal within the time window t. Represents the component of a spatial geometric vector in the zenith direction;

[0134] Step S25: Establish a candidate service satellite scoring mechanism. Within each time window, candidate service satellites are determined based on the aforementioned elevation angle, slant range, and link capacity. The scoring formula is as follows:

[0135] ;

[0136] ;

[0137] in, Indicates the satellite score for candidate services. This indicates the link capacity. Through this scoring mechanism, candidate service satellites within each time window are obtained, providing a basis for subsequent satellite handover indicators and QoS calculations.

[0138] Step S26: Determine satellite visibility based on elevation angle threshold and construct dynamic network connectivity; the determination formula is:

[0139] ;

[0140] in, This indicates whether satellite s is visible to the ground terminal within the time window t. Indicates the preset elevation angle threshold;

[0141] when At that time, it is determined that there is a communicable access link between the ground terminal and satellite s; when At that time, it is determined that there is no communicable access link between the ground terminal and satellite s. Through the above steps, the dynamic network connection relationship within each time window is obtained.

[0142] like Figure 3The diagram shows a time-varying topology. In one embodiment of the present invention, step S3 further includes the following:

[0143] Step S31: Represent the dynamic network connectivity within each time window as a time-varying topology, with the following formula:

[0144] ;

[0145] in, This represents the time-varying topology within a time window t. Represents a set of nodes. Represents a set of links.

[0146] Indicates node or link characteristics, Indicates link weight;

[0147] The time-varying topology corresponding to multiple consecutive time windows is represented as follows:

[0148] ;

[0149] Where g represents a continuous time-varying topology sequence of a low-Earth orbit satellite communication network;

[0150] Step S32: Filter out invisible satellites and retain visible satellites; extract the number of visible satellites, calculate the highest elevation angle and nearest slant range among them, then determine the candidate service satellites, calculate the candidate service satellite switching flag, and the candidate service satellites can be represented as:

[0151] ;

[0152] in, This represents the candidate service satellites within the time window t. Indicates the score for selecting candidate service satellites;

[0153] The switching indicator for candidate service satellites is:

[0154] ;

[0155] in, Indicates the switching indicator for candidate service satellites. Indicates the candidate service satellites for the previous time window;

[0156] After confirming the candidate service satellites, the slant range change rate and elevation angle change rate are calculated, and a time-varying topology sensing feature vector is constructed. The feature vector is represented as follows:

[0157] ;

[0158] in, This represents the average elevation angle of the first k candidate satellites. This represents the average slant range of the top k candidate satellites; ultimately, time-varying topology sensing features describing the changes of the low-Earth orbit satellite communication network over time are obtained.

[0159] like Figure 4 The diagram shown is a schematic of a QoS simulation tag. In one embodiment of the present invention, step S4 includes the following:

[0160] Step S41: Perform simulation based on the time-varying topology sensing features to obtain a series of QoS tags, and calculate the packet loss rate tag using the following formula:

[0161] ;

[0162] in, Indicates packet loss rate, Indicates the base packet loss rate. This represents the packet loss rate scaling factor. This represents the Sigmoid function. Indicates link utilization. Indicates the congestion threshold. Indicates the satellite's elevation angle. Indicates the low elevation angle threshold. This indicates the candidate service satellite switching flag, and a, b, and c represent weight parameters;

[0163] The formula for calculating throughput labels is:

[0164] ;

[0165] in, Indicates throughput;

[0166] The formula for calculating jitter tags is:

[0167] ;

[0168] in, This indicates jitter, and k represents the length of the sliding window.

[0169] like Figure 5 The diagram shown is a schematic of a multilayer perceptron model. In one embodiment of the present invention, step S5 includes the following:

[0170] Step S51: Divide the low-Earth orbit satellite dataset into a classification dataset and a regression dataset. The classification dataset retains the entire time window and is used to predict link availability and QoS anomalies. The regression dataset retains only visible satellites and is used to predict end-to-end latency, jitter, throughput, and packet loss rate. Input the divided datasets into a multilayer perceptron, where the formula for calculating the hidden layer of the multilayer perceptron is:

[0171] ;

[0172] in, This represents the output of the Lth hidden layer. Let L represent the weight matrix of the Lth layer. This represents the bias vector of the Lth layer. Represents a nonlinear activation function;

[0173] Step S52: Calculate the link availability and QoS anomaly status predicted by the classification model, and predict the final QoS index using the regression model. The calculation formula is as follows:

[0174] ;

[0175] in, Represents a regression model. Indicates the parameters of the regression model. Indicates end-to-end delay. Indicates predicted jitter. Indicates the predicted throughput. This represents the predicted packet loss rate; furthermore, the classification model uses the binary cross-entropy loss function:

[0176] ;

[0177] The regression model uses the mean squared error loss function:

[0178] ;

[0179] Where N represents the number of samples, and m represents the number of classification tasks. This represents the label of the i-th sample on the j-th task. This represents the corresponding predicted probability. This represents the predicted QoS metric vector for the i-th sample. This represents the QoS label vector of the i-th sample; ultimately, the final QoS index prediction result of the low-Earth orbit satellite communication network is obtained.

[0180] According to a second aspect of the present invention, the present invention proposes a QoS calculation system for low-Earth orbit satellite communication networks based on time-varying topology awareness, for performing a QoS calculation method for low-Earth orbit satellite communication networks based on time-varying topology awareness as described in any one of the present invention, characterized in that it includes the following:

[0181] The information acquisition module is used to acquire the orbital position information of two-line low-Earth orbit satellites, and to calculate the spatial position of each low-Earth orbit satellite within a continuous time window using the SGP4 orbital propagation model, thereby constructing a dynamic time sequence of low-Earth orbit satellites for subsequent QoS calculations.

[0182] The dynamic network connection module is used to set the location of the ground terminal, propose a candidate service satellite mechanism, and construct a dynamic network connection based on the spatial geometric relationship between this mechanism and the ground terminal.

[0183] The time-varying module is used to treat the dynamic network connection relationships within each time window as a time-varying topology and extract parameter features from the continuous time-varying topology.

[0184] The QoS tag module is used to perform simulations by combining the time-varying topology sensing features described in S3, obtain a series of QoS tags, and combine the tags to generate a low-orbit satellite dataset.

[0185] The QoS calculation module divides the low-Earth orbit satellite dataset and inputs it into the final multilayer perceptron model to calculate the QoS index.

[0186] According to a third aspect of the present invention, the present invention proposes a QoS calculation system for a low-Earth orbit satellite communication network based on time-varying topology awareness, comprising an electronic device, wherein the electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the computer program, it implements a QoS calculation method for a low-Earth orbit satellite communication network based on time-varying topology awareness as described in any one of the present invention.

[0187] According to a fourth aspect of the present invention, the present invention proposes a QoS calculation system for a low-Earth orbit satellite communication network based on time-varying topology awareness, comprising a computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, it implements a QoS calculation method for a low-Earth orbit satellite communication network based on time-varying topology awareness as described in any one of the present invention.

[0188] To further verify the feasibility and superiority of this method, the present invention provides the following simulation examples for illustration and analysis:

[0189] Environment setup: The computing environment of this embodiment of the invention is an i7-13700F processor and an NVIDIA GeForce RTX4060. The programming language used is Python, the network is built using PyTorch, PyCharm is used as the compiler, Adam is used as the optimizer, and the learning rate is set to 0.00001.

[0190] The satellite data used in the experiment was two-line element (TLE) data for low-Earth orbit (LEO) satellites, sourced from the NORAD satellite orbit element dataset provided by CelesTrak. First, the original TLE text file was parsed, extracting the basic orbital information of each LEO satellite according to a three-line structure: "Satellite Name—First Line Orbital Element—Second Line Orbital Element." Starlink satellites related to the LEO satellite communication network were selected as the primary research objects. Subsequently, the parsed TLE data was organized into a basic satellite information table, obtaining fields such as orbital period, approximate orbital altitude, perigee altitude, apogee altitude, and whether it belongs to a Starlink satellite for each satellite. These fields were used for subsequent orbit propagation and network topology construction.

[0191] The evaluation metrics used in this experimental case include: Pearson correlation coefficient (PCC), which measures the linear correlation between predicted and actual values, with a range of values... Mean absolute error (MAE) measures the magnitude of the average deviation; mean square error (MSE) measures the average squared deviation.

[0192] In summary, this invention provides a time-varying topology-aware QoS calculation method for low-Earth orbit (LEO) satellite communication networks to address the problem of QoS calculation under dynamically changing conditions. This method constructs dynamic network connections based on the spatial geometric relationship between satellites and ground terminals, thereby enabling dynamic calculation of the QoS of LEO satellite communication networks.

[0193] It should be understood that the embodiments and specific implementations described herein are for illustrative purposes only and are not intended to limit the scope of protection of this invention. Those skilled in the art can modify, substitute, or improve the QoS tag simulation method and QoS calculation model based on the basic principles of this invention, for example, replacing TLE data with other ephemeris data, or replacing the multilayer perceptron model with a temporal convolutional network; any modifications, substitutions, and improvements made within the principles of this invention should be included within the scope of protection of this invention.

[0194] The above are preferred embodiments of the present invention. Any changes made to the technical solution of the present invention that do not exceed the scope of the technical solution of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for calculating QoS in low-Earth orbit satellite communication networks based on time-varying topology awareness, characterized in that, Includes the following: Step S1: Obtain the orbital position information of two-line low-Earth orbit satellites, and use the SGP4 orbital propagation model to calculate the spatial position of each low-Earth orbit satellite within a continuous time window, and construct the dynamic time series sequence of low-Earth orbit satellites for subsequent QoS calculation; Step S2: Set the location of the ground terminal, propose a candidate service satellite mechanism, and construct a dynamic network connection relationship based on the spatial geometric relationship between this mechanism and the ground terminal; Step S3: Use the dynamic network connection relationships within each time window as a time-varying topology, and extract parameter features from the continuous time-varying topology. Step S4: Combine the time-varying topology sensing features described in S3 to perform simulation, obtain a series of QoS tags, and combine the tags to generate a low-Earth orbit satellite dataset; Step S5: Divide the low-Earth orbit satellite dataset and input it into the final multilayer perceptron model to calculate the QoS index.

2. The QoS calculation method for low-Earth orbit satellite communication networks based on time-varying topology awareness according to claim 1, characterized in that, Step S1 includes the following: Step S11: Obtain the low-Earth orbit satellite orbital position information, which includes two-line orbital element (TLE) data, and the TLE data includes satellite number, launch time, and security classification. Step S12: Construct a time window sequence according to a preset time interval. Let the time window sequence be: ; in, Let represent the i-th time window, and n represent the total number of time windows; Step S13: Calculate the position and motion of the low-Earth orbit satellite under different time windows using the SGP4 orbital propagation model. The calculation formula is as follows: ; in, This represents the two-row orbital elements corresponding to satellite s. This indicates the spatial position of satellite s within time window t. This represents the velocity of satellite s within time window t. Through the above steps, we obtain the spatial position data of low-orbit satellites at different times, which is used to construct the subsequent network topology.

3. The QoS calculation method for low-Earth orbit satellite communication networks based on time-varying topology awareness according to claim 1, characterized in that, Step S2 includes the following: Step S21: Set the ground terminal location. The spatial location of the ground terminal is set as follows: ; in, Indicates the latitude of the ground terminal. Indicates the longitude of the ground terminal. Indicates the height of the ground terminal; Step S22: Calculate the relative spatial geometric vector between the low-Earth orbit satellite and the ground terminal. The calculation formula is as follows: ; in, This represents the spatial geometric vector of satellite s relative to the ground terminal within the time window t; Step S23: Calculate the slant distance between the satellite and the ground terminal. The calculation formula is as follows: ; in, This represents the slant distance between satellite s and the ground terminal; Step S24: Calculate the elevation angle of the satellite relative to the ground terminal. In the local northeast-northeast coordinate system, the formula for calculating the elevation angle is: ; in, This represents the elevation angle of satellite s relative to the ground terminal within the time window t. This represents the component of a spatial geometric vector in the zenith direction; Step S25: Establish a candidate service satellite scoring mechanism. Within each time window, candidate service satellites are determined based on the aforementioned elevation angle, slant range, and link capacity. The scoring formula is as follows: ; ; in, Indicates the satellite score for candidate services. This indicates the link capacity. Through this scoring mechanism, candidate service satellites within each time window are obtained, providing a basis for subsequent satellite handover indicators and QoS calculations. Step S26: Determine satellite visibility based on elevation angle threshold and construct dynamic network connectivity; the determination formula is: ; in, This indicates whether satellite s is visible to the ground terminal within the time window t. Indicates the preset elevation angle threshold; when At that time, it is determined that there is a communicable access link between the ground terminal and satellite s; when At that time, it was determined that there was no communicable access link between the ground terminal and the satellite s, and finally the dynamic network connection relationship within each time window was obtained.

4. The QoS calculation method for low-Earth orbit satellite communication networks based on time-varying topology awareness according to claim 1, characterized in that, Step S3 includes the following: Step S31: Represent the dynamic network connectivity within each time window as a time-varying topology, with the following formula: ; in, This represents the time-varying topology within a time window t. Represents a set of nodes. Represents a set of links. Indicates node or link characteristics, Indicates link weight; The time-varying topology corresponding to multiple consecutive time windows is represented as follows: ; Where g represents a continuous time-varying topology sequence of a low-Earth orbit satellite communication network; Step S32: Filter out invisible satellites and retain visible satellites; extract the number of visible satellites, calculate the highest elevation angle and nearest slant range among them, then determine the candidate service satellites, calculate the candidate service satellite switching flag, and the candidate service satellites can be represented as: ; in, This represents the candidate service satellites within the time window t. Indicates the score for selecting candidate service satellites; The switching indicator for candidate service satellites is: ; in, Indicates the switching indicator for candidate service satellites. Indicates the candidate service satellites for the previous time window; After confirming the candidate service satellites, the slant range change rate and elevation angle change rate are calculated, and a time-varying topology sensing feature vector is constructed. The feature vector is represented as follows: ; in, This represents the average elevation angle of the first k candidate satellites. This represents the average slant range of the top k candidate satellites; ultimately, time-varying topology sensing features describing the changes of the low-Earth orbit satellite communication network over time are obtained.

5. The QoS calculation method for low-Earth orbit satellite communication networks based on time-varying topology awareness according to claim 1, characterized in that, Step S4 includes the following: Step S41: Perform simulation based on the time-varying topology sensing features to obtain a series of QoS tags, and calculate the packet loss rate tag using the following formula: ; in, Indicates packet loss rate, Indicates the base packet loss rate. This represents the packet loss rate scaling factor. This represents the Sigmoid function. Indicates link utilization. Indicates the congestion threshold. Indicates the satellite's elevation angle. Indicates the low elevation angle threshold. This indicates the candidate service satellite switching flag, and a, b, and c represent weight parameters; The formula for calculating throughput labels is: ; in, Indicates throughput; The formula for calculating jitter tags is: ; in, This indicates jitter, and k represents the length of the sliding window.

6. The QoS calculation method for low-Earth orbit satellite communication networks based on time-varying topology awareness according to claim 1, characterized in that, Step S5 includes the following: Step S51: Divide the low-Earth orbit satellite dataset into a classification dataset and a regression dataset. The classification dataset retains the entire time window and is used to predict link availability and QoS anomalies. The regression dataset retains only visible satellites and is used to predict end-to-end latency, jitter, throughput, and packet loss rate. Input the divided datasets into a multilayer perceptron, where the formula for calculating the hidden layer of the multilayer perceptron is: ; in, This represents the output of the Lth hidden layer. Let L represent the weight matrix of the Lth layer. This represents the bias vector of the Lth layer. Represents a nonlinear activation function; Step S52: Calculate the link availability and QoS anomaly status predicted by the classification model, and predict the final QoS index using the regression model. The calculation formula is as follows: ; in, Represents a regression model. Indicates the parameters of the regression model. Indicates end-to-end delay. Indicates predicted jitter. Indicates the predicted throughput. This represents the predicted packet loss rate; furthermore, the classification model uses the binary cross-entropy loss function: ; The regression model uses the mean squared error loss function: ; Where N represents the number of samples, and m represents the number of classification tasks. This represents the label of the i-th sample on the j-th task. This represents the corresponding predicted probability. This represents the predicted QoS metric vector for the i-th sample. This represents the QoS label vector of the i-th sample; ultimately, the final QoS index prediction result of the low-Earth orbit satellite communication network is obtained.

7. A QoS calculation system for low-Earth orbit satellite communication networks based on time-varying topology awareness, used to execute the QoS calculation method for low-Earth orbit satellite communication networks based on time-varying topology awareness as described in any one of claims 1 to 6, characterized in that, Includes the following: The information acquisition module is used to acquire the orbital position information of two-line low-Earth orbit satellites, and to calculate the spatial position of each low-Earth orbit satellite within a continuous time window using the SGP4 orbital propagation model, thereby constructing a dynamic time sequence of low-Earth orbit satellites for subsequent QoS calculations. The dynamic network connection module is used to set the location of the ground terminal, propose a candidate service satellite mechanism, and construct a dynamic network connection based on the spatial geometric relationship between this mechanism and the ground terminal. The time-varying module is used to treat the dynamic network connection relationships within each time window as a time-varying topology and extract parameter features from the continuous time-varying topology. The QoS tag module is used to perform simulations by combining the time-varying topology sensing features described in S3, obtain a series of QoS tags, and combine the tags to generate a low-orbit satellite dataset. The QoS calculation module divides the low-Earth orbit satellite dataset and inputs it into the final multilayer perceptron model to calculate the QoS index.

8. A QoS calculation system for a low-Earth orbit satellite communication network based on time-varying topology awareness, comprising an electronic device, wherein the electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements a QoS calculation method for low-Earth orbit satellite communication networks based on time-varying topology awareness as described in any one of claims 1 to 6.

9. A QoS calculation system for a low-Earth orbit satellite communication network based on time-varying topology awareness, comprising a computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements a QoS calculation method for low-orbit satellite communication networks based on time-varying topology awareness as described in any one of claims 1 to 6.