Demonstration and identification simulation platform and method for low earth orbit satellite network access flow
By designing a simulation platform for low-orbit satellite networks, using WPF framework and port recognition and machine learning recognition methods, the problem of difficult access traffic recognition in low-orbit satellite networks is solved, and high-accuracy traffic recognition and network management optimization are achieved.
Patent Information
- Application Number
- CN202510099512.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-21
- Publication Date
- 2025-06-24
AI Technical Summary
It is difficult to identify access traffic in low-orbit satellite networks, and it is difficult for the prior art to accurately identify different types of access traffic, affecting network management and optimization.
A demonstration and identification simulation platform for low-orbit satellite network access traffic was designed, developed using WPF framework, and the modules communicate with each other through UI interface modules and event realization modules to realize traffic recognition. The platform includes satellite motion trajectory area, node parameter selection area, traffic real-time curve area and traffic recognition area. It uses port recognition and machine learning recognition methods to improve the accuracy of traffic recognition.
It realizes high accuracy identification of low-orbit satellite network access traffic, simplifies operational processes, improves work efficiency and the availability of simulation platforms, and provides support for the optimization and evaluation of satellite communication networks.
Smart Images

Figure CN120200909A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of satellite network simulation, and particularly relates to a demonstration and recognition simulation platform and method for low-earth orbit satellite network access traffic. Background Art
[0002] As of 2023, the number of global Internet users is approximately 5.3 billion, accounting for 65.4% of the global population. With a series of countries such as the United States, China, and the European Union increasing their investment in low-earth orbit satellites, on the basis of the huge scale of original ground Internet users, satellite Internet is showing an accelerating development trend. As the scale of satellite Internet users expands, satellite network application services are becoming increasingly rich. Different satellite services require different network bandwidths and transmission delays. Identifying different types of services can provide a suitable satellite network for them, thus meeting different needs, which helps the development of emerging services. At the same time, the service identification and classification of satellite networks are of great significance to satellite network information security. It can be applied to intrusion detection systems to detect satellite network traffic in real time. Once abnormal network traffic is found, relevant measures can be taken immediately to minimize losses. It can also predict the future usage of satellite networks, helping managers to make reasonable network plans to avoid network congestion. Therefore, as the scale of satellite Internet expands day by day, satellite traffic identification and classification become increasingly important.
[0003] With the wide deployment of low-earth orbit satellite communication networks, the types and complexity of satellite access services have increased significantly. The performance of different types of access traffic in the satellite communication environment varies significantly, and it is impossible to simply follow the model of ground networks. At the same time, the diversification of traffic types and the dynamic changes in service loads in satellite networks make accurate traffic identification an important prerequisite for satellite network management and optimization. Summary of the Invention
[0004] The main purpose of the present invention is to propose a demonstration and recognition simulation platform and method for low-earth orbit satellite network access traffic, aiming to improve the accuracy of traffic identification and provide support for the optimization and evaluation of satellite communication networks.
[0005] To achieve the above object, the present invention provides a demonstration and recognition simulation platform for low-earth orbit satellite network access traffic, including: a front-end UI interface module and a back-end event implementation module. The demonstration and recognition simulation platform for low-earth orbit satellite network access traffic is based on the WPF (Windows Presentation Foundation) application framework to integrate the front-end UI (User Interface) interface module and the back-end event implementation module. The UI interface module and the event implementation module communicate and cooperate with each other.
[0006] A further technical solution of the present invention is that the UI interface is provided with four regions: a satellite movement trajectory region, a node parameter selection region, an imported traffic real-time curve region, and an imported traffic recognition region. Among them,
[0007] Two types of controls are added to the satellite movement trajectory region. One is for satellite scene display, and the other is for binding to the trigger event of satellite scene loading or closing.
[0008] Three types of controls are added to the node parameter selection region. One is for manually inputting the satellite number parameter of the node, one is for selecting the service type of the node, and one is for binding to the trigger event of accessing satellite traffic import.
[0009] Three types of controls are added to the imported traffic real-time curve region. One is for real-time analyzing the increasing rate of the imported traffic and drawing a line chart, one is for real-time displaying the longitude and latitude position information of the imported traffic, and one is for binding to the trigger event of clearing all data and stopping analysis.
[0010] Three types of controls are added to the imported traffic recognition region divided in the upper right of the UI interface. One is for the selection of the recognition method, one is for displaying the recognition result of the imported traffic, and one is for binding to the trigger event of recognizing the imported traffic using the selected recognition method.
[0011] A further technical solution of the present invention is that the event implementation module belongs to the backend in the WPF framework, is used to receive trigger events from the UI interface, run corresponding functions according to the trigger events, and feedback the results to the corresponding controls in the UI interface.
[0012] A further technical solution of the present invention is that the event implementation module includes a satellite scene loading module;
[0013] The satellite scene loading module is used to import the pre-saved low-earth orbit satellite constellation scene file and provide it to the AxAgUiAxVOCntrl control in the UI interface when the trigger event in the UI interface is to load the movement trajectory scenes of all satellites, and the AxAgUiAxVOCntrl control performs the loading;
[0014] When the trigger event is to load the movement trajectory scenes of several selected satellites, according to the satellite numbers already input in the UI interface, immediately create the scenes of these several satellites and provide them to the AxAgUiAxVOCntrl control in the UI interface, and the AxAgUiAxVOCntrl control performs the loading;
[0015] When the trigger event is to close the satellite movement trajectory scene, stop providing the scene to the STK control in the UI interface.
[0016] A further technical solution of the present invention is that the event implementation module further includes a satellite traffic import module;
[0017] When the event triggered in the UI interface is to import the satellite traffic of a certain node, the satellite traffic import module is used to call the python import script according to the satellite number and service type parameters of the corresponding node in the UI interface, import the satellite traffic with the same parameters as this node, and at the same time, it will also call the python packet capture script to continuously write the imported traffic into a pcapng type data packet file. For the data packet file with an increasing file size, this part calculates the increasing rate of the data packet size in real time by means of regularly detecting the data packet size, and adds it to an array. The data in the array will be displayed on the CartesianChart control in real time to form a real-time curve graph of the imported traffic;
[0018] When the event triggered in the UI interface is to clear all data and stop the script from running, the data in the array will be cleared, the curve graph will be reset to zero, and the import script and the packet capture script will also terminate running.
[0019] A further technical solution of the present invention is that the event implementation module further includes a satellite traffic identification module;
[0020] When all the satellite traffic is imported, if port identification is selected in the UI interface and the event triggered is to identify the imported traffic of a certain node using the port number, the satellite traffic identification module is used to directly read the port number of the corresponding data packet and judge the service type of the imported traffic according to the known port number mapping table;
[0021] If machine learning identification is selected in the UI interface and the event triggered is to identify the imported traffic of a certain node using machine learning methods, the python machine learning identification script is called to extract the feature vectors from the corresponding data packets and substitute them into the trained model for prediction, thereby judging the service type of the imported traffic; among them, the multi-layer perceptron classifier provided by the sklearn library is used for training the model.
[0022] A further technical solution of the present invention is that the design and operation of the UI interface include the following steps:
[0023] Step S10, design the basic structure of the UI interface, that is, create and set the window, set the layout container, and add area dividing lines;
[0024] Step S20, divide the upper left part of the UI interface into a satellite movement trajectory area, and add two controls, one for satellite scene display and the other for binding with the trigger event of satellite scene loading or closing;
[0025] Step S30: Divide the lower left part of the UI interface into node parameter selection areas, and add three controls. One is for manually inputting the satellite number parameter of the node, one is for selecting the service type of the node, and one is for binding to the trigger event of importing satellite traffic access;
[0026] Step S40: Divide the lower right part of the UI interface into an imported traffic real-time curve area, and add three controls. One is for analyzing the increasing rate of the imported traffic in real time and drawing a line chart, one is for displaying the longitude and latitude position information of the imported traffic in real time, and one is for binding to the trigger events of clearing all data and stopping analysis;
[0027] Step S50: Divide the upper right part of the UI interface into an imported traffic recognition area, and add three controls. One is for selecting the recognition method, one is for displaying the recognition result of the imported traffic, and one is for binding to the trigger event of recognizing the imported traffic using the selected recognition method.
[0028] To achieve the above object, the present invention also proposes a demonstration and recognition simulation method for low-earth orbit satellite network access traffic, and the method includes the following steps:
[0029] Step S100: Develop a demonstration and recognition simulation platform for low-earth orbit satellite network access traffic as described in any one of claims 1 to 8 using the WPF framework in the Windows platform;
[0030] Step S200: Import different types of real and reliable satellite access traffic through each node;
[0031] Step S300: After the traffic import is completed, use two methods of port recognition and machine learning recognition to accurately recognize the imported traffic;
[0032] Step S400: Intuitively control the operation of the entire system and observe the operation results through the UI interface.
[0033] The beneficial effects of the present invention's demonstration and recognition simulation platform and method for low-earth orbit satellite network access traffic are:
[0034] The present invention develops a fully functional demonstration and recognition simulation platform for satellite access traffic through the WPF framework of C#, imports different types of real and reliable satellite access traffic, uses port or machine learning methods to achieve high-accuracy traffic recognition, and designs a UI interface to clearly display the entire process, verifying the integrity of the simulation platform function. The present invention has the following advantages:
[0035] 1. The present invention develops an integrated simulation analysis platform that can import different types of real and reliable satellite access traffic and accurately recognize it.
[0036] 2. The simulation platform of the present invention controls the system operation intuitively through the UI interface, and displays the curve of the imported traffic and the recognition result of the imported traffic in real time, simplifies the operation process, and improves the work efficiency and the usability of the simulation platform. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on the structures shown in these drawings.
[0038] Figure 1 is the system architecture diagram of the demonstration and recognition simulation platform for the access traffic of low-earth orbit satellite networks of the present invention;
[0039] Figure 2 is the schematic diagram of the training accuracy and confusion matrix of the double-layer neural network;
[0040] Figure 3 is the overall schematic diagram of the UI interface;
[0041] Figure 4 is the schematic diagram of the parameter setting of the imported traffic;
[0042] Figure 5 is the schematic diagram of the real-time traffic curve;
[0043] Figure 6 is the schematic diagram of the recognition results of four node ports;
[0044] Figure 7 is the recognition result of machine learning of four nodes.
[0045] The realization, functional features and advantages of the object of the present invention will be further described in conjunction with the embodiments with reference to the drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.
[0047] The present invention proposes a demonstration and recognition simulation platform for low-Earth orbit satellite network access traffic. By importing different types of real and reliable satellite access traffic, the present invention can achieve high-accuracy traffic recognition for the imported traffic using port or machine learning methods, and designs a user interface to clearly display the entire process, providing a functionally complete simulation platform to support the performance optimization and evaluation of satellite communication networks.
[0048] The present invention uses WPF (a framework for building desktop applications) in the Windows platform to develop a UI user interface, and imports different types of real and reliable satellite access traffic through each node. After the traffic import is completed, two methods, namely port recognition and machine learning recognition, are used to accurately recognize the imported traffic. Among them, the UI interface can intuitively control the operation of the entire system and observe the operation results, improving the usability of the simulation platform.
[0049] Specifically, please refer to Figure 1 , a preferred embodiment of the demonstration and recognition simulation platform for low-Earth orbit satellite network access traffic of the present invention includes: a front-end UI interface module and a back-end event implementation module. The demonstration and recognition simulation platform for low-Earth orbit satellite network access traffic separates the front-end UI interface module and its code from the back-end event implementation module logic code based on the WPF application framework, and the UI interface module and the event implementation module communicate and cooperate with each other.
[0050] This embodiment separates the front-end UI interface and its code from the back-end logic code based on the WPF application framework in C# language. At the same time, the front-end and the back-end can communicate and cooperate with each other. This separation makes the development of large-scale applications more modular and easier to maintain, and developers can easily expand existing controls or create completely custom controls to meet personalized needs. In view of the above characteristics, this embodiment selects the WPF framework to develop the satellite access traffic demonstration and recognition simulation platform. As Figure 1 shown, this simulation platform is developed on a PC and consists of two modules, which will be introduced in turn below.
[0051] I. UI Interface Module
[0052] In this embodiment, the UI interface module belongs to the front-end UI interface and its code part in the WPF framework. The code of each part corresponds one-to-one with the controls in the UI interface, and supports changing the size and position of the controls in both the UI interface and the code. Its controls can bind trigger events to control the operation of the corresponding logic code in the back-end event implementation module and obtain feedback.
[0053] The UI interface is provided with four areas: a satellite motion trajectory area, a node parameter selection area, a real-time curve area for imported traffic, and an imported traffic recognition area.
[0054] Among them, two types of controls are added to the satellite motion trajectory area, one for satellite scene display and one for binding to the trigger event of satellite scene loading or closing.
[0055] Three types of controls are added to the node parameter selection area. One is for manually inputting the satellite number parameter of the node, one is for selecting the service type of the node, and one is for binding to the trigger event of satellite traffic import.
[0056] Three types of controls are added to the imported traffic real-time curve area. One is for analyzing the increasing rate of the imported traffic in real time and drawing a line chart, one is for displaying the longitude and latitude position information of the imported traffic in real time, and one is for binding to the trigger event of clearing all data and stopping analysis.
[0057] Three types of controls are added to the imported traffic recognition area divided in the upper right of the UI interface. One is for selecting the recognition method, one is for displaying the recognition result of the imported traffic, and one is for binding to the trigger event of recognizing the imported traffic using the selected recognition method.
[0058] The design and operation of the UI interface module include the following steps:
[0059] Step S10: Design the basic structure of the UI interface, that is, create and set the window, set the layout container, and add area dividing lines;
[0060] Step S20: Divide the upper left of the UI interface into the satellite motion trajectory area and add two types of controls, one for satellite scene display and one for binding to the trigger event of satellite scene loading or closing;
[0061] Step S30: Divide the lower left of the UI interface into the node parameter selection area and add three types of controls, one for manually inputting the satellite number parameter of the node, one for selecting the service type of the node, and one for binding to the trigger event of satellite traffic import;
[0062] Step S40: Divide the lower right of the UI interface into the imported traffic real-time curve area and add three types of controls, one for analyzing the increasing rate of the imported traffic in real time and drawing a line chart, one for displaying the longitude and latitude position information of the imported traffic in real time, and one for binding to the trigger event of clearing all data and stopping analysis;
[0063] Step S50: Divide the upper right of the UI interface into the imported traffic recognition area and add three types of controls, one for selecting the recognition method, one for displaying the recognition result of the imported traffic, and one for binding to the trigger event of recognizing the imported traffic using the selected recognition method.
[0064] The following elaborates in detail on the design and steps of the UI interface module.
[0065] 1. Basic structure of the UI interface:
[0066] 1) Create and set the window
[0067] Use the Window control to create a window and set properties such as the title, initial window size, and position for the window.
[0068] 2) Set the layout container
[0069] Use the Grid control as the layout container, which allows us to divide the interface by defining rows and columns, facilitating the arrangement and organization of elements.
[0070] 3) Add area dividing lines
[0071] Use the Line control to draw a vertical line and a horizontal line in the interface as dividing lines, dividing the interface into different areas for visual distinction.
[0072] 2. Functional controls for each area of the UI interface:
[0073] The UI interface is designed into four areas. The controls that may be used in each area are: the Label control, which is used to display static text, representing the explanation of the function of a certain part of the control. This control has a simple function and a large number, so it will not be described in the following areas; the TextBlock control, which is used to display dynamic text, representing the result display after a triggering event occurs; the Button control, which is bound to different triggering events and can add text descriptions; the ComboBox control, which provides different options for the user; the TextBox control, which allows the user to manually input in a certain form when the number of options is large.
[0074] 1) Satellite motion trajectory area
[0075] Add an AxAgUiAxVOCntrl control provided by STK (Satellite Tool Kit) in the upper left area. This control can be used after installing the STK software and adding relevant references. It is usually used to display the visualization content of STK simulation data, such as satellite orbits, sensors, platforms, etc. Then add Button controls such as "All Satellites", "Selected Satellites", and "Close Scene". The triggering event bound to "All Satellites" is to load the motion trajectory scene of all satellites, the triggering event bound to "Selected Satellites" is to load the motion trajectory scene of several selected satellites, and the triggering event bound to "Close Scene" is to close the satellite motion trajectory scene.
[0076] 2) Parameter selection area for each node
[0077] Add several identical graphics in different colors to the lower left area. Different colors correspond to different nodes. Configure a TextBox control and a ComboBox control for each node, which are used to input the satellite number of the node and select the service type of the node respectively. Then configure a "Click to Import" Button control for each node. The trigger event bound to "Click to Import" is to import the satellite traffic of a certain node.
[0078] 3) Import traffic real-time curve area
[0079] Add the same number of CartesianChart controls as the nodes to the lower right area. This control needs to install LiveCharts.Wpf in the NuGet package before it can be used. It is suitable for creating beautiful and dynamic data visualization charts. Then configure a TextBlock control and a "Clear Data" Button control for each CartesianChart control. The TextBlock control is used to display the longitude and latitude information in the imported traffic. The trigger event bound to "Clear Data" is to clear all data and stop the script from running.
[0080] 4) Import traffic recognition area
[0081] Add a ComboBox control to the upper right area to select the recognition method. Then add the same number of "Click to Recognize" Button controls as the nodes. Each Button control is configured with several TextBlock controls to display the proportion of each type of service in the recognition result. The trigger event bound to "Click to Recognize" is to recognize the imported traffic of a certain node using the port number or to recognize the imported traffic of a certain node using machine learning methods, which changes according to the recognition method in the TextBox control.
[0082] II. Event implementation module:
[0083] In this embodiment, the event implementation module belongs to the logical code part of the backend in the WPF framework, which is used to receive the trigger events from the UI interface, run the corresponding function code according to the trigger events, and feedback the results to the corresponding controls in the UI interface. The event implementation module includes three parts: the satellite scene loading module, the satellite traffic importing module, and the satellite traffic recognizing module. The following introduces these three parts.
[0084] 1. Satellite scene loading module, which has three functions.
[0085] The described satellite scene loading module is used to import the pre-saved low-earth orbit satellite constellation scene file when the trigger event on the UI interface is to load the motion trajectory scenes of all satellites, and provide it to the AxAgUiAxVOCntrl control in the UI interface for loading.
[0086] When the trigger event is to load the motion trajectory scenes of several selected satellites, according to the satellite numbers already input in the UI interface, the scenes of these several satellites are created immediately and provided to the AxAgUiAxVOCntrl control in the UI interface for loading.
[0087] When the trigger event is to close the satellite motion trajectory scene, stop providing the scene to the STK control in the UI interface.
[0088] 2. Import satellite traffic module, which has three functions.
[0089] The described satellite traffic import module is used to, when the trigger event in the UI interface is to import the satellite traffic of a certain node, call the python import script according to the satellite number and service type parameters of the corresponding node in the UI interface to import the satellite traffic with the same parameters as this node. At the same time, it will also call the python packet capture script to continuously write the imported traffic into a pcapng type data packet file. For the data packet file with the continuously increasing file size, this part calculates the increasing rate of the data packet size in real time by the method of regularly detecting the data packet size and adds it to an array. The data in the array will be displayed on the CartesianChart control in real time to form a real-time curve graph of the imported traffic.
[0090] When the trigger events in the UI interface are to clear all data and stop the script from running, the data in the array will be cleared, the curve graph will be reset to zero, and the import script and the packet capture script will also terminate running. To ensure that the satellite traffic of all nodes to be simulated can be imported simultaneously and packet capture can be performed, this embodiment adopts a multi-threaded method to ensure that each python script can run simultaneously to improve efficiency and performance.
[0091] 3. Satellite traffic identification module, which has two functions.
[0092] The described satellite traffic identification module is used to, after all the satellite traffic is imported, if port identification is selected in the UI interface and the trigger event is to identify the imported traffic of a certain node using the port number, directly read the port number of the corresponding data packet and judge the service type of the imported traffic according to the known port number mapping table;
[0093] If machine learning recognition is selected in the UI interface and the trigger event is to identify the imported traffic of a certain node using machine learning methods, the python machine learning recognition script is called to extract the feature vectors from the corresponding data packets and substitute them into the trained model for prediction, thereby determining the service type of the imported traffic. The trained model uses the multi-layer perceptron classifier provided by the sklearn library, which is a feedforward artificial neural network composed of an input layer, a hidden layer, and an output layer, and is trained using the backpropagation algorithm.
[0094] There may be multiple data streams here, so there are multiple data packet files that need to be recognized. Therefore, the proportion of different service types can be calculated to determine the accuracy of this recognition. The python machine learning recognition script also runs in a multi-threaded manner.
[0095] The usage steps of the UI interface in this embodiment are as follows:
[0096] 1) Display the satellite motion trajectory
[0097] It is located in the upper left corner of the UI interface. There are two display states. The first is the display of the low-earth orbit satellite constellation. Click the "All Satellites" Button control, and the low-earth orbit satellite constellation scene imported from the backend will be loaded in the AxAgUiAxVOCntrl control to display the motion trajectories of 1584 satellites. The second is the display of several selected satellites. Click the "Selected Satellites" Button control, and the scene of several satellites created from the backend will be loaded in the STK control to display the motion trajectories of these several satellites. The number of satellites is equal to the number of nodes. The numbers of the selected satellites need to be input in the TextBox control in the form of Sati_j, where i is the number of orbits and j is the number of satellites in each orbit. Finally, click the "Close Scene" Button control to stop loading the scene and restore the initial state.
[0098] 2) Select the parameters of the imported traffic for each node
[0099] It is located in the lower left corner of the UI interface. There are two types of parameters. The first is the satellite number parameter, which is input in the TextBox control in the form of Sati_j. The second is the satellite service type parameter, which is selected from "Web Service", "Voice Service", or "Video Service" in the ComboBox control. After the parameters of several nodes are selected respectively, click the "Click to Import" Button control, and the backend will simultaneously call all the python import and packet capture scripts.
[0100] 3) Display the real-time curve of the imported traffic
[0101] It is located in the lower right corner of the UI interface. After clicking the "Click to Import" Button control, the data of the increasing rate array with real-time increase in quantity in the back-end program will be obtained and displayed in the CartesianChart control in the form of a line chart, thereby obtaining the real-time change curve of the imported traffic. At the same time, the satellite longitude and latitude information in the imported traffic will also be displayed in real time in the TextBlock control. When the traffic curve is 0 for a long time, it means that the traffic import is completed. At this time, clicking the "Clear Data" Button control can reset the traffic curve.
[0102] 4) Display the recognition results of the imported traffic of each node
[0103] It is located in the upper right corner of the UI interface. After the traffic of each node is imported, select one of "Port Recognition" or "Machine Learning" in the ComboBox control. After the selection is completed, click the "Click to Recognize" Button control. If it is "Port Recognition", the port numbers of the data packets of each node will be read simultaneously to judge the traffic type; if it is "Machine Learning", the python machine learning recognition script will be called simultaneously to recognize the data packets of each node. Finally, the recognition ratios of the three services will be displayed in the TextBlock control, and the recognition accuracy will be judged accordingly.
[0104] The following analyzes the simulation results of the demonstration and recognition simulation platform for the low-earth orbit satellite network access traffic of the present invention.
[0105] The parameters of the PC side of the simulation platform are shown in Table 1. The perceptron classifier model with two hidden layers is selected for training in machine learning recognition. The sample quantity and model parameters are shown in Table 2. The average sending time, average interval time, variance of sending time, and variance of interval time of the imported traffic are selected as feature vectors. The sample ratios of the three services are the same, and the training results are shown in Figure 2 . The ranges of the number of nodes, the number of satellites, and the satellite number Sati_j in the UI interface are shown in Table 3. By running the simulation platform, the following figures can be obtained. The overall UI interface is shown in Figure 3 , the parameter settings and real-time curve of the imported traffic in the UI interface are shown in Figure 4 and Figure 5 , and the results of the final two recognition methods are shown in Figure 6 and Figure 7 .
[0106] Table 1 PC-side parameters
[0107]
[0108] Table 2 Multilayer perceptron classifier model parameters
[0109]
[0110] Table 3 PC-side Parameters
[0111]
[0112]
[0113] Analysis Figure 2 In the end, the training accuracy rate of the double-layer neural network model reached 95.43%. It can be clearly seen from the confusion matrix that most recognition errors occurred between web and voice services, while the video service achieved a very high recognition accuracy rate.
[0114] Analysis Figure 3 In the upper left part, it can be seen that the satellite motion trajectory map is operating normally, and the satellite number is the same as the setting.
[0115] Analysis Figure 4 and Figure 5 Since the selected satellites are pairwise symmetric, the latitudes and longitudes of the satellites at Node 1 and Node 2 are symmetric, and the latitudes and longitudes of the satellites at Node 3 and Node 4 are symmetric. This is the same as the pattern reflected by the latitude and longitude data of the four-node satellites in the UI interface: (-64.974, -34.168), (115.026, 34.168), (-179.794, 28.93), (0.206, -28.93). From the latitude and longitude, it can be known that Node 1 is in the Patagonia region of Argentina, Node 2 is in Jinan City, Shandong Province, China, Node 3 is in the central Pacific Ocean, and Node 4 is in the southern Atlantic Ocean. The land traffic is greater than the ocean traffic, which is consistent with the peak of the traffic curve.
[0116] Analysis Figure 6 and Figure 7 The port recognition can achieve an accuracy rate of 100%, which is consistent with the characteristics of port recognition - when the port number of the known traffic is available, it can accurately identify according to the port number of the unencrypted traffic; the machine learning recognition of each node reached the accuracy rates of 95.2%, 95.2%, 100%, and 100% respectively, which is consistent with the training results.
[0117] In summary, all the functions of the simulation platform can be realized.
[0118] The beneficial effects of the present invention's demonstration and recognition simulation platform for low-earth orbit satellite network access traffic are as follows:
[0119] The present invention develops a complete-function demonstration and recognition simulation platform for satellite access traffic through the WPF framework of C#. By importing real and reliable different types of satellite access traffic, it uses port or machine learning methods to achieve high-accuracy traffic recognition, and designs a UI interface to clearly display the whole process, verifying the integrity of the simulation platform's functions. The present invention has the following advantages:
[0120] 1. The present invention develops an integrated simulation analysis platform that can import real and reliable satellite access traffic of different types and accurately identify it.
[0121] 2. The simulation platform of the present invention intuitively controls the system operation through the UI interface, and real-time displays the curve of the imported traffic and the recognition result of the imported traffic, simplifies the operation process, and improves the work efficiency and the usability of the simulation platform.
[0122] To achieve the above object, the present invention also proposes a demonstration and recognition simulation method for low-earth orbit satellite network access traffic, and the method includes the following steps:
[0123] Step S100, develop the demonstration and recognition simulation platform for low-earth orbit satellite network access traffic as described in the above embodiment using the WPF framework in the Windows platform. The demonstration and recognition simulation platform for low-earth orbit satellite network access traffic has been elaborated in detail above and will not be repeated here.
[0124] Step S200, import real and reliable satellite access traffic of different types through each node.
[0125] Step S300, after the traffic import is completed, accurately identify the imported traffic through two methods: port recognition and machine learning recognition.
[0126] Step S400, intuitively control the operation of the entire system and observe the operation results through the UI interface.
[0127] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. All equivalent structural transformations made under the concept of the present invention by using the content of the specification and drawings of the present invention, or directly / indirectly applied to other related technical fields are included in the patent protection scope of the present invention.
Claims
1. A demonstration and identification simulation platform for low-orbit satellite network access traffic, characterized in that: include: The front-end UI interface module and the back-end event implementation module, the low-orbit satellite network access traffic demonstration and identification simulation platform separates the front-end UI interface module and the back-end event implementation module based on the WPF application framework, and the UI interface module and the event implementation module communicate and collaborate with each other.
2. The demonstration and identification simulation platform for low-orbit satellite network access traffic according to claim 1 is characterized in that: The UI interface is set up with four areas: satellite motion trajectory area, node parameter selection area, imported traffic real-time curve area and imported traffic identification area. The satellite motion track area is added with two controls, one for satellite scene display, and the other for binding with the trigger event of satellite scene loading or closing; Three controls are added to the parameter selection area of each node, one for manually inputting the satellite number parameter of the node, one for selecting the service type of the node, and one for binding with the trigger event of importing access satellite traffic; The imported traffic real-time curve area is added with three controls, one for real-time analysis of the increase rate of the imported traffic and drawing a line graph, one for real-time display of the latitude and longitude position information of the imported traffic, and one for binding with the trigger event of clearing all data and stopping analysis; The upper right portion of the UI interface is divided into an imported traffic identification area and three controls are added thereto, one for selecting an identification method, one for displaying the identification result of the imported traffic, and one for binding to a trigger event for identifying the imported traffic using the selected identification method.
3. The demonstration and identification simulation platform for low-orbit satellite network access traffic according to claim 2 is characterized in that: The event implementation module belongs to the back end of the WPF framework, and is used to receive trigger events from the UI interface, run corresponding functions according to the trigger events, and feed back the results to the corresponding controls in the UI interface.
4. The demonstration and identification simulation platform for low-orbit satellite network access traffic according to claim 3 is characterized in that: The event realization module includes a satellite scene loading module; The satellite scene loading module is used to import the pre-saved low-orbit satellite constellation scene file when the trigger event of the UI interface is to load the motion trajectory scene of all satellites, and provide it to the AxAgUiAxVOCntrl control in the UI interface, and the AxAgUiAxVOCntrl control is loaded; When the trigger event is to load the motion trajectory scenes of several selected satellites, the scenes of these several satellites are created immediately according to the satellite numbers that have been entered in the UI interface, and provided to the AxAgUiAxVOCntrl control in the UI interface, and the AxAgUiAxVOCntrl control is loaded; When the trigger event is to close the satellite motion trajectory scene, stop providing the scene to the STK controls in the UI interface.
5. The demonstration and identification simulation platform for low-orbit satellite network access traffic according to claim 4 is characterized in that: The event realization module also includes a satellite traffic import module; The import satellite traffic module is used to trigger an event in the UI interface, which is to import the satellite traffic of a certain node. According to the satellite number and service type parameters of the corresponding node in the UI interface, the python import script is called to import the satellite traffic with the same parameters as this node. At the same time, the python packet capture script is also called to continuously write the imported traffic into a data packet file of the pcapng type. For the data packet file with an increasing file size, this part calculates the increase rate of the data packet size in real time by a method of regularly detecting the data packet size, and adds it to the array. The data in the array will be displayed in real time on the CartesianChart control to form a real-time curve chart of the imported traffic; When the trigger event in the UI interface is to clear all data and stop the script, the data in the array will be cleared, the curve chart will be reset to zero, and the import script and packet capture script will also terminate.
6. The demonstration and identification simulation platform for low-orbit satellite network access traffic according to claim 5 is characterized in that: The event realization module also includes a satellite traffic identification module; The satellite traffic identification module is used to directly read the port number of the corresponding data packet after all satellite traffic has been imported. If port identification is selected in the UI interface, and the trigger event is to identify the imported traffic of a certain node using the port number, the service type of the imported traffic is determined according to the known port number mapping table; If machine learning identification is selected in the UI interface, and the trigger event is the use of machine learning methods to identify the imported traffic of a certain node, the python machine learning identification script is called to extract the feature vector from the corresponding data packet and substitute it into the trained model for prediction, thereby determining the service type of the imported traffic; The training model uses the multi-layer perceptron classifier provided by the sklearn library.
7. The demonstration and identification simulation platform for low-orbit satellite network access traffic according to any one of claims 1 to 6, characterized in that: The design and operation of the UI interface includes the following steps: Step S10, designing the basic structure of the UI interface, namely, creating and setting windows, setting layout containers, and adding area dividing lines; Step S20, dividing the upper left corner of the UI interface into a satellite motion track area, adding two controls, one for satellite scene display, and the other for binding with a trigger event for loading or closing a satellite scene; Step S30: divide the lower left corner of the UI interface into a node parameter selection area, and add three controls: one for manually inputting the satellite number parameter of the node, one for selecting the service type of the node, and one for binding with the trigger event of importing access satellite traffic; Step S40, divide the lower right corner of the UI interface into an imported traffic real-time curve area, add three controls, one for real-time analysis of the increase rate of the imported traffic and drawing a line graph, one for real-time display of the latitude and longitude location information of the imported traffic, and one for binding with a trigger event for clearing all data and stopping analysis; Step S50, divide the upper right corner of the UI interface into an imported traffic identification area, and add three controls, one for selecting an identification method, one for displaying the identification result of the imported traffic, and one for binding to a trigger event for identifying the imported traffic using the selected identification method.
8. A demonstration and identification simulation method for low-orbit satellite network access traffic, characterized in that: The method comprises the following steps: Step S100, using the WPF framework in the Windows platform to develop a demonstration and identification simulation platform for low-orbit satellite network access traffic as described in any one of claims 1 to 7; Step S200, importing real and reliable satellite access traffic of different types through each node; Step S300, after the traffic is imported, the imported traffic is identified with high accuracy by using two methods: port identification and machine learning identification; Step S400: The entire system operation can be intuitively controlled and the operation results can be observed through the UI interface.