A method and system for assisting decision making for space segment passage configuration

Abstract

The application discloses a kind of space section passage configuration auxiliary decision-making method and system, comprising: constructing directed weighted node-connection graph, node indicates satellite, and the directed line between node indicates signal direction;According to the connection relationship in directed weighted node-connection graph, attribute data is obtained, and attribute data includes at least one of the following: carrier signal bandwidth, carrier signal power, carrier signal modulation mode;Output space section passage best configuration.Through the application, satellite transponder multi-satellite signal transmission passage configuration situation can be optimized, and the efficiency of space section passage configuration can be effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of satellite resource overall scheduling, and in particular to a space segment channel configuration auxiliary decision method and system. BACKGROUND

[0002] Satellite resource overall scheduling is a system activity implemented by a satellite operator in a certain window of time for the purpose of guaranteeing the quality of service (Qos) of users, which is a whole planning, system optimization, unified configuration and macro-control system activity implemented by a satellite operator in a certain window of time for the purpose of guaranteeing the quality of service (Qos) of users, in the face of challenges such as limited on-orbit resources of a communication satellite network, channel resources, satellite power consumption constraints, user priority and task burstiness. Multi-satellite networking scheduling can achieve high accuracy and timeliness of resource overall scheduling strategy. How to visualize satellite resource overall scheduling is a problem to be solved. SUMMARY

[0003] The present application provides a space segment channel configuration auxiliary decision method and system, which aims to realize the visualization of space segment channel configuration, can optimize the configuration of satellite transponder multi-satellite signal transmission channels, and can effectively improve the efficiency of space segment channel configuration.

[0004] In a first aspect, a space segment channel configuration auxiliary decision method is provided, comprising:

[0005] Constructing a directed and weighted node-connection graph, wherein the nodes represent satellites, and the directed lines between the nodes represent signal directions;

[0006] According to the connection relationship in the directed and weighted node-connection graph, attribute data is obtained, which includes at least one of the following: carrier signal bandwidth, carrier signal power, carrier signal modulation mode;

[0007] Outputting the optimal configuration of the space segment channel.

[0008] In combination with the first aspect, in some implementations of the first aspect, the outputting of the optimal configuration of the space segment channel comprises:

[0009] By pairwise carrier comparison, a matrix is established to determine n factors of the criterion layer and n weights corresponding to the n factors, the weights being used to represent the influence degree of the factors, and the n factors including at least one of the following: carrier bandwidth, carrier power, carrier priority, carrier modulation mode;

[0010] According to the n weights, the optimal configuration of the space segment channel is outputted.

[0011] In combination with the first aspect, in some implementations of the first aspect, the method further comprises:

[0012] Visual coding through multiple visual cognitive channels according to a specified coding mode.

[0013] With reference to the first aspect, in some implementations of the first aspect, the output space segment channel optimal configuration comprises:

[0014] The output space segment channel optimal configuration is according to a relative proportion of inter-channel connection numbers, the relative proportion of the inter-channel connection numbers being represented by a matrix, 0 in the matrix representing no channel between nodes, and 1 representing a channel between nodes.

[0015] With reference to the first aspect, in some implementations of the first aspect, the output space segment channel optimal configuration comprises:

[0016] The output space segment channel optimal configuration is according to attribute dimensions contained in a satellite carrier signal bearing service relationship, the attribute dimensions comprising at least one of the following: a satellite itself, a transponder quantity, a transponder coverage, and a transponder frequency band.

[0017] With reference to the first aspect, in some implementations of the first aspect, the output space segment channel optimal configuration comprises:

[0018] The overall satellite channel configuration is optimized by an attention flow network, nodes of the attention flow network being satellites, directed edges representing user jumps, and weight values representing attention flow.

[0019] With reference to the first aspect, in some implementations of the first aspect, the attention flow network satisfies at least one of the following:

[0020] The attention flow network is constructed according to custom weight data divided according to sessions;

[0021] The attention flow network is trained according to attribute weight levels, website dwell times, and click counts.

[0022] With reference to the first aspect, in some implementations of the first aspect, the method further comprises:

[0023] Outputting a node importance evaluation, the node importance evaluation being represented by an importance score, the importance score representing a value of importance or popularity of a satellite.

[0024] With reference to the first aspect, in some implementations of the first aspect, the method further comprises:

[0025] Outputting a simulation result, the simulation result comprising at least one of the following:

[0026] Outputting the simulation result to a specified file;

[0027] Visualizing the simulation result in a visual graphical interface;

[0028] The simulation results are outputted in a chart form.

[0029] In a second aspect, a space segment channel configuration decision-making assistance system is provided for performing the method as described in any one of the implementation manners of the first aspect.

[0030] Compared with the prior art, the scheme provided in the application has at least the following beneficial technical effects:

[0031] In view of the problem that a resource overall scheduling cost function is not accurate due to dynamic changes of multiple influence factors such as carrier properties, ground station properties, satellite state, frequency interference, uneven time and space distribution of services, and user use preferences, and the like, of a traditional satellite resource overall scheduling algorithm, an intelligent communication satellite network resource overall scheduling algorithm is developed based on deep reinforcement learning, an intelligent scheduling module is introduced, and a communication satellite network resource overall scheduling model under multi-objective constraints is constructed, so as to overall optimize service allocation, resource deployment, user site distribution, communication demand, and communication satellite network properties, and improve resource utilization while ensuring service quality. A communication satellite network service capability evaluation system based on meta-learning is constructed, the state set, behavior set, cost function, average performance index, and behavior function of the communication satellite network are taken as an evaluation module, and the overall structure of the communication satellite network service capability evaluation system is formed.

[0032] The application constructs an intelligent resource scheduling system based on reinforcement learning, which is used to process changes of satellite coverage, satellite frequency bands, and service properties (for example, priority changes caused by task burstiness). Due to the scaling of satellite communication networks, the complexity of satellite properties (including transponders, satellite coverage, satellite life, and the like), the differentiation of different users and different service demands, the high variability of satellite channel environments, the scarcity of on-board resources, and the difficulty of resource overall scheduling configuration, the difficulty of resource overall scheduling configuration will increase. The system designs a communication satellite network resource overall maximum task priority scheduling algorithm, which first analyzes the overall consideration of the target satellite, that is, whether to first occupy the capacity of a satellite according to the list order, then occupies the capacity of all satellites according to the specified principle, and fills the remaining idle frequency bands according to the list order, then considers whether to preferentially ensure the consistency of frequency points during service transfer, whether to ignore the use constraints of uplink stations during service transfer, and whether to supplement the remaining capacity of the satellite so that all idle carriers can be utilized, the algorithm pursues to obtain the maximum task priority sum of the task, and the task priority sum is proportional to the task benefit.

[0033] The application establishes a user preference-oriented communication satellite network service capability evaluation system. The user preference-oriented communication satellite network service capability evaluation system is customized according to customer needs. First, a communication satellite network service capability evaluation system based on convolutional neural network and meta-learning is constructed. The evaluation module uses different means for information mining according to different magnitudes of data. For example, in the user business request, each user business has different preferences for the resources allocated by the request, including commonly used satellites, commonly used transponders, commonly used frequency points and polarization modes, commonly used uplink ground stations, and ground station weather conditions. Reasonable global decision planning is given in real time, quickly and reliably. Different schemes are scored according to different weights defined by the user, improving the rationality of resource scheduling and providing customized services for users, and better improving the effect of user satisfaction. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 A schematic diagram of a space segment channel configuration auxiliary decision method provided for an embodiment of the application. DETAILED DESCRIPTION

[0035] The application will be described in further detail below with reference to the drawings and specific embodiments.

[0036] Figure 1 A schematic diagram of a space segment channel configuration auxiliary decision method provided for an embodiment of the application.

[0037] Step 1: Visual analysis of space segment transponder configuration data representing complex relationships of satellite transponders using a directed and weighted node-connection graph. The nodes of the directed and weighted node-connection graph can represent satellites, and the directed lines between the nodes can represent signal directions. The transponder configuration can be the number of transponders and the gain setting.

[0038] Step 2: Obtain multi-attribute data through the connection relationship of the space segment transponder. The multi-attribute data can include carrier signal bandwidth, carrier signal power, and carrier signal modulation mode.

[0039] Step 3: Output the optimal configuration of the space segment channel. In one embodiment, the optimal configuration of the space segment channel can be displayed in combination with the directed and weighted node-connection graph.

[0040] The traditional layout drawing algorithm first groups the network nodes, and the nodes in each group use the same vertical coordinate value. Then, the root node (satellite body) of the group is found, and the horizontal coordinate position (carrier destination) is assigned to the nodes in the group according to the position of the root node. The process ends after recursively traversing all nodes. Then, the Graphviz programming technology (rendering environment-code-NULL node filling) can be used to complete the visualization of the complex relationship network channel data and obtain the data visualization structure.

[0041] In the scheme provided in the embodiments of the present application, the traditional layout drawing algorithm can be optimized and designed through the directed weighted model. Compared with the traditional node connection graph layout, the satellite transponder multi-attribute network data has better cognitive efficiency, and can clearly determine the input variables such as the direction of satellite carrier transmission, carrier signal bandwidth, carrier signal power, carrier signal modulation mode and the like.

[0042] The traditional heuristic channel configuration method only considers the graph structure characteristics of the network, only considers the correlation of a pair of variables, lacks the application of implicit characteristic information such as the direction of satellite carrier transmission, carrier signal bandwidth, carrier signal power, carrier signal modulation mode, and most of the methods are based on undirected and unweighted networks.

[0043] In the scheme provided in the embodiments of the present application, when the connection relationship between nodes contains multiple attributes such as start and end direction, weight (weighted according to the importance of the information transmitted by the carrier), carrier bandwidth, carrier power, a directed weighted graph layout as shown in step 1 needs to be used for drawing.

[0044] In the scheme provided in the embodiments of the present application, the weight configuration algorithm can use the AHP hierarchical analysis method. This method is a qualitative and quantitative weight calculation research method, which uses the method of comparing two carriers, establishes a matrix, determines n factors (indicators) in the criterion layer, such as carrier bandwidth, carrier power, carrier priority, carrier modulation mode, and then compares their influence on the target to determine the proportion of each factor in the layer relative to a certain criterion. The principle of the higher the weight, the higher the weight of the vector sum, and finally the importance of each factor is calculated.

[0045] The traditional drawing algorithm models the relationship network through the FR algorithm (i.e. the nodes in the undirected graph are simulated as satellites, and the position relationship between nodes is calculated by simulating the length and direction of the carrier signal between satellites. After continuous iteration calculation, the system finally enters a dynamic balance state), and solves the distribution of the entire relationship network in the carrier vector balance or energy vector balance state to complete the drawing of the graph layout. This method has obvious decline in drawing efficiency and presentation effect when facing large-scale satellite node set network data. In addition, the simple dynamic balance model cannot solve the problem of mutual shielding between connection lines, and it is difficult to meet the drawing requirements of the above multi-attribute association relationship.

[0046] In the scheme provided in the embodiments of the present application, in order to effectively solve the limitations of the simple mechanical model (FR algorithm), the relative proportion of the number of connections between channels and the attribute dimension (represented by weight level) contained in the satellite carrier signal business relationship can be used to complete the drawing requirements of the multi-attribute association relationship of the space segment.

[0047] The relative proportion of the number of inter-channel connections depends on the size of the satellite relationship network (i.e. the number of satellites, the relative positions of the satellites, and the different coverage of the satellites). The number of inter-channel connections is expressed in the form of an n-dimensional matrix. For a large-scale satellite multi-dimensional multi-network channel configuration, the number of inter-channel connections has a direct impact on the drawing efficiency of the connection graph layout.

[0048] In the scheme provided in the embodiments of the present application, when the relationship between satellite nodes is multi-attribute data, a simple node-connection graph layout cannot completely present the data structure information, and more visual cognitive channels need to be mobilized for visual coding, and meanwhile, consistency between the coding channels is ensured by specifying the coding when the database is established, so as to achieve a reasonable directed weighted graph.

[0049] In the scheme provided in the embodiments of the present application, for a large-scale satellite multi-dimensional multi-network channel configuration, the dimensions include the satellite itself, the number of transponders, the coverage of the transponders, and the frequency bands of the transponders. Due to the large number of nodes, for a multi-attribute connection graph, the weight of the directed edge can be used as the main reference factor for the accuracy of visual cognition when drawing the connection.

[0050] In the scheme provided in the embodiments of the present application, based on the deficiencies of the traditional heuristic channel configuration method (which does not comprehensively consider the node attributes and edge attributes in the learning network), an attention flow network can be used to optimize the overall satellite channel configuration. Thus, the flow of attention flow (i.e. vector carrier signal) on the satellite transponder resources can be studied from the overall perspective. It can comprehensively consider the hyperlink structure (matrix size) between satellite resources.

[0051] The so-called attention flow network is a weighted directed graph, in which the nodes are the above-mentioned satellites, the directed edges represent user jumps, and the weight values represent the flow of attention (i.e. how many users jump from one node to another). At the same time, there are two special nodes in the entire network: source and sink, which represent the entrance (representing the carrier ON) and the exit (representing the carrier OFF) of the attention flow.

[0052] In one scheme provided in the embodiments of the present application, the attention flow network can be directly constructed according to the self-defined weight data divided according to the session (usually a fixed time interval, one hour), and the current data is read in the database through the fixed time interval and is processed through coding. In the process of learning node features, the local structure of the force flow network, the attribute weight level, the website dwell time and the click count and other features (obtained through the previous monitoring data point) are considered, which can capture more information about the network and effectively improve the prediction performance, and output the best spatial segment channel auxiliary decision configuration group.

[0053] Before the visual analysis data is displayed, the system basic parameters can be configured.

[0054] For example, nodes and carriers can be configured. Nodes can be specified, i.e., the satellite itself. Carriers typically include two types: randomly generated and specified. Random generation is based on Monte Carlo simulation to randomly sample the system, using the Student-t distribution as a basis, and constructing a quasi-normal distribution according to this model. It involves stratified sampling from the historical carrier data distribution to obtain estimates for configuring the system. Specified generation involves configuring carriers with defined characteristics from external input.

[0055] For example, in the complex satellite network topology generation area setting, users can set the size of the area based on the satellite network composed of all satellites.

[0056] For example, in setting statistical parameters, some constants and variables in the calculated statistical parameters are set, including the confidence interval of the structure, the probability of type I error, the probability of type II error, and the significance level of the sample statistic.

[0057] For example, other parameter settings, including custom inputs, allow you to exclude a specific satellite from the complex satellite network topology generation area.

[0058] For example, topology interface configuration mainly involves configuring some functions of the graphical display interface, including satellite node size and color, edge color, node label font and color, edge thickness display, etc., with the aim of providing users with a convenient and user-friendly interface.

[0059] In some embodiments, node importance evaluation can be implemented as a separate module. Evaluating the importance of nodes in a directed weighted complex satellite network involves identifying the n most important satellites for the current configuration, specifically including important node mining algorithms. The important node mining algorithm module incorporates several classic degree-based and betweenness-based important node mining algorithms.

[0060] This application proposes the GENI model based on GNN networks. Given the importance scores of certain nodes in a complex network, the importance score represents the importance or popularity of a satellite, such as satellite nodes that attract a lot of attention and search traffic (obtained through front-end monitoring data points). Therefore, the page views on the front-end page can be used as the importance score of the corresponding entity. Based on the optimal space segment channel auxiliary decision configuration group obtained in step 7, a correlation analysis is performed with the obtained important nodes to output the node importance evaluation.

[0061] In some embodiments, three output modes can be adopted to output the simulation results. Firstly, the results are output to a specified file: some calculated parameter results and information in the simulation process are saved to a specified file, facilitating company research and analysis and post-processing; secondly, the results are displayed in a visual graphical interface: after the simulation is completed, the structure can be directly displayed in a complex network topology window; thirdly, various analysis data are displayed in the form of charts: after a large amount of simulation is performed, the statistical results can be displayed in the form of charts for analysis and comparison, and the simulation platform can make charts according to the setting of chart types.

[0062] In some embodiments, according to the processed data, a new satellite channel configuration test log file can be generated, and after the compression and dumping work is completed, the processed test log file is marked and updated in the operation management file. Specifically, according to a preset minimum saving time m of original test data, the data before m days before the processing time is compressed when the test time is performed, and the data before m days before the processing time is not included, wherein: in a preset saving time interval Ts, if there is more than one test record, the first test record is saved and the remaining test records are deleted.

[0063] The application also provides a space segment channel configuration auxiliary decision-making system, which comprises a target weight level, a weight level sorting, channel configuration information, a spectrum inspection and carrier analysis server, a spectrum front-end interface, a storage module, a spectrum data compression management module, and a carrier frequency domain distribution characteristic parameter analysis module.

[0064] Although the present application has been disclosed with reference to the preferred embodiments above, it is not intended to limit the present application, and any person skilled in the art can make possible changes and modifications to the technical solutions of the present application by using the disclosed methods and technical contents without departing from the spirit and scope of the present application. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application, which does not deviate from the technical solutions of the present application, belongs to the protection scope of the present application. The contents not described in detail in the specification of the present application are known to those skilled in the art.

Claims

1. A spatial segment channel configuration auxiliary decision-making method, characterized in that, include: Construct a directed weighted node-connection graph, where nodes represent satellites and directed line segments between nodes represent signal directions; Based on the connection relationships in the directed weighted node-connection graph, attribute data is obtained, including: carrier signal bandwidth, carrier signal power, and carrier signal modulation method. The optimal configuration of the output spatial segment channel includes: establishing a matrix by comparing each pair of carriers, determining n factors of the criterion layer and n weights corresponding to the n factors, wherein the weights are used to represent the degree of influence of the factors, and the n factors include: carrier bandwidth, carrier power, carrier priority, and carrier modulation method; and outputting the optimal configuration of the spatial segment channel according to the n weights. The method further includes: based on the relative proportion of the number of connections between channels and the attribute dimensions included in the service relationship undertaken by the satellite carrier signal, completing the drawing requirements of the multi-attribute association relationship of the space segment configuration. The relative proportion of the number of connections between channels is represented by a matrix, where 0 indicates that there is no channel between nodes and 1 indicates that there is a channel between nodes. The attribute dimensions include at least one of the following: satellite itself, number of transponders, transponder coverage, and transponder frequency band. The overall satellite channel configuration is optimized through an attention flow network. The nodes of the attention flow network are satellites, the directed edges represent user jumps, and the weight values ​​represent the attention flow. The attention flow network has two nodes, a source and a sink, which represent the inlet and outlet of attention flow, respectively. The inlet of attention flow represents carrier ON, and the outlet of attention flow represents carrier OFF. The attention flow network is constructed according to the custom weight data of the session division; the attention flow network is trained according to the attribute weight level, the time spent on the website, and the number of clicks. It reads the current data from the database at fixed time intervals, encodes it, and outputs the optimal spatial segment channel auxiliary decision configuration group.

2. The method according to claim 1, characterized in that, The method further includes: Visual encoding is performed through multiple visual cognitive channels according to a specified encoding method.

3. The method according to claim 1, characterized in that, The method further includes: Output node importance evaluation, which is represented by an importance score, representing the value of the satellite's importance or popularity.

4. The method according to claim 1, characterized in that, The method further includes: Output simulation results, including at least one of the following: Output the simulation results to the specified file; The simulation results are presented in a visual graphical interface; The simulation results are output in the form of charts.

5. A spatial segment channel configuration auxiliary decision-making system, characterized in that, Used to perform the method as described in any one of claims 1 to 4.

Images