Data building method and system based on star-ground collaborative computing

Abstract

The embodiment of the application discloses a data building method and system based on star-ground cooperative calculation, and belongs to the technical field of data processing. The embodiment of the application can comprehensively master the condition of the star-ground cooperative network matrix by acquiring ground side sensing data and satellite side state data. Mining coding vectors can effectively extract data features and reduce data redundancy, providing an accurate basis for subsequent integration. The first vector integration and the second vector integration adopt different interaction strategies, which can deeply fuse star-ground data, so that the obtained star-ground network embedding integration vector can fully reflect the complex relationship of the star-ground cooperative network. The decision information obtained by judging the data building situation according to the vector is helpful to optimize the data building of the star-ground cooperative network matrix, improve the rationality of network resource allocation, enhance the ability of the network to respond to complex environmental changes, and ensure the efficient and stable operation of the star-ground cooperative network.

Description

Technical Field

[0001] This application belongs to the field of data processing technology, specifically relating to a data construction method and system based on satellite-ground collaborative computing. Background Technology

[0002] In the field of space-ground collaborative networks, with the continuous development of satellite and terrestrial network technologies, effectively managing and optimizing the data construction of the space-ground collaborative network matrix has become an urgent technical problem to be solved. Traditional network management methods struggle to comprehensively consider the status information of both terrestrial network equipment and satellites, resulting in insufficient data integration and an inability to accurately reflect the overall status of the space-ground collaborative network. For example, in terms of resource allocation, the lack of consideration for the overall situation of the space and ground systems can easily lead to resource waste or resource shortages in certain areas. Furthermore, traditional methods cannot adapt to dynamic changes in the network in a timely manner. Summary of the Invention

[0003] This application provides a data construction method and system based on satellite-ground collaborative computing, which can solve the technical problems involved in the background art.

[0004] This application provides a data construction method based on satellite-ground collaborative computing, applied to a satellite-ground collaborative data construction system. The method includes: acquiring ground-side sensing data and multiple sets of satellite-side status data corresponding to a satellite-ground collaborative network matrix, wherein the ground-side sensing data is monitoring data of ground network equipment associated with the satellite-ground collaborative network matrix; mining ground network embedding encoding vectors corresponding to the ground-side sensing data and mining satellite network embedding encoding vectors corresponding to the multiple sets of satellite-side status data; performing a first vector integration on the multiple satellite network embedding encoding vectors and the ground network embedding encoding vectors respectively to obtain a first satellite-ground network embedding integration vector corresponding to the multiple satellite network embedding encoding vectors respectively; performing a second vector integration on the first satellite-ground network embedding integration vectors corresponding to the multiple satellite network embedding encoding vectors respectively to obtain a second satellite-ground network embedding integration vector corresponding to the satellite-ground collaborative network matrix, wherein the vector feature interaction strategies of the first vector integration and the second vector integration are different; and performing data construction status judgment on the satellite-ground collaborative network matrix based on the second satellite-ground network embedding integration vector to obtain construction status decision information corresponding to the satellite-ground collaborative network matrix.

[0005] This application provides a satellite-ground collaborative data construction system, including at least one processor and a memory; the memory stores computer execution instructions; the at least one processor executes the computer execution instructions stored in the memory, causing the at least one processor to perform the above-described method.

[0006] This application provides a readable storage medium on which a program or instruction is stored, and when the program or instruction is executed by a processor, it implements the steps of the above method.

[0007] This application embodiment, by acquiring ground-side sensing data and satellite-side status data, can comprehensively grasp the status of the space-ground cooperative network matrix. Mining coded vectors can effectively extract data features, reduce data redundancy, and provide a precise foundation for subsequent integration. The first vector integration and the second vector integration employ different interaction strategies, enabling deep fusion of space-ground data. This allows the resulting space-ground network embedding integration vector to fully reflect the complex relationships within the space-ground cooperative network. Decision information derived from data construction and situational assessment based on this vector helps optimize the data construction of the space-ground cooperative network matrix, improves the rationality of network resource allocation, enhances the network's ability to cope with complex environmental changes, and ensures the efficient and stable operation of the space-ground cooperative network. Attached Figure Description

[0008] Figure 1 This is a flowchart illustrating a data construction method based on satellite-ground collaborative computing provided in an embodiment of this application.

[0009] Figure 2 This is a schematic diagram of the structure of a satellite-ground collaborative data construction system provided in an embodiment of this application. Detailed Implementation

[0010] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of the embodiments of this application.

[0011] In the embodiments of this application, the terms "first," "second," etc., are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, the first object can be one or more. Furthermore, in the embodiments of this application, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects have an "or" relationship.

[0012] Figure 1 A data construction method based on satellite-ground collaborative computing is shown, which is applied to a satellite-ground collaborative data construction system. The method includes the following steps 101-105.

[0013] Step 101: The satellite-ground collaborative data construction system acquires ground-side sensing data and multiple sets of satellite-side status data corresponding to the satellite-ground collaborative network matrix. The ground-side sensing data is monitoring data of ground network equipment associated with the satellite-ground collaborative network matrix.

[0014] Step 102: The satellite-ground collaborative data construction system mines the ground network embedding coding vectors corresponding to the ground-side sensing data, and mines the satellite network embedding coding vectors corresponding to the multiple sets of satellite-side status data.

[0015] Step 103: The satellite-ground collaborative data construction system performs a first vector integration on the multiple satellite network embedding coding vectors and the ground network embedding coding vectors respectively to obtain the first satellite-ground network embedding integration vectors corresponding to the multiple satellite network embedding coding vectors respectively.

[0016] Step 104: The satellite-ground collaborative data construction system performs a second vector integration on the first satellite-ground network embedding integration vectors corresponding to the multiple satellite network embedding coding vectors, to obtain the second satellite-ground network embedding integration vector corresponding to the satellite-ground collaborative network matrix. The vector feature interaction strategies of the first vector integration and the second vector integration are different.

[0017] Step 105: The satellite-ground collaborative data construction system performs data construction status judgment on the satellite-ground collaborative network matrix based on the second satellite-ground network embedding integration vector, and obtains the construction status decision information corresponding to the satellite-ground collaborative network matrix.

[0018] In this embodiment, the satellite-ground collaborative network matrix is ​​a network architecture that organically combines satellite and terrestrial networks. In this architecture, specific connections, communication links, and interaction modes exist between the nodes of the satellite and terrestrial networks. It forms the fundamental framework of the entire satellite-ground collaborative data construction system, providing structural support for data flow, interaction, and related operations.

[0019] First, in step 101, the space-ground collaborative data construction system begins acquiring relevant data. This involves ground-side sensing data and multiple sets of satellite-side status data. Ground-side sensing data refers to monitoring data of ground network equipment associated with the space-ground collaborative network matrix. For example, ground network equipment monitoring data can include operational status data of ground base stations, such as base station temperature (e.g., normal operating temperature range between 10-35 degrees Celsius), power consumption (potentially consuming 1-5 kilowatts per hour, varying depending on equipment size and traffic volume), and the number of user connections within the signal coverage area (potentially reaching thousands of user connections during peak periods). This data reflects the actual operation of ground network equipment within the space-ground collaborative network matrix. Satellite-side status data can include satellite orbital position information (e.g., an orbital altitude of approximately 500 kilometers and an orbital inclination of 45 degrees), the operational status of various sensors on the satellite (e.g., a sensor sampling frequency of 10 times per second), and the satellite's energy reserves (e.g., initial energy reserves can support its operation for 5 years, gradually depleting with use), among other data.

[0020] Next, in step 102, the satellite-ground collaborative data construction system performs data mining operations on the acquired data. For ground-side sensing data, the system mines the corresponding ground network embedding coding vector. This coding vector is a representation of the ground-side sensing data after feature extraction and encoding. For example, data such as the temperature, power consumption, and number of user connections of ground base stations can be encoded according to a certain algorithm. For example, temperature values ​​can be encoded in intervals of 5 degrees Celsius, power consumption in intervals of 1 kilowatt, and the number of user connections in intervals of 1000. This transforms the complex status information of ground base stations into a specific coding vector. Similarly, for satellite-side status data, the system mines the satellite network embedding coding vector corresponding to each set of satellite-side status data. Taking satellite orbital position, sensor operating status, and energy reserves as examples, a similar interval division method can be used for encoding. For example, orbital altitude can be encoded in intervals of 100 kilometers, sensor sampling frequency in intervals of 5 times per second, and energy reserves in intervals based on the annual consumption ratio, thus obtaining the satellite network embedding coding vector.

[0021] Next, in step 103, the satellite-ground collaborative data construction system performs first vector integration on multiple satellite network embedded coding vectors and ground network embedded coding vectors respectively, resulting in first satellite-ground network embedded integrated vectors corresponding to the multiple satellite network embedded coding vectors. Vector integration here is an operation that merges coding vectors from different sources. For example, there are three sets of satellite network embedded coding vectors: coding vector V1 corresponding to satellite 1, coding vector V2 corresponding to satellite 2, coding vector V3 corresponding to satellite 3, and ground network embedded coding vector G. During the first vector integration, a weighted summation method may be used (this is just an example; a more complex algorithm may be used in practice). For example, assigning a weight of 0.3 to V1, 0.3 to V2, 0.3 to V3, and 0.1 to G, then the first satellite-ground network embedded integrated vector I1 corresponding to satellite 1 is I1 = 0.3V1 + 0.1G, the first satellite-ground network embedded integrated vector I2 corresponding to satellite 2 is I2 = 0.3V2 + 0.1G, and the first satellite-ground network embedded integrated vector I3 corresponding to satellite 3 is I3 = 0.3V3 + 0.1G. This integration method initially merges the characteristics of satellite networks and terrestrial networks, forming a first satellite-terrestrial network embedding integration vector for each satellite.

[0022] Subsequently, in step 104, the satellite-ground collaborative data construction system performs a second vector integration on the first satellite-ground network embedding integration vectors corresponding to the multiple satellite network embedding coding vectors, resulting in the second satellite-ground network embedding integration vector corresponding to the satellite-ground collaborative network matrix. Here, the vector feature interaction strategies for the first and second vector integrations differ. In the second vector integration process, a different weighting method or fusion algorithm than the first vector integration may be used. For example, continuing with the first satellite-ground network embedding integration vectors I1, I2, and I3 corresponding to satellites 1, 2, and 3, a more complex nonlinear fusion algorithm may be used in the second vector integration. For instance, based on the overall structural characteristics of the satellite-ground collaborative network matrix and the importance distribution of the data, I1, I2, and I3 may be recombined and fused. Some features in I1 may be amplified (e.g., multiplied by 2), some features in I2 may be reduced (e.g., multiplied by 0.5), and some features in I3 may be adjusted (e.g., taken as their square root). Then, these adjusted features are summed or subjected to other complex combination operations to finally obtain the second satellite-ground network embedding integration vector J corresponding to the satellite-ground collaborative network matrix.

[0023] Finally, in step 105, the satellite-ground collaborative data construction system performs data construction status judgment on the satellite-ground collaborative network matrix based on the second satellite-ground network embedding integration vector, obtaining the construction status decision information corresponding to the satellite-ground collaborative network matrix. This data construction status judgment is based on the comprehensive feature information contained in the second satellite-ground network embedding integration vector to determine the current state and development trend of the satellite-ground collaborative network matrix. For example, if some feature values ​​in the second satellite-ground network embedding integration vector J indicate a large rate of change in the satellite's orbital position (exceeding a preset threshold, such as a 0.1-degree change in orbital angle per day), and simultaneously the number of user connections at ground base stations is growing rapidly (more than 100 user connections per hour), then the system may determine that the satellite-ground collaborative network matrix is ​​in a rapidly changing state, requiring adjustments to the data construction strategy, such as increasing the data transmission bandwidth between the satellite and ground base stations (from the original 10Mbps to 20Mbps) or optimizing data transmission routing. This construction status decision information provides a basis for decision-making regarding the data construction of the satellite-ground collaborative network matrix, ensuring that the entire satellite-ground collaborative data construction system can effectively manage data and optimize the network according to the actual situation.

[0024] The satellite-ground collaborative data construction system, through the above series of steps, comprehensively performs data construction-related operations on the satellite-ground collaborative network matrix, from data acquisition, encoding vector mining, vector integration to final situational assessment. This provides effective technical support for the stable operation, optimized development, and efficient data management of the satellite-ground collaborative network.

[0025] To further elaborate on this process, the richness and accuracy of the data acquired in step 101 are crucial for the implementation of the entire invention application. In addition to the typical data types mentioned earlier, ground-side sensing data may also include information such as link latency between ground network devices (under normal circumstances, link latency may be between 1 and 10 milliseconds, varying between different links due to distance and device performance), and data packet loss rate (typically expected to be below 1%). Satellite-side status data may also cover the occupancy of satellite communication frequency bands (e.g., a frequency band utilization rate of approximately 30%), and the remaining capacity of data storage devices on the satellite (e.g., total capacity 1TB, remaining capacity 0.3TB), etc. Accurate acquisition of this data provides a comprehensive foundation for subsequent operations.

[0026] In step 102, when mining the encoding vector, various data mining techniques can be employed to improve the effectiveness of the encoding. For example, for mining terrestrial network embedding encoding vectors, cluster analysis can be used to classify terrestrial network devices with similar characteristics, and then encoding can be performed based on the classification results. For instance, terrestrial base stations can be divided into high, medium, and low categories according to the population density of their coverage areas, and encoded using numbers 1, 2, and 3 respectively. Then, other data features (such as temperature and power consumption) can be combined to form a complete terrestrial network embedding encoding vector. For mining satellite network embedding encoding vectors, the special characteristics of satellites, such as their orbital type (geostationary orbit, low Earth orbit, etc.), may need to be considered. Satellites with different orbital types may have different weights or encoding rules during encoding to reflect their different roles in the satellite-ground cooperative network matrix.

[0027] In the first vector integration process of step 103, besides the weighted summation approach, a neural network-based integration method can also be used. For example, a feedforward neural network can be constructed, using satellite network embedding vectors and terrestrial network embedding vectors as inputs to neurons in the input layer. Data processing is then performed through the hidden layers of the neural network (e.g., a hidden layer with three neurons, each using a different activation function such as sigmoid, tanh, and the modified linear unit function (ReLU)). Finally, the first satellite-terrestrial network embedding integration vector is obtained at the output layer. This approach can better uncover the nonlinear relationships between data, improving the effectiveness of vector integration.

[0028] In the second vector integration step 104, the differences in vector feature interaction strategies become more apparent. An optimization strategy based on genetic algorithms can be used to determine the interaction methods between different vector features. For example, each feature in the first satellite-to-ground network embedding integration vector can be considered as a gene in a genetic algorithm. Through selection, crossover, and mutation operations, the optimal feature combination can be found to construct the second satellite-to-ground network embedding integration vector. This approach can adaptively adjust the vector integration strategy according to the complex environment and requirements of the satellite-to-ground cooperative network matrix, improving the system's flexibility and adaptability.

[0029] In step 105, the second satellite-to-ground network embedding integration vector, which forms the basis for situational awareness determination, contains rich information. Besides the previously mentioned information such as satellite orbital position and the number of ground base station user connections, it may also include the bit error rate of data transmission between the satellite and the ground network (e.g., the bit error rate should be below 0.01% under normal circumstances), the overall load balancing of the network (assessed by calculating the load ratio of different nodes), etc. When making situational awareness determination based on this information, a fuzzy logic approach can be used. For example, the values ​​of each feature can be mapped to different fuzzy sets (e.g., "low," "medium," "high," etc.), and then a judgment can be made according to predefined fuzzy rules (e.g., "if the rate of change of satellite orbital position is high and the growth of the number of ground base station user connections is high, then the situation is rapid expansion") to obtain accurate situational awareness decision information. This fuzzy logic method can handle some imprecise and uncertain information and is better adapted to the complex and ever-changing reality of satellite-to-ground cooperative network matrices.

[0030] Throughout the implementation of this invention application, the satellite-ground collaborative data construction system manages and optimizes the data construction of the satellite-ground collaborative network matrix from multiple perspectives through carefully designed steps, ensuring that the system can operate effectively and make reasonable decisions in different operating environments.

[0031] To elaborate further on the details of each step, in step 101, the process by which the satellite-ground collaborative data construction system acquires ground-side sensing data corresponding to the satellite-ground collaborative network matrix and multiple sets of satellite-side status data is a complex multi-source data acquisition process. For ground-side sensing data, the system needs to establish reliable communication connections with numerous ground network devices to acquire the data. For example, ground network devices may be distributed in different geographical locations, from city centers to remote mountainous areas, and the system needs to use multiple communication protocols (such as TCP / IP) to ensure accurate data transmission. When acquiring base station temperature data, the analog signal may be converted into a digital signal by a temperature sensor and then sent to the satellite-ground collaborative data construction system via the base station's internal communication module. For satellite-side status data, the system needs to communicate with the satellite. The satellite sends its own status data to the ground receiving station via the satellite communication link, and then forwards it to the satellite-ground collaborative data construction system. During this process, data transmission may be affected by various factors, such as atmospheric interference and solar activity. To ensure data integrity, data verification and error correction techniques, such as Cyclic Redundancy Check (CRC), may be used.

[0032] In step 102, the process of mining encoded vectors involves a deep understanding of the data and feature extraction. For mining terrestrial network embedded encoded vectors, in addition to the previously mentioned clustering analysis and interval partitioning coding methods, Principal Component Analysis (PCA) can also be used. PCA helps identify the main components in terrestrial sensing data, reducing high-dimensional data to a low-dimensional space, and then encoding it in that lower-dimensional space. For example, when processing various state data (such as temperature, power consumption, number of user connections, link latency, etc.) involving multiple terrestrial base stations, PCA can find the principal components that best represent the changes in these data, and then encode them based on the values ​​of these principal components. For mining satellite network embedded encoded vectors, considering the spatiotemporal characteristics of satellite data, wavelet analysis can be used. Wavelet analysis can analyze the characteristics of satellite data at different scales and time frequencies, such as the long-term and short-term trends of satellite orbital positions and the periodic changes in sensor operating states. Encoding is then performed based on the results of wavelet analysis, resulting in encoded vectors that more accurately reflect the satellite-side state data.

[0033] In the first vector ensemble in step 103, in addition to the weighted summation and neural network methods mentioned earlier, the idea of ​​Support Vector Machine (SVM) can also be used to construct the ensemble model. The satellite network embedding encoding vector and the terrestrial network embedding encoding vector are considered as different feature vectors, and an optimal hyperplane is found through SVM to integrate these vectors. In this process, the kernel function of the SVM (such as a linear kernel function, Gaussian kernel function, etc.) needs to be determined according to the distribution characteristics of the data. For example, if the data exhibits linearly separable features in the feature space, a linear kernel function can be chosen; if the data exhibits a non-linear distribution, a Gaussian kernel function may be chosen. This SVM-based vector ensemble method can improve the generalization ability of vector ensemble to a certain extent, making the obtained first satellite-terrestrial network embedding ensemble vector more representative.

[0034] In the second vector integration step 104, the design of the vector feature interaction strategy is based on the consideration of optimizing the overall performance of the satellite-ground cooperative network matrix. Besides optimization strategies based on genetic algorithms, a particle swarm optimization (PSO) approach can also be used. PSO treats the features embedded in the integration vector of each first satellite-ground network as the position and velocity of particles, and finds the optimal feature combination by observing the movement of the particle swarm in the search space. In the PSO algorithm, particles continuously adjust their position and velocity based on their own experience and the experience of the swarm, ultimately finding the optimal solution. This method can quickly converge to a better vector feature interaction strategy, improving the efficiency and effectiveness of the second vector integration.

[0035] In step 105, when determining the data construction situation of the satellite-ground cooperative network matrix based on the second satellite-ground network embedding integration vector, in addition to the fuzzy logic method, a decision tree model can also be used. Each feature in the second satellite-ground network embedding integration vector is used as an input node of the decision tree, and the decision tree model is constructed by learning from a large amount of historical data. For example, based on past satellite-ground cooperative network matrix operation data, a decision tree is constructed, where internal nodes represent different feature judgment conditions (such as whether the rate of change of satellite orbit position is greater than a certain threshold), and leaf nodes represent different situation outcomes (such as stable situation, expanding situation, contracting situation, etc.). When a new second satellite-ground network embedding integration vector is input, the corresponding construction situation decision information is obtained through the decision path of the decision tree. This decision tree model has the advantages of being intuitive, easy to understand and interpret, and can provide a clear decision basis for the data construction of the satellite-ground cooperative network matrix.

[0036] By continuously optimizing the technical means and methods in each step, the satellite-ground collaborative data construction system can more accurately manage the data construction of the satellite-ground collaborative network matrix, improve the operational efficiency and stability of the entire satellite-ground collaborative network, and provide a solid technical guarantee for future satellite-ground collaborative communication and data interaction.

[0037] In summary, this application's embodiments, by acquiring ground-side sensing data and satellite-side status data, can comprehensively grasp the status of the space-ground cooperative network matrix. Mining coded vectors effectively extracts data features, reduces data redundancy, and provides a precise foundation for subsequent integration. The first and second vector integrations employ different interaction strategies, enabling deep fusion of space-ground data. This allows the resulting space-ground network embedding integration vector to fully reflect the complex relationships within the space-ground cooperative network. Decision information derived from data construction and situational assessment based on these vectors helps optimize the data construction of the space-ground cooperative network matrix, improves the rationality of network resource allocation, enhances the network's ability to cope with complex environmental changes, and ensures the efficient and stable operation of the space-ground cooperative network.

[0038] In a preferred embodiment, the step of performing a first vector integration on the plurality of satellite network embedding coding vectors and the ground network embedding coding vectors respectively to obtain a first satellite-ground network embedding integration vector corresponding to the plurality of satellite network embedding coding vectors respectively includes: performing network state attribute fusion on the plurality of satellite network embedding coding vectors and the ground network embedding coding vectors respectively to obtain a first satellite-ground network embedding integration vector corresponding to the plurality of satellite network embedding coding vectors respectively.

[0039] Optionally, before fusing the network state attributes of the plurality of satellite network embedding coding vectors with the ground network embedding coding vectors to obtain the first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding coding vectors, the method further includes: performing vector pooling processing on the plurality of satellite network embedding coding vectors to obtain satellite network embedding pooled vectors corresponding to the plurality of satellite network embedding coding vectors, wherein the vector pooling processing is used to compress the feature size corresponding to the satellite network embedding coding vectors. Based on this, fusing the network state attributes of the plurality of satellite network embedding coding vectors with the ground network embedding coding vectors to obtain the first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding coding vectors includes: fusing the network state attributes of the satellite network embedding pooled vectors corresponding to the plurality of satellite network embedding coding vectors with the ground network embedding coding vectors to obtain the first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding coding vectors.

[0040] In the preferred embodiment described above, the core of the process of performing a first vector integration on multiple satellite network embedding coding vectors and terrestrial network embedding coding vectors to obtain a first satellite-terrestrial network embedding integration vector corresponding to each of the multiple satellite network embedding coding vectors lies in the fusion of network state attributes.

[0041] First, before fusing network state attributes, there is an operation that performs vector pooling on multiple satellite network embedding vectors. The purpose of vector pooling is to compress the feature size corresponding to the satellite network embedding vectors, which is an important step in optimizing data structure and improving processing efficiency.

[0042] Taking a satellite network embedding coding vector as an example, it might contain 100 feature dimensions. These feature dimensions cover various aspects of the satellite's state information. For instance, the satellite's orbital position information might occupy 5 dimensions, representing the three-dimensional coordinates of the orbit and the two components of the orbital velocity. The satellite might be equipped with 20 sensors, each with its operating status represented by 2 dimensions, resulting in a total of 40 dimensions describing the sensor status. The satellite's energy reserves are represented by 5 dimensions indicating the status of different energy reserve modules. Additionally, the satellite's communication frequency band occupancy is represented by 10 dimensions indicating the occupancy ratio of different frequency bands, and other dimensions are used to represent other satellite state information, thus forming this vector of 100 feature dimensions.

[0043] Vector pooling can compress this 100-feature-dimension vector into, for example, a 30-feature-dimension satellite network embedding pooled vector. This compression process is carefully designed and based on in-depth analysis and selection of the importance of features in the satellite network embedding encoding vector. For example, in the representation of satellite sensor states, there may be some sensor state dimensions that are relatively minor in the current network state. For instance, some sensor state dimensions that are only used for auxiliary monitoring and have little impact on the situational awareness of the satellite-ground cooperative network matrix data construction can be merged or discarded. Sensor state dimensions that monitor minor changes in the satellite's internal environment have low importance in the overall network resource allocation and data construction situational awareness, and can be discarded, thus retaining the feature dimensions that are most critical to the situational awareness of the satellite-ground cooperative network matrix data construction, such as sensor state dimensions that directly affect communication and data transmission.

[0044] Next, network state attribute fusion is performed. Taking the previously obtained satellite network embedding pooling vector and terrestrial network embedding coding vector as examples, network state attribute fusion is an operation that deeply combines satellite networks and terrestrial networks at the network state attribute level.

[0045] The terrestrial network embedding coding vector also contains a wealth of information. Taking a terrestrial base station as an example, its terrestrial network embedding coding vector includes information on various aspects such as the base station's temperature, power consumption, the number of user connections within its signal coverage area, and the link status between the terrestrial base station and other base stations. Specifically, the base station's temperature may be represented by three dimensions to indicate the temperature of different parts. For example, one dimension represents the temperature of the core components of the base station equipment, one dimension represents the temperature of the base station's outer casing, and another dimension represents the temperature near the base station's heat dissipation components. Power consumption is represented by two dimensions to indicate the power consumption of different modules. For example, one dimension represents the power consumption of the signal transmission module, and another dimension represents the power consumption of the data processing module inside the base station. The number of user connections within the signal coverage area is represented by five dimensions to indicate the distribution of user connections at different times. For example, one dimension represents the number of user connections during the morning peak hours, one dimension represents the number of user connections during the midday hours, and so on. The link status between the terrestrial base station and other base stations is represented by ten dimensions to indicate the link's bandwidth, latency, packet loss rate, etc. For example, five dimensions are used to indicate the bandwidth of the link with adjacent base stations, three dimensions represent the link latency, and two dimensions represent the packet loss rate.

[0046] When fusing network state attributes, the satellite orbital position information embedded in the satellite network's pooled vector is correlated and fused with the base station geographical location information embedded in the ground network's coded vector. For example, a satellite with an orbital altitude of 500 kilometers and an orbital inclination of 45 degrees can cover the area where a specific ground base station is located, such as a base station located at longitude 120 degrees and latitude 30 degrees. This coverage relationship attribute will be reflected in the first integrated satellite-ground network embedding vector after fusion, possibly represented by a specific coded value or attribute identifier. This coded value or attribute identifier can be a numerical value or symbol generated according to predefined rules; for example, if the satellite orbit completely covers the area where the base station is located, it is represented by the value 1, and if it partially covers the area, it is represented by 0.5, etc.

[0047] For example, the energy reserves of satellites can be considered in conjunction with the power consumption of ground base stations. For instance, a satellite's initial energy reserves might support its operation for 5 years, and with the current remaining energy, it is expected to support it for another 2 years. Meanwhile, the power consumption of a ground base station fluctuates between 1 and 5 kilowatts per hour. When satellite energy reserves are low, ground base stations can appropriately adjust their power consumption strategies to reduce their dependence on the overall resources of the satellite-ground collaborative network. This relationship will be reflected in the fused vector as an attribute reflecting the resource coordination strategy. For example, an attribute value range can be set. If satellite energy reserves are low and ground base station power consumption is high, this attribute value might be in a higher range, such as 0.8-1, indicating that a power consumption strategy adjustment is needed as soon as possible. If satellite energy reserves are sufficient and ground base station power consumption is within the normal range, this attribute value might be in the range of 0.2-0.5, indicating that the current resource status is good and no adjustment is needed.

[0048] The operational status of satellite sensors and the number of user connections within the signal coverage area of ​​ground base stations are also analyzed through attribute fusion. For example, satellite sensors can detect environmental information about the ground area, such as atmospheric humidity fluctuating between 30% and 80%, while the number of user connections at ground base stations may show different trends under different atmospheric humidity conditions. During the fusion process, this correlation between environmental factors and the number of user connections is incorporated into the first satellite-to-ground network embedding integration vector. For instance, when the atmospheric humidity is in the 30%-50% range, a certain attribute value in the first satellite-to-ground network embedding integration vector obtained through fusion will indicate to the ground base station that it should anticipate a potential increase or decrease in the number of user connections, thereby making adjustments to resource allocation in advance. For example, if the attribute value indicates a potential increase in the number of user connections, the ground base station can increase signal transmission power or optimize signal allocation strategies in advance to meet the potentially increased user demand.

[0049] Similarly, the occupancy of satellite communication frequency bands and the link status between ground base stations are also deeply integrated into network state attributes. For example, if the satellite communication frequency band occupancy rate is around 30%, the link bandwidth between the ground base station and its neighboring base station is 10 Mbps, the latency is 5 milliseconds, and the packet loss rate is 0.5%, then if the satellite communication frequency band occupancy is tight, the fused vector will reflect the requirements for adjusting the link status of the ground base station. For example, the ground base station may need to adjust its data transmission frequency band or optimize link routing to avoid conflicts with satellite communication or to better utilize satellite communication resources. If the satellite communication frequency band occupancy rate exceeds 50%, a certain attribute value in the fused vector may trigger the ground base station to adjust its link strategy, such as adjusting the link bandwidth from 10 Mbps to 8 Mbps, to reduce potential interference with the satellite communication frequency band, while simultaneously optimizing internal data processing flows and reducing the packet loss rate.

[0050] This fusion is comprehensive and multi-dimensional. By fusing network state attributes through satellite network embedded pooling vectors and terrestrial network embedded coding vectors, the resulting first satellite-terrestrial network embedded integrated vectors, corresponding to multiple satellite network embedded coding vectors, can fully reflect the comprehensive state attributes of the satellite-terrestrial collaborative network. It is no longer a simple patchwork of satellite and terrestrial network data, but rather an organic combination at the network state attribute level, deeply exploring the intrinsic connection between the two and embodying this connection in an orderly manner in the integrated vector.

[0051] The aforementioned technical solution provides a more accurate and effective data foundation for the entire satellite-ground collaborative network data construction. In the subsequent second vector integration and the final satellite-ground collaborative network matrix data construction and situational assessment processes, based on this first satellite-ground network embedded integration vector that accurately reflects the comprehensive state attributes of the satellite-ground collaborative network, the operational status of the satellite-ground collaborative network can be grasped more precisely, enabling more rational resource allocation decisions, optimizing the overall architecture and operational efficiency of the satellite-ground collaborative network, and improving its ability to cope with complex environmental changes. This enhances the stability and adaptability of the satellite-ground collaborative network, ensuring its efficient and stable operation in various complex working scenarios.

[0052] As can be seen, based on the above technical solution, vector pooling reduces data redundancy and improves data processing speed and efficiency by compressing the feature size of the satellite network embedded coding vector. Numerically, this is reflected in compressing the original 100 feature dimensions into 30 dimensions, significantly reducing the amount of data. Secondly, the network state attribute fusion comprehensively and deeply considers the interrelationships between various state attributes of the satellite network and the terrestrial network. For example, it integrates satellite orbit and base station location, satellite energy and base station power consumption, satellite sensor status and base station user connection count, satellite communication frequency band and base station link status, and other aspects. This allows the resulting first satellite-ground network embedded integrated vector to accurately and comprehensively reflect the overall state of the satellite-ground collaborative network. Subsequent operations based on such a vector help to more accurately grasp the operational status of the satellite-ground collaborative network, providing a reliable basis for resource allocation and network architecture optimization, thereby improving the stability, adaptability, and overall operational efficiency of the satellite-ground collaborative network and ensuring its reliable operation in complex environments.

[0053] In one exemplary technical solution, the step of performing a second vector integration on the first satellite-ground network embedding integration vectors corresponding to the plurality of satellite network embedding coding vectors to obtain the second satellite-ground network embedding integration vector corresponding to the satellite-ground cooperative network matrix includes: inputting the first satellite-ground network embedding integration vectors corresponding to the plurality of satellite network embedding coding vectors into a depth residual model to generate the second satellite-ground network embedding integration vector corresponding to the satellite-ground cooperative network matrix.

[0054] It is understandable that the core of the process of performing a second vector integration on the first satellite-ground network embedding integration vector corresponding to the embedding coding vectors of multiple satellite networks to obtain the second satellite-ground cooperative network matrix lies in using the deep residual model for operation.

[0055] First, let's delve deeper into the understanding of the first satellite-to-ground network embedding integration vector. This vector, obtained through a series of complex operations, carries fused information on numerous state attributes from both the satellite and terrestrial networks.

[0056] Let's illustrate the composition of the first satellite-to-ground network embedding integration vector with a practical example. Consider a specific satellite-to-ground cooperative network scenario where one of the first satellite-to-ground network embedding integration vectors contains rich dimensional information. In the dimension representing the coverage relationship between the satellite orbit and the ground base station, the value is 0.75. This means there is a specific coverage ratio between the satellite orbit and the ground base station. The value of 0.75 indicates that the satellite orbit has a high degree of coverage for the base station, approximately covering 75% of the area where the base station is located. This value is of significant reference value for subsequently assessing the data transmission potential and signal coverage quality between the satellite and the ground base station.

[0057] The value is 0.3, reflecting the relationship between satellite energy reserves and ground base station power consumption. This value indicates that the satellite's energy reserves are in a relatively low equilibrium state relative to the ground base station's power consumption. For example, if the satellite's initial energy reserves can support its operation for 5 years, and it has already consumed a significant amount of energy, leaving relatively limited energy, while the ground base station's power consumption is at a normal level, then the value of 0.3 reflects the need for careful consideration of energy utilization, and it may be necessary to consider optimizing base station power consumption or adjusting the satellite's energy management strategy.

[0058] In the dimension relating satellite sensor status to the number of user connections at terrestrial base stations, the value is -0.4. This negative number indicates a decreasing trend in the number of user connections at terrestrial base stations based on environmental information detected by satellite sensors or other relevant factors. For example, if satellite sensors detect potential severe weather in a certain area, this weather condition may affect signal propagation at terrestrial base stations, leading to a decrease in user connections. The value of -0.4 is a quantification of this correlation.

[0059] The value for the dimension reflecting the relationship between satellite communication frequency band occupancy and ground base station link status is 0.65. This indicates that satellite communication frequency band occupancy has a relatively significant impact on the link status of ground base stations. For example, a high satellite communication frequency band occupancy rate may affect the link bandwidth, latency, and packet loss rate of ground base stations. The value of 0.65 indicates that this impact is at a moderate to high level, requiring attention to the link status and possible corresponding adjustment measures.

[0060] These are just some examples of the dimensions in the first satellite-to-ground network embedding integration vector. In reality, it may contain more dimensions to comprehensively reflect the various state attribute relationships of the satellite-to-ground cooperative network. Under different satellite-to-ground cooperative network conditions, the values ​​of these dimensions will vary depending on the actual network state attribute fusion results, and the value of each dimension plays a unique role in the state evaluation and subsequent operations of the entire satellite-to-ground cooperative network.

[0061] The following section details how to embed these first-level satellite-to-ground networks into an integrated vector input deep residual model to generate a second-level satellite-to-ground network embedded integrated vector. The deep residual model is a neural network model with a special structure that offers many advantages when handling complex input data.

[0062] The deep residual model contains multiple hidden layers, the structure and parameter settings of which are carefully designed based on the characteristics and processing requirements of the satellite-ground cooperative network data. When the first satellite-ground network embedding ensemble vector is input into the deep residual model, it first enters the input layer. For example, there are three first satellite-ground network embedding ensemble vectors, each with multiple dimensional features as mentioned above. These vectors are received in the input layer and prepared to be passed to the first hidden layer.

[0063] In the first hidden layer, the model performs feature extraction and transformation operations on the input vector. Neurons in the hidden layer process the data by performing operations such as weighted summation on each feature dimension of the input. For example, a neuron in the first hidden layer might assign a weight of 0.15 to the dimension representing the relationship between satellite orbit and ground base station coverage in the first satellite-to-ground network embedding ensemble vector. This means that when processing information in this dimension, the neuron will multiply the value of this dimension by 0.15 as part of the calculation. A weight of -0.08 is assigned to the dimension reflecting the relationship between satellite energy reserves and ground base station power consumption; a weight of 0.25 is assigned to the dimension relating satellite sensor status to the number of ground base station user connections; and a weight of 0.3 is assigned to the dimension reflecting the relationship between satellite communication frequency band occupancy and ground base station link status. Then, a specific activation function (e.g., the ReLU function) is used to perform a non-linear transformation on the weighted summation result to obtain the neuron's output. The ReLU function sets values ​​less than 0 to 0 and leaves values ​​greater than 0 unchanged, thus increasing the model's non-linear expressive power.

[0064] Taking the previously mentioned values ​​of the first satellite-to-ground network embedded integration vector as an example, for the dimension value 0.75 representing the relationship between satellite orbit and ground base station coverage, after processing by this neuron, the calculated result is 0.75 * 0.15 = 0.1125. After further processing by the ReLU function, if the result is greater than 0, it remains 0.1125. For the dimension value 0.3 reflecting the relationship between satellite energy reserves and ground base station power consumption, the calculated result is 0.3 * (-0.08) = -0.024, which becomes 0 after processing by the ReLU function. For the dimension value -0.4 involving the correlation between satellite sensor status and the number of ground base station user connections, the calculated result is -0.4 * 0.25 = -0.1, which becomes 0 after processing by the ReLU function. For the dimension value 0.65 reflecting the relationship between satellite communication frequency band occupancy and ground base station link status, the calculated result is 0.65 * 0.3 = 0.195, which becomes 0.195 after processing by the ReLU function. In this way, each neuron in the first hidden layer performs a similar operation on the input first star-ground network embedding ensemble vector, thereby obtaining a set of transformed feature representations.

[0065] As data is passed through the deep residual model, subsequent hidden layers further process the features output by the preceding hidden layers. In this process, residual connections in the deep residual model play a crucial role. Residual connections allow information to be directly passed from earlier layers to later layers, avoiding the vanishing or exploding gradient problems found in deep networks.

[0066] For example, in the second hidden layer, some neurons, in addition to receiving the output of the first hidden layer as input, also directly receive certain features from the original first satellite-to-ground network embedding ensemble vector. These two pieces of information are then combined. For instance, when processing features representing the relationship between satellite orbits and ground base station coverage, a neuron in the second hidden layer receives not only the feature value 0.1125 processed by the first hidden layer, but also the feature value 0.75 directly from the original first satellite-to-ground network embedding ensemble vector. These two values ​​are then combined according to certain rules (e.g., weighted averaging or other combination methods) to obtain a new value as the neuron's output. This residual connection method allows the model to better preserve key features from the original input information, preventing excessive information loss or distortion during multi-layer network transmission.

[0067] After processing through multiple hidden layers, the second satellite-ground network embedding ensemble vector corresponding to the satellite-ground cooperative network matrix is ​​finally obtained in the output layer of the deep residual model. This second satellite-ground network embedding ensemble vector is also a vector with multiple dimensional features, and these features are the result of deep mining and transformation of the first satellite-ground network embedding ensemble vector by the deep residual model.

[0068] For example, the output of the second satellite-to-ground network embedding integration vector has a value of 0.82 in the dimension representing the overall stability of the satellite-to-ground cooperative network. This value may have changed compared to the relevant dimension value in the first satellite-to-ground network embedding integration vector (if there is a similar representation). 0.82 indicates that the network is in a relatively stable state. This may be because the deep residual model takes into account more factors during processing, such as the cooperative relationship of various state attributes between satellites and ground base stations, potential risks in the network, etc., thus obtaining a more accurate assessment of the overall network stability.

[0069] The value is 0.55 in the dimension reflecting the efficiency of satellite-to-ground data transmission. This value indicates that the data transmission efficiency is at a moderate level, which may be the result of a comprehensive analysis of various factors in the deep residual model, such as satellite communication frequency band occupancy, ground base station link status, and coverage relationship between satellites and base stations. For example, although the satellite orbit provides high coverage for the base station, the satellite communication frequency band occupancy has certain limitations on data transmission, resulting in an overall moderate transmission efficiency.

[0070] The value of 0.72 in the dimension reflecting the degree of satellite-ground resource synergy optimization indicates a high degree of resource synergy optimization. This means that the deep residual model, through analysis of satellite energy reserves, ground base station power consumption, and other related resource utilization dimensions in the first satellite-ground network embedded integration vector, judges that the overall resource synergy optimization between satellite and ground has reached a good level. This may be because effective resource allocation strategies have been adopted in some aspects, such as adjusting ground base station power consumption according to satellite energy conditions, or optimizing satellite communication resource allocation according to the user connection requirements of ground base stations.

[0071] These values ​​are calculated by the deep residual model based on the input first satellite-to-ground network embedding ensemble vector, as well as its own network structure and parameters. They comprehensively reflect the overall state of the satellite-to-ground cooperative network matrix after processing by the deep residual model. This technique of converting the first satellite-to-ground network embedding ensemble vector into a second satellite-to-ground network embedding ensemble vector using a deep residual model makes the state representation of the satellite-to-ground cooperative network more accurate and comprehensive. The deep residual model can uncover deeper feature relationships in the first satellite-to-ground network embedding ensemble vector, and the residual connections ensure the effectiveness and stability of the model. This provides a more reliable basis for subsequent data construction and situational judgment of the satellite-to-ground cooperative network matrix based on the second satellite-to-ground network embedding ensemble vector.

[0072] In the actual operation of satellite-ground cooperative networks, such accurate state representation and analysis are crucial. For example, when data transmission optimization is required, strategies such as satellite communication frequency band allocation and ground base station link optimization can be adjusted based on the data transmission efficiency dimension value in the second satellite-ground network embedding integration vector. To improve the overall stability of the network, adjustments can be made to satellite orbits and ground base station layout optimization based on the relevant values ​​in the second satellite-ground network embedding integration vector.

[0073] During the data construction process of the satellite-ground collaborative network, the second satellite-ground network embedding integration vector is obtained by performing a second vector integration on the first satellite-ground network embedding integration vector through a deep residual model. This can more accurately reflect the status of the satellite-ground collaborative network and provide strong support for the management, optimization and stable operation of the entire satellite-ground collaborative network.

[0074] Therefore, firstly, by using a deep residual model to process the first satellite-to-ground network embedding ensemble vector to obtain the second satellite-to-ground network embedding ensemble vector, the structural characteristics of the deep residual model enable it to effectively mine deep-level feature relationships in the data. The multi-dimensional features in the first satellite-to-ground network embedding ensemble vector, after processing, are represented in the second satellite-to-ground network embedding ensemble vector with new values ​​reflecting different aspects of the satellite-to-ground collaborative network's state, such as stability, transmission efficiency, and the degree of resource collaborative optimization. This helps to more accurately grasp the overall state of the satellite-to-ground collaborative network and provides a more reliable basis for situational awareness. Secondly, the residual connection mechanism in the deep residual model avoids the gradient vanishing or gradient exploding problems common in deep networks, ensuring the model's effectiveness and stability. This improves the reliability and accuracy of the entire satellite-to-ground collaborative network data construction process, helps optimize the management and operation of the satellite-to-ground collaborative network, and enables it to better adapt to complex and ever-changing network environments.

[0075] In another exemplary technical solution, before performing a second vector integration on the first satellite-to-ground network embedding integration vectors corresponding to the plurality of satellite network embedding coding vectors to obtain the second satellite-to-ground network embedding integration vector corresponding to the satellite-to-ground cooperative network matrix, the method further includes: obtaining at least one set of first satellite-side state data corresponding to a target multi-task information and observation scheduling server from the plurality of sets of satellite-side state data, wherein the target multi-task information and observation scheduling server refers to a multi-task information and observation scheduling server that has a spatiotemporal relationship with the satellite-to-ground cooperative network matrix; determining a first satellite-to-ground cooperative confidence coefficient corresponding to the at least one set of first satellite-side state data; and determining a second satellite-to-ground cooperative confidence coefficient corresponding to the second satellite-side state data, wherein the second satellite-side state data refers to satellite-side state data other than the at least one set of first satellite-side state data from the plurality of sets of satellite-side state data, and the first satellite-to-ground cooperative confidence coefficient is different from the second satellite-to-ground cooperative confidence coefficient.

[0076] Based on this, the step of performing a second vector integration on the first satellite-to-ground network embedding integration vectors corresponding to the plurality of satellite network embedding coding vectors respectively to obtain the second satellite-to-ground network embedding integration vector corresponding to the satellite-to-ground cooperative network matrix includes: performing a first vector integration on the satellite network embedding coding vectors corresponding to the at least one set of first satellite-side state data and the ground network embedding coding vectors according to the first satellite-to-ground cooperative confidence coefficient to obtain the first satellite-to-ground network embedding integration vector corresponding to the at least one set of first satellite-side state data; and performing a first vector integration on the satellite network embedding coding vectors corresponding to the second satellite-side state data and the ground network embedding coding vectors according to the second satellite-to-ground cooperative confidence coefficient to obtain the first satellite-to-ground network embedding integration vector corresponding to the second satellite-side state data.

[0077] In the following steps, determining the first satellite-ground collaborative confidence coefficient corresponding to the at least one set of first satellite-side state data includes: determining the number of sets of the at least one set of first satellite-side state data; determining the frequent item feature and the edge item feature of the probability heatmap based on the number of sets; configuring the first satellite-ground collaborative confidence coefficient of the at least one set of first satellite-side state data according to the time priority interval of the at least one set of first satellite-side state data using probability heatmap statistical rules, wherein the first satellite-side state data located within the time priority interval is configured with the frequent item feature of the probability heatmap, and the first satellite-side state data located at the boundary of the time priority interval is configured with the edge item feature of the probability heatmap.

[0078] In this exemplary technical solution, the overall technical solution revolves around a series of operations before the second vector integration and the specific methods of the second vector integration. These operations and methods are crucial for the data construction of the satellite-ground cooperative network matrix.

[0079] First, let's describe the operation of acquiring at least one set of first-satellite-side state data corresponding to the target multi-task information management server. In the complex architecture of the satellite-ground cooperative network, the multi-task information management server plays a crucial role, responsible for coordinating and managing various tasks and resources within the network. The target multi-task information management server refers to the multi-task information management server that has a spatiotemporal relationship with the satellite-ground cooperative network matrix. This spatiotemporal relationship is a complex concept involving multiple factors.

[0080] From a spatial perspective, in a space-ground collaborative network, the vast areas covered by the satellite network and the distribution areas of the terrestrial network are intertwined. Different multi-task information scheduling servers may be responsible for different spatial regions. For example, a certain multi-task information scheduling server may primarily be responsible for an area within a specific longitude and latitude range, which may include the coverage areas of multiple terrestrial base stations and satellites. When there is close interaction between the satellite network and the terrestrial network in this area in terms of data exchange and task scheduling, this multi-task information scheduling server has a spatial relationship with the space-ground collaborative network matrix.

[0081] From a temporal perspective, task execution exhibits both temporal sequence and periodicity. Within the same time period, different tasks may require different resource scheduling and coordination. For example, during a specific time period, a large number of data transmission tasks may occur from satellites to ground base stations. In this case, the multi-task information management scheduling server responsible for scheduling these tasks has a temporal relationship with the satellite-ground cooperative network matrix. When a multi-task information management scheduling server has this close spatial and temporal relationship with the satellite-ground cooperative network matrix, it is identified as the target multi-task information management scheduling server.

[0082] At least one set of first satellite-side status data is obtained from multiple sets of satellite-side status data corresponding to the target multi-mission information monitoring and scheduling server. This set of data contains rich satellite status information associated with the target multi-mission information monitoring and scheduling server. This satellite status information can cover multiple aspects, such as the satellite's orbital parameters, the operational status of various devices on the satellite, and the satellite's communication link status. Taking the satellite's orbital parameters as an example, it may include information such as the satellite's orbital altitude, orbital inclination, and orbital period. For example, a satellite with an orbital altitude of 800 kilometers, an orbital inclination of 50 degrees, and an orbital period of 100 minutes are all part of the satellite-side status data and are closely related to the scheduling tasks of the target multi-mission information monitoring and scheduling server. For example, orbital altitude affects the communication distance and signal propagation delay between the satellite and the ground base station; orbital inclination may affect the satellite's coverage range and coverage time for different areas on the ground; and the orbital period is related to the satellite's revisit cycle and data acquisition frequency.

[0083] Then, the first satellite-to-ground cooperation confidence coefficient corresponding to at least one set of first satellite-side state data and the second satellite-to-ground cooperation confidence coefficient corresponding to the second satellite-side state data are determined (the second satellite-side state data refers to the satellite-side state data other than at least one set of first satellite-side state data among multiple sets of satellite-side state data, and the first satellite-to-ground cooperation confidence coefficient and the second satellite-to-ground cooperation confidence coefficient are different).

[0084] Determining the first satellite-to-ground coordination confidence coefficient corresponding to at least one set of first satellite-side state data involves several steps. First, the number of sets of at least one set of first satellite-side state data is determined. This operation is based on the classification and statistics of the satellite-side state data. For example, after careful analysis and statistics, the number of sets of at least one set of first satellite-side state data is determined to be 5. These 5 sets of data are processed differently in subsequent operations, and determining the number of sets provides the basic information for subsequently determining the confidence coefficient.

[0085] Next, the frequent and marginal features of the probability heatmap are determined based on the number of groups. In this process, the probability heatmap is a tool used to describe data distribution and relationships. According to pre-defined rules, when the number of groups is 5, the frequent feature of the probability heatmap may be determined as a value indicating high correlation or importance, such as 0.75. This value indicates that in the satellite-ground collaborative network, this portion of the first satellite-side state data associated with the target multi-mission information and monitoring server has a high correlation or importance in certain key factors. For example, this data may be closely related to the current primary mission (such as high-priority data transmission missions) or play an important role in maintaining the key functions of the satellite-ground collaborative network (such as communication coverage in a specific area).

[0086] The marginal features of the probability heatmap might be a value indicating relatively low correlation or importance, such as 0.15. This means that while these data play a role in the operation and data construction of the entire space-ground collaborative network, their importance or correlation is relatively low. For example, this data might be auxiliary monitoring data, or data that only plays a role under special circumstances.

[0087] Then, based on the time-priority interval of at least one set of first satellite-side state data, the first satellite-ground cooperation confidence coefficient is configured for at least one set of first satellite-side state data according to probabilistic heatmap statistical rules. The time-priority interval is a time range set according to the mission requirements and data characteristics of the satellite-ground cooperation network. For example, the time-priority interval is set to [0, 20] (where the value represents a certain time unit, such as minutes). The first satellite-side state data within this time-priority interval is configured with the frequent term feature of the probabilistic heatmap, i.e., 0.75.

[0088] For example, for a set of satellite-side status data with a timestamp of 10 minutes, this data is configured with a first satellite-to-ground collaboration confidence coefficient of 0.75. This is because satellite-side status data within this time frame may be closely related to currently ongoing critical missions or important network conditions. Satellite-side status data located at the boundaries of time-priority intervals are configured with a probability heatmap edge term feature, i.e., 0.15. For example, satellite-side status data with timestamps of 0 or 20 minutes are configured with a first satellite-to-ground collaboration confidence coefficient of 0.15. This method of configuring confidence coefficients based on time-priority intervals better considers the dynamic characteristics of satellite-side status data over time and its relevance to satellite-to-ground collaborative network tasks.

[0089] Based on the confidence coefficients determined above, a first vector integration operation is performed. According to the first satellite-ground coordination confidence coefficient, the satellite network embedding coding vectors and ground network embedding coding vectors corresponding to at least one set of first satellite-side state data are integrated into a first satellite-ground network embedding integration vector corresponding to at least one set of first satellite-side state data.

[0090] To illustrate, consider a dimension in the satellite network embedding vector representing the satellite's energy reserves, with a value of 80% (meaning the satellite has 80% of its energy remaining). A dimension in the terrestrial network embedding vector represents the available bandwidth of the terrestrial base station, with a value of 50 Mbps. When the first satellite-terrestrial cooperation confidence coefficient is 0.75, the satellite energy reserve dimension will have a relatively high weight in the integrated vector during the first vector integration. Specifically, it might participate in the construction of the first satellite-terrestrial network embedding integrated vector using a calculation method of 0.75 * 80 = 60. Conversely, the available bandwidth of the terrestrial base station will have a relatively low weight, participating in the construction using a calculation method of 0.25 * 50 = 12.5 (for example, the sum of the overall weights is 1; this is just an example of weight allocation). In this way, at least one set of first satellite-side state data corresponding to the first satellite-terrestrial network embedding integrated vector is obtained. This vector integrates information from both the satellite and terrestrial networks, and the weights of the different dimensions are adjusted according to the confidence coefficient.

[0091] Similarly, based on the second satellite-to-ground cooperation confidence coefficient, the satellite network embedding coding vector and the ground network embedding coding vector corresponding to the second satellite-side state data are integrated into a first vector to obtain the first satellite-to-ground network embedding integration vector corresponding to the second satellite-side state data. For example, if the second satellite-to-ground cooperation confidence coefficient is 0.3, for the satellite energy reserve dimension (still 80%) in the satellite network embedding coding vector and the available bandwidth dimension of the ground base station in the ground network embedding coding vector (still 50Mbps), during the first vector integration, the satellite energy reserve dimension participates in the construction of the first satellite-to-ground network embedding integration vector according to a certain calculation method of 0.3*80=24, and the available bandwidth dimension of the ground base station participates in the construction according to a certain calculation method of 0.7*50=35 (again, for example, the sum of the overall weights is 1).

[0092] After performing the first vector integration for different satellite-side state data, different first satellite-to-ground network embedding integration vectors are obtained. These vectors will then be used for the second vector integration to obtain the second satellite-to-ground network embedding integration vector corresponding to the satellite-to-ground cooperative network matrix. In this way, satellite-side state data under different confidence coefficients are treated differently in the first vector integration, thus affecting the generation of the final second satellite-to-ground network embedding integration vector. This operation based on different confidence coefficients can more accurately reflect the relationship between different satellite state data and the ground network in the satellite-to-ground cooperative network, thereby improving the accuracy and effectiveness of the entire satellite-to-ground cooperative network matrix data construction.

[0093] In the actual operation of the space-ground collaborative network, the impact of different satellite status data on the entire network varies due to factors such as their different relationships with the target multi-mission information and perception scheduling server and differences in timing. For example, during an urgent data transmission mission, satellite status data closely related to the target multi-mission information and perception scheduling server and within the time priority interval (such as good satellite communication link status and sufficient power) is crucial to ensuring the smooth progress of the mission. This data is given greater weight in vector integration with a higher confidence coefficient, thus better reflecting its importance when constructing the first space-ground network embedding integration vector and the final second space-ground network embedding integration vector. On the other hand, some relatively minor satellite status data (such as the status data of certain auxiliary monitoring equipment), although also participating in the entire data construction process, have a relatively small weight in vector integration due to their lower confidence coefficient, and will not cause excessive interference to critical missions and the main functions of the network.

[0094] Thus, by determining different confidence coefficients for satellite-ground cooperation before the first vector integration, and processing the satellite-side state data differently, followed by the first and subsequent second vector integrations, the resulting second satellite-ground network embedding integration vector better reflects the complex relationships within the satellite-ground cooperative network. This facilitates more rational use of satellite state data during the satellite-ground cooperative network data construction process, improving the quality of data construction and providing a more reliable basis for subsequent operations based on the second satellite-ground network embedding integration vector (such as data construction situation assessment). Simultaneously, this approach also helps optimize resource allocation within the satellite-ground cooperative network, improving its overall performance and stability, enabling it to better adapt to complex and ever-changing network environments and mission requirements.

[0095] This approach offers several advantages. First, by determining the confidence coefficients of satellite-side state data related to the target multi-task information scheduling server, the importance of different satellite state data to the space-ground collaborative network can be differentiated. For example, different confidence coefficients can be determined based on time priority intervals and the number of groups, making data processing more targeted and better adaptable to the complex task requirements and dynamic changes of the space-ground collaborative network. Second, performing the first vector integration based on different confidence coefficients allows for a reasonable adjustment of the weights of satellite network and ground network information in the integrated vector, more accurately reflecting their relationship and improving the accuracy of the first space-ground network embedding integration vector. This helps to obtain a second space-ground network embedding integration vector that better reflects the actual situation of the space-ground collaborative network in the subsequent second vector integration. Finally, this precise method of constructing the second space-ground network embedding integration vector provides a more reliable basis for building the space-ground collaborative network matrix data, helping to optimize network management and operation, and improve the overall network performance, stability, and adaptability to complex tasks and environmental changes.

[0096] In some preferred embodiments, the satellite-ground cooperative network matrix includes multiple network matrix blocks, wherein the x-th network matrix block corresponds to the x-th group of satellite-side state data, and x is a positive integer; before mining the satellite network embedding encoding vectors corresponding to the multiple groups of satellite-side state data, the method further includes: determining the y-th group of initial satellite-side state data in the x-th network matrix block as the x-th group of satellite-side state data, where y is a positive integer.

[0097] In practical applications, the structural characteristics of the space-ground cooperative network matrix become a crucial foundation for the entire technical solution. The space-ground cooperative network matrix comprises multiple network matrix blocks; this block structure is designed for more precise processing and analysis of data and state relationships within the space-ground cooperative network.

[0098] Each network matrix block corresponds to a set of satellite-side status data. This correspondence is not arbitrary but based on the operational mechanism and data association logic of the satellite-ground collaborative network. For example, the first network matrix block corresponds to the first set of satellite-side status data, the second network matrix block corresponds to the second set of satellite-side status data, and so on, where x is a positive integer. This correspondence runs through the entire data processing flow of the technical solution.

[0099] Before mining the satellite network embedding vectors corresponding to multiple sets of satellite-side state data, there is an operation that identifies the y-th initial satellite-side state data in the x-th network matrix block as the x-th set of satellite-side state data, where y is a positive integer. The significance of this operation is to filter out data sets closely related to a specific network matrix block from numerous initial satellite-side state data sets, so as to perform more accurate satellite network embedding vector mining subsequently.

[0100] Taking a specific satellite-ground collaborative network matrix structure as an example, this network matrix is ​​divided into 5 network matrix blocks. The third network matrix block contains multiple sets of initial satellite-side state data. This initial satellite-side state data covers various aspects of the satellite's information. For example, one set of initial satellite-side state data might include the satellite's orbital position information, such as an orbital altitude of 700 kilometers and an orbital inclination of 45 degrees; the satellite's communication status, such as the communication frequency band being the X-band and a signal strength of -80 dBm; and the satellite's energy status, such as an energy reserve of 60%. Within this network matrix block, through specific rules and logic, a particular set of initial satellite-side state data is determined as the third set of satellite-side state data corresponding to this network matrix block.

[0101] This determination process is based on multiple factors. From the perspective of satellite orbital position, satellites at certain orbital positions may have a more direct connection to the ground areas or communication links involved in a specific network matrix segment. For example, a satellite with an orbital altitude of 700 kilometers and an orbital inclination of 45 degrees may have a unique geometric relationship when covering the ground area corresponding to a specific network matrix segment, making its status data more critical to that segment. From a communication status perspective, satellites communicating in the X-band with a signal strength of -80dBm may play an important role in communication tasks related to that network matrix segment, and their communication status data is indispensable for understanding and processing the network characteristics of that segment. Similarly, the status of a satellite with 60% energy reserves may also be closely related to task scheduling and resource allocation within that network matrix segment. For example, some tasks may require satellites to have a certain amount of energy reserves to execute, and 60% energy reserves may just meet some of the task requirements within that network matrix segment.

[0102] Once the x-th set of satellite-side state data is determined, the satellite network embedding coding vector can be mined. This mining process transforms the satellite-side state data into a form that allows for better data processing and analysis within the space-ground collaborative network. For each set of satellite-side state data, the mined satellite network embedding coding vector contains an abstraction and encoding of the satellite-side state data characteristics.

[0103] Taking the previously mentioned third set of satellite-side status data as an example, when mining the satellite network embedding coding vector, the information about a satellite's orbital altitude of 700 kilometers might be encoded as a specific value or coding segment. For instance, in the coding system, an orbital altitude between 500 and 1000 kilometers is encoded as a specific coding segment, which might correspond to a number or a set of binary codes, such as 101. An orbital inclination of 45 degrees might be encoded as another specific coding form, such as 011. For a communication frequency band of X-band, it might be encoded as 110, and a signal strength of -80dBm might be encoded as a code divided according to the signal strength range, such as 001. A satellite energy reserve of 60% would also be encoded as a code corresponding to the energy reserve percentage, such as 100. These codes combined constitute the satellite network embedding coding vector corresponding to the third set of satellite-side status data.

[0104] By transforming the satellite-side state data corresponding to each network matrix block into satellite network embedded coding vectors in this way, subsequent data processing operations can be performed more effectively. For example, when fusing or integrating with terrestrial network embedded coding vectors, these coding vectors can participate in the operation in a unified form that is easy to calculate and analyze. Moreover, this coding method can better reflect the inherent relationship between satellite-side state data and the various blocks of the satellite-ground cooperative network matrix, providing a more accurate and effective data foundation for the construction of the entire satellite-ground cooperative network.

[0105] This technical approach, which associates network matrix blocks with satellite-side state data, determines the satellite-side state data through a specific method, and then mines the corresponding satellite network embedded coding vectors, helps improve the accuracy and effectiveness of data processing in space-ground collaborative networks. In the actual operation of space-ground collaborative networks, this precise data processing can better adapt to complex network environments and diverse mission requirements. For example, when allocating resources, accurate satellite network embedded coding vectors allow for more rational allocation of satellite and ground network resources; in terms of network optimization, it enables more targeted optimization and adjustment of satellite and ground networks associated with specific network matrix blocks, thereby improving the overall performance and stability of the space-ground collaborative network.

[0106] In an alternative technical approach, the step of performing data construction situation discrimination on the satellite-ground cooperative network matrix based on the second satellite-ground network embedding integration vector to obtain the construction situation decision information corresponding to the satellite-ground cooperative network matrix includes: inputting the second satellite-ground network embedding integration vector into a pre-tuned data construction situation discrimination network to generate the construction situation decision information corresponding to the satellite-ground cooperative network matrix.

[0107] Furthermore, before the step of embedding the second satellite-ground network into the integrated vector input data to build a situational awareness network and generating the construction situational decision information corresponding to the satellite-ground cooperative network matrix, the method further includes: obtaining ground network equipment monitoring data examples corresponding to the satellite-ground cooperative network matrix example and multiple sets of satellite-side status data examples, wherein the ground network equipment monitoring data examples are the ground network equipment monitoring data involved in the data construction task process of the satellite-ground cooperative network matrix example, and the satellite-ground cooperative network matrix example carries construction situational decision annotations; mining the ground network embedding encoding vector corresponding to the ground network equipment monitoring data examples, and mining the satellite network embedding encoding vectors corresponding to the multiple sets of satellite-side status data examples respectively; and combining multiple satellite... The satellite network embedding encoding vector and the ground network embedding encoding vector are respectively integrated using the first vector to obtain the first satellite-ground network embedding integration vector corresponding to the multiple sets of satellite-side state data examples. The first satellite-ground network embedding integration vector corresponding to the multiple sets of satellite-side state data examples is then integrated using the second vector to obtain the second satellite-ground network embedding integration vector corresponding to the satellite-ground cooperative network matrix example. The second satellite-ground network embedding integration vector is input into the initial situation discrimination network to generate the construction situation prediction information corresponding to the satellite-ground cooperative network matrix example. The initial situation discrimination network is then debugged based on the difference between the construction situation prediction information and the construction situation decision annotation to obtain the data construction situation discrimination network.

[0108] In this alternative technical approach, the entire technical solution revolves around using the second satellite-ground network embedded integration vector to perform data construction situation judgment on the satellite-ground cooperative network matrix in order to obtain construction situation decision information, which includes several key operational steps and the network debugging process.

[0109] First, the core operation of determining the data construction status of the satellite-ground cooperative network matrix based on the second satellite-ground network embedding integration vector is explained. This involves inputting the second satellite-ground network embedding integration vector into a pre-tuned data construction status determination network to generate the corresponding construction status decision information for the satellite-ground cooperative network matrix. This data construction status determination network is obtained through a series of pre-tuning processes and is specifically designed to accurately determine the data construction status of the satellite-ground cooperative network matrix based on the input second satellite-ground network embedding integration vector and output the corresponding decision information.

[0110] Before embedding the second satellite-to-ground network into the integrated vector input data to build the situational awareness network, a series of operations are required to construct and debug this network. The first step is to acquire ground network equipment monitoring data examples corresponding to the satellite-to-ground cooperative network matrix example, as well as multiple sets of satellite-side status data examples. Here, the satellite-to-ground cooperative network matrix example is a representative example of a satellite-to-ground cooperative network structure, and the ground network equipment monitoring data involved in the data construction process is defined as ground network equipment monitoring data examples. These ground network equipment monitoring data contain multiple aspects of information, such as the operational status data of ground base stations. Taking a ground base station as an example, its operational status data might include a temperature of 30 degrees Celsius, a power consumption of 2 kilowatts, and 500 user connections within its signal coverage area. Simultaneously, there are multiple sets of satellite-side status data examples, which contain various status information about the satellites. For example, a satellite's orbital altitude is 600 kilometers, the sampling frequency of a sensor on the satellite is 10 times per second, and the satellite's energy reserve is 70%. Furthermore, the example of the satellite-ground cooperative network matrix carries a situational decision annotation. This annotation is a known and accurate judgment result of the situational decision of the satellite-ground cooperative network matrix data in this example, which is used for subsequent debugging of the initial situational decision network.

[0111] Next, the process involves mining the terrestrial network embedding code vector corresponding to the terrestrial network equipment monitoring data examples, and mining the satellite network embedding code vectors corresponding to multiple sets of satellite-side status data examples. For the terrestrial base station data in the terrestrial network equipment monitoring data examples, different data types are encoded during the mining of the terrestrial network embedding code vector. For example, a temperature of 30 degrees Celsius might be encoded as a specific value according to a pre-defined temperature range encoding rule, such as 10 for temperatures between 25-35 degrees Celsius; a power consumption of 2 kilowatts might be encoded as 01 according to a power consumption range encoding rule, such as 01 for temperatures between 1-3 kilowatts; and a user connection count of 500 might be encoded as 11 according to a user connection count range encoding rule, such as 11 for connections between 300-600. These codes combine to form the terrestrial network embedding code vector. The satellite data in the satellite-side status data examples are encoded similarly. For example, a satellite orbital altitude of 600 kilometers is encoded as 010 according to orbital altitude range coding rules, such as 500-700 kilometers; a satellite sensor sampling frequency of 10 times per second is encoded as 101 according to sampling frequency range coding rules, such as 5-15 times per second; and a satellite energy reserve of 70% is encoded as 110 according to energy reserve percentage range coding rules, such as 60%-80%. These codes constitute the satellite network embedded coding vector.

[0112] Then, multiple satellite network embedding coding vectors and ground network embedding coding vectors are respectively subjected to first vector integration to obtain first satellite-ground network embedding integration vectors corresponding to multiple sets of satellite-side state data examples. Taking two sets of satellite network embedding coding vectors and ground network embedding coding vectors as examples, during the first vector integration, the elements in each vector are combined and calculated according to certain integration rules. For example, for the elements in the satellite network embedding coding vector of satellite 1 and the elements in the ground network embedding coding vector, summation or other calculation operations may be performed according to a certain weight allocation. For example, the satellite network embedding coding vector of satellite 1 is [010, 101, 110], and the ground network embedding coding vector is [10, 01, 11]. During the first vector integration, the element at each corresponding position is calculated according to a specific weight. For example, the weight of the first element is 0.3, the weight of the second element is 0.3, and the weight of the third element is 0.4. Then, after calculation, the first element in the first satellite-ground network embedding integration vector corresponding to satellite 1 is 0.3*010+0.1*10=001 (the actual calculation can be more complex depending on the specific coding and weight definition). The complete first satellite-ground network embedding integration vector is obtained by analogy.

[0113] Next, the first satellite-to-ground network embedding integration vectors corresponding to the multiple sets of satellite-side state data examples are subjected to second vector integration to obtain the second satellite-to-ground network embedding integration vector corresponding to the satellite-to-ground cooperative network matrix example. Again, taking two sets of first satellite-to-ground network embedding integration vectors as an example, for instance, first satellite-to-ground network embedding integration vector 1 is [001, 110, 011], and first satellite-to-ground network embedding integration vector 2 is [101, 010, 100]. During the second vector integration, different rules than those used in the first vector integration are employed. For example, some elements may be scaled up or down before summing or other complex combination operations are performed. For instance, the first element 001 in first satellite-to-ground network embedding integration vector 1 is multiplied by 2 to obtain 010, and the first element 101 in first satellite-to-ground network embedding integration vector 2 is divided by 2 to obtain 010. Then, these two adjusted elements are added together to obtain 010 + 010 = 100 (this is just an example; the actual operation is more complex). This process is repeated for other elements to finally obtain the second satellite-to-ground network embedding integration vector corresponding to the satellite-to-ground cooperative network matrix example.

[0114] The second satellite-to-ground network is then embedded into the integrated vector input to the initial situational awareness network, generating situational awareness prediction information corresponding to the satellite-to-ground cooperative network matrix example. This initial situational awareness network is a preliminarily constructed network for situational assessment; its structure and parameters may be based on some general design principles or empirical settings. When the second satellite-to-ground network is embedded into the integrated vector, the network outputs situational awareness prediction information based on its internal computational logic. This prediction information is a preliminary judgment result of the situational awareness of the satellite-to-ground cooperative network matrix example.

[0115] Finally, the initial situational awareness network was debugged based on the difference between the situational awareness prediction information and the situational awareness decision annotation, resulting in a data-based situational awareness network. Due to the preliminary nature of the initial situational awareness network, its output situational awareness prediction information may differ from the known accurate situational awareness decision annotation. For example, the situational awareness prediction information might indicate that the satellite-ground cooperative network matrix example is in a stable data-based situational awareness, while the situational awareness decision annotation indicates that it is in a data-based situational awareness that needs optimization. By analyzing this difference, the structural parameters of the initial situational awareness network (such as the connection weights of neurons, the parameters of the activation function, etc.) were adjusted. After multiple such adjustments, using a large number of satellite-ground cooperative network matrix examples and their corresponding ground network equipment monitoring data examples, satellite-side status data examples, and situational awareness decision annotations, the initial situational awareness network was continuously optimized until the difference between its output situational awareness prediction information and the situational awareness decision annotation reached an acceptable range, thus obtaining a pre-tuned data-based situational awareness network. This network can more accurately determine the data-based situational awareness of the satellite-ground cooperative network matrix based on the input second satellite-ground network embedding ensemble vector and generate corresponding situational awareness decision information.

[0116] This design has several advantages. First, by using an example of a satellite-ground cooperative network matrix and its related data to build and debug the situational awareness network, the accuracy of network judgments can be improved. For example, the integrated vector obtained by mining and encoding vectors from the example data can be used to debug the network. Second, adjusting the network based on the differences between the situational awareness prediction information and the decision annotations can continuously optimize network performance. Finally, the pre-debugged network can accurately generate situational awareness decision information for the satellite-ground cooperative network matrix based on the integrated vectors embedded in the second satellite-ground network, which helps to effectively manage the construction of satellite-ground cooperative network data.

[0117] In some independent embodiments, after determining the data construction status of the satellite-ground cooperative network matrix based on the second satellite-ground network embedding integration vector and obtaining the construction status decision information corresponding to the satellite-ground cooperative network matrix, the method further includes: based on the construction status decision information, constructing a data relationship network for the ground-side sensing data and the multiple sets of satellite-side status data to generate a full-network sensing data relationship network corresponding to the satellite-ground cooperative network matrix.

[0118] In detail, based on the established situational decision-making information, a data relationship network is constructed for the ground-side sensing data and the multiple sets of satellite-side status data to generate the full-network sensing data relationship network corresponding to the satellite-ground cooperative network matrix. Specifically, this includes: extracting the ground network sensing data topology and satellite network status data topology of the satellite-ground cooperative network matrix using the established situational decision-making information. The ground network sensing data topology is used to represent the ground-side sensing data corresponding to the satellite-ground cooperative network matrix, and the satellite network status data topology is used to represent the multiple sets of satellite-side status data corresponding to the satellite-ground cooperative network matrix; and for the ground network sensing data topology, based on the data topology associated with the ground network sensing data topology in the satellite network status data topology... A first set of network-wide sensing and linkage nodes is obtained, which represents the ground network sensing data topology that integrates the satellite network status data topology. For the satellite network status data topology, a second set of network-wide sensing and linkage nodes is obtained based on the data topology node set associated with the satellite network status data topology in the ground network sensing data topology. This second set of network-wide sensing and linkage nodes represents the satellite network status data topology that integrates the ground network sensing data topology. A relationship is constructed between the first and second sets of network-wide sensing and linkage nodes to obtain a network-wide sensing and linkage relationship topology. Based on this network-wide sensing and linkage relationship topology, a network-wide sensing data relationship network corresponding to the satellite-ground collaborative network matrix is ​​generated.

[0119] In the above-mentioned independent embodiments, after obtaining the construction status decision information by performing data construction status judgment on the satellite-ground cooperative network matrix based on the second satellite-ground network embedded integration vector, the entire technical solution further performs a series of operations based on the construction status decision information to generate the full-network perception data relationship network corresponding to the satellite-ground cooperative network matrix.

[0120] First, after obtaining the situational decision-making information, the ground-side sensing data topology and satellite-side status data topology of the satellite-ground cooperative network matrix are extracted based on this information. The ground-side sensing data topology is used to characterize the ground-side sensing data corresponding to the satellite-ground cooperative network matrix, while the satellite-side status data topology is used to characterize multiple sets of satellite-side status data corresponding to the satellite-ground cooperative network matrix.

[0121] Taking a specific satellite-ground cooperative network matrix as an example, ground-side sensing data contains multiple aspects of information, reflecting the operational status of ground network equipment and its relationships with other equipment. For instance, consider a group of ground base stations, where the temperature of one base station is 25 degrees Celsius. This temperature data is an important component of the vast amount of ground-side sensing data. In terms of power consumption, this base station consumes 3 kilowatts, reflecting its energy consumption during operation. Furthermore, the number of user connections within the signal coverage area is 800, reflecting the base station's load and service capacity within the area. These data are inherently interconnected, and the construction of the ground network sensing data topology is based on this data and the relationships between them.

[0122] When constructing the ground-based sensing data topology, elements such as temperature, power consumption, and the number of user connections might be used as nodes, with the relationships between them forming edges. For example, temperature can affect the performance of base station equipment, thus affecting power consumption. Therefore, in the topology, there would be an edge between the temperature node and the power consumption node, representing this influence. Similarly, changes in the number of user connections can lead to fluctuations in power consumption, so there would also be an edge between the user connection number node and the power consumption node. This topology clearly demonstrates the relationships between ground-side sensing data, providing a foundation for subsequent operations.

[0123] For satellite network status data topology, satellite-side status data also contains a wealth of diverse information. For example, the orbital altitude of a satellite is 700 kilometers; this orbital altitude data has a significant impact on the satellite's coverage area and communication latency. The sampling frequency of a sensor on the satellite is 15 times per second; this sampling frequency reflects the satellite's ability and frequency of collecting data from the external environment or other sources. The satellite's energy reserve is 65%; this data reflects the satellite's energy status and plays a crucial role in the satellite's continuous operation and functional execution.

[0124] When constructing the satellite network state data topology, the topology is built using satellite orbital altitude, sensor sampling frequency, and energy reserves as nodes, and the relationships between them as edges. For example, satellite orbital altitude can affect sensor sampling frequency because different orbital altitudes correspond to different observation angles and environmental conditions, thus affecting sensor efficiency. Therefore, in the satellite network state data topology, there is an edge between the orbital altitude node and the sampling frequency node, representing this influence. Satellite energy reserves are also related to sensor operation; the amount of energy reserves limits sensor operating time and sampling frequency, so there is also an edge between the energy reserve node and the sampling frequency node.

[0125] Next, based on the set of data topology nodes associated with the ground network sensing data topology in the satellite network status data topology, the first set of network-wide sensing linkage nodes is obtained. This operation aims to integrate relevant information from the satellite network status data topology into the ground network sensing data topology, thereby constructing a more comprehensive set of nodes that reflects the satellite-ground collaborative network relationship.

[0126] For example, a satellite orbital altitude node in the satellite network status data topology may be associated with a ground base station coverage area node in the ground network sensing data topology. When a satellite orbits at an altitude of 700 kilometers, its covered ground area is specific, and this coverage area overlaps with or influences the coverage area of ​​a ground base station. This is because the satellite's coverage area affects the strength and quality of the satellite signals received by the ground base station, thus affecting related data such as the number of user connections. Integrating these associated nodes forms the first set of network-wide sensing and linkage nodes. This node set not only includes the original nodes in the ground network sensing data topology but also incorporates nodes from the satellite network status data topology related to the ground network sensing data topology; it represents a ground network sensing data topology that integrates the satellite network status data topology.

[0127] This process requires careful analysis of the relationship between each node in the satellite network status data topology and the ground network sensing data topology. For example, while the satellite's energy storage node may seem unrelated to the ground base station's temperature node, an indirect connection can be found through the satellite's communication function (satellite energy storage affects communication duration and frequency, thus affecting the stability of the ground base station's reception of satellite signals, which in turn may affect the heat dissipation of base station equipment, indirectly affecting temperature). These indirectly related nodes are then rationally integrated into the first set of network-wide sensing and linkage nodes, enabling this node set to more comprehensively reflect the complex relationships within the satellite-ground collaborative network.

[0128] Similarly, for the satellite network status data topology, a second set of network-wide sensing and linkage nodes is obtained based on the set of data topology nodes associated with the satellite network status data topology in the ground network sensing data topology. For example, the power consumption nodes of ground base stations in the ground network sensing data topology may be associated with the satellite energy reserve nodes in the satellite network status data topology. The power consumption of the ground base station is 3 kilowatts, and this power consumption requirement may affect the satellite's energy allocation strategy. If the power consumption requirement of the ground base station is high, the satellite may need to adjust its communication power or data transmission strategy, thereby affecting the satellite's energy consumption. Therefore, the satellite's energy reserve needs to be reasonably planned and adjusted according to the power consumption of the ground base station.

[0129] Integrating these related nodes yields the second set of network-wide sensing and linkage nodes, which represents the satellite network status data topology that integrates the ground network sensing data topology. When constructing this node set, various indirect relationships also need to be considered. For example, the number of user connections at ground base stations may affect the power consumption of ground base stations, thereby affecting the satellite's energy reserves, and then the sampling frequency of satellite sensors through the satellite's energy reserves. Therefore, the second set of network-wide sensing and linkage nodes needs to include all these related nodes to comprehensively reflect this complex chain of relationships.

[0130] Then, the relationships between the first and second sets of network-wide sensing and linkage nodes are constructed to obtain the network-wide sensing and linkage topology. This process further sorts out and constructs the relationships between two sets of nodes that have already integrated some information, forming a more complete and accurate topology that reflects various data relationships in the space-ground collaborative network.

[0131] For example, there might be an indirect relationship between the ground base station temperature node in the first set of network-wide sensing linkage nodes and the satellite sensor sampling frequency node in the second set of network-wide sensing linkage nodes. When the ground base station temperature is 25 degrees Celsius, the base station equipment performance is in a certain state, which affects the signal transmission quality and efficiency of the base station. The satellite sensor sampling frequency is 15 times per second, and the information collected by the satellite sensor is transmitted to the ground base station via the satellite communication link. If the signal transmission quality and efficiency of the ground base station changes, it may affect the transmission and processing of the information collected by the satellite sensor, potentially requiring the satellite to adjust the sensor's sampling frequency. By analyzing their relationships with other nodes (e.g., ground base station temperature may affect the number of user connections, which in turn is related to the satellite communication workload, and the satellite communication workload affects the satellite sensor's sampling frequency), relationship edges are constructed within the network-wide sensing linkage topology.

[0132] When constructing the topology of the network-wide sensing and linkage relationships, similar relationship analysis needs to be performed on each node in the first and second sets of network-wide sensing and linkage nodes. For example, consider the relationship between the ground base station power consumption node in the first set and the satellite orbital altitude node in the second set. Changes in ground base station power consumption may affect the demand for satellite communication, thereby influencing the satellite's orbital adjustment strategy (although this influence may be long-term and complex, involving the synergistic effects of multiple intermediate links and other factors). Therefore, relationship edges between them also need to be constructed in the network-wide sensing and linkage topology.

[0133] Finally, based on the topology of the network-wide sensing linkage relationship, a network-wide sensing data relationship network corresponding to the space-ground collaborative network matrix is ​​generated. This network-wide sensing data relationship network is a comprehensive representation of the complex relationships between ground-side sensing data and satellite-side status data in the space-ground collaborative network matrix.

[0134] It integrates various nodes (including data nodes from ground base stations and status data nodes from satellites) and their relationships (direct and indirect) in a network format. In this comprehensive sensing data relationship network, each node has a specific meaning and function, and the edges between nodes represent the strength and type of their relationship. For example, the weight of the edge connecting the ground base station temperature node and the satellite sensor sampling frequency node might represent the degree of influence of this indirect relationship; this weight can be obtained through the analysis and calculation of a large amount of real-world data.

[0135] This technical solution, based on constructing a network of perception data relationships across the entire network using situational decision-making information, helps to deeply understand the internal structure and operational mechanism of the space-ground collaborative network matrix. By clarifying the relationship between the ground network perception data topology and the satellite network status data topology, and constructing a network-wide perception linkage topology, it is possible to better grasp the mutual influence and collaborative relationship between the ground and satellite data in the space-ground collaborative network, thereby providing a strong basis for the optimization, management, and fault diagnosis of the space-ground collaborative network.

[0136] For example, in network optimization, the layout of ground base stations or the orbital parameters of satellites can be adjusted based on the relationships between nodes in the network-wide sensing data relationship network. If the relationship between the power consumption nodes of ground base stations and the energy storage nodes of satellites is found to have a significant impact on the energy consumption of the entire network, the energy efficiency of the entire satellite-ground collaborative network can be improved by optimizing the power management strategy of ground base stations or adjusting the energy allocation method of satellites. In terms of fault diagnosis, when network problems occur, the nodes and relationships that may be causing the problems can be quickly located through the network-wide sensing data relationship network. If the number of user connections at ground base stations suddenly drops, by analyzing the network-wide sensing data relationship network, it is possible to trace the problem back to a problem with the satellite communication link, or that the temperature of the ground base station is too high, affecting the equipment performance, thereby enabling targeted troubleshooting and repair.

[0137] This design, firstly, utilizes situational decision-making information to extract the topology of ground-based sensing data and satellite-based status data, enabling accurate and comprehensive characterization of both ground-side sensing data and satellite-side status data, laying a solid foundation for subsequent operations. Secondly, the process of constructing the first and second sets of network-wide sensing linkage nodes deeply integrates information related to the satellite-ground data topology. This goes beyond simple node merging; it fully considers the direct and indirect relationships between satellite and ground, greatly deepening the connection between them and expanding the understanding of the satellite-ground collaborative network beyond its isolated components. Thirdly, constructing the network-wide sensing linkage topology and ultimately generating a network-wide sensing data relationship network comprehensively and meticulously presents the complex relationships between satellite and ground data. This comprehensiveness contributes to a deeper understanding of the internal structure and operational mechanism of the satellite-ground collaborative network. Finally, this technical solution provides strong support for the management, optimization, and fault diagnosis of the satellite-ground collaborative network, playing a significant role in improving network performance, optimizing resource allocation, and rapidly locating and resolving network faults.

[0138] Furthermore, Figure 2 This is a schematic diagram of the structure of a satellite-ground collaborative data construction system 200 provided in an embodiment of this application. Figure 2 The satellite-ground collaborative data construction system 200 shown includes a processor 210, which can call and run computer programs from memory to implement the methods in the embodiments of this application.

[0139] Optionally, such as Figure 2 As shown, the satellite-ground collaborative data construction system 200 may further include a memory 230. The processor 210 can retrieve and run computer programs from the memory 230 to implement the methods described in this embodiment.

[0140] The memory 230 can be a separate device independent of the processor 210, or it can be integrated into the processor 210.

[0141] Optionally, such as Figure 2 As shown, the satellite-ground collaborative data construction system 200 may also include a transceiver 220. The processor 210 can control the transceiver 220 to interact with other devices. Specifically, it can send information or data to other devices or receive information or data sent by other devices.

[0142] Optionally, the space-ground collaborative data construction system 200 can implement the corresponding processes of the storage engine or components (such as processing modules) in the storage engine or devices with deployed storage engines in the various methods of the embodiments of this application. For the sake of brevity, these will not be elaborated here.

[0143] It should be understood that the processor in this application embodiment may be an integrated circuit chip with signal processing capabilities.

[0144] It is understood that the memory in the embodiments of this application may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. It should be noted that the memory in the systems and methods described herein is intended to include, but is not limited to, suitable types of memory.

[0145] Based on the above, a readable storage medium is provided, on which a program or instructions are stored, and when the program or instructions are executed by a processor, the steps of the above method are implemented.

[0146] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0147] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solutions of the embodiments of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0148] The embodiments of this application have been described above with reference to the accompanying drawings. However, the embodiments of this application are not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the embodiments of this application without departing from the spirit and scope of protection of the embodiments of this application, and all of these forms are within the protection scope of the embodiments of this application.

Claims

1. A data building method based on satellite-ground cooperative computing, characterized in that, The method is applied to a star-ground cooperative data building system, and the method comprises: obtaining ground side perception data corresponding to a star-ground cooperative network matrix and a plurality of satellite side state data, wherein the ground side perception data is ground network equipment monitoring data associated with the star-ground cooperative network matrix; mining ground network embedding encoding vectors corresponding to the ground side perception data, and mining satellite network embedding encoding vectors corresponding to the plurality of satellite side state data respectively; performing first vector integration on a plurality of satellite network embedding encoding vectors and the ground network embedding encoding vector respectively to obtain first star-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding encoding vectors respectively; performing second vector integration on the first star-ground network embedding integrated vectors corresponding to the plurality of satellite side state data examples respectively to obtain second star-ground network embedding integrated vectors corresponding to the star-ground cooperative network matrix example; inputting the second star-ground network embedding integrated vectors into an initial situation discrimination network to generate building situation prediction information corresponding to the star-ground cooperative network matrix example; debugging the initial situation discrimination network based on the difference between the building situation prediction information and the building situation decision annotation to obtain the data building situation discrimination network. the first vector integration on a plurality of satellite network embedding encoding vectors and the ground network embedding encoding vector respectively to obtain first star-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding encoding vectors respectively, comprising: ​ ​ ​ ​ ​ 2. The method of claim 1, wherein, ​ The plurality of satellite network embedding code vectors are respectively fused with the ground network embedding code vectors to obtain first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding code vectors.

3. The method of claim 2, wherein, Before the plurality of satellite network embedding code vectors are respectively fused with the ground network embedding code vectors to obtain first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding code vectors, the method further includes: The plurality of satellite network embedding code vectors are respectively subjected to vector pooling processing to obtain satellite network embedding pooled vectors corresponding to the plurality of satellite network embedding code vectors, and the vector pooling processing is used to compress feature sizes corresponding to the satellite network embedding code vectors; The plurality of satellite network embedding code vectors are respectively fused with the ground network embedding code vectors to obtain first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding code vectors. The plurality of satellite network embedding code vectors are respectively fused with the ground network embedding code vectors to obtain first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding code vectors.

4. The method according to any one of claims 1 to 3, characterized in that, The plurality of satellite network embedding code vectors are respectively fused with the ground network embedding code vectors to obtain first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding code vectors. The plurality of satellite network embedding code vectors are respectively fused with the ground network embedding code vectors to obtain first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding code vectors.

5. The method according to any one of claims 1 to 3, wherein The plurality of satellite network embedding code vectors are respectively fused with the ground network embedding code vectors to obtain first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding code vectors. The plurality of satellite network embedding code vectors are respectively fused with the ground network embedding code vectors to obtain first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding code vectors. The plurality of satellite network embedding code vectors are respectively fused with the ground network embedding code vectors to obtain first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding code vectors. The plurality of satellite network embedding code vectors are respectively fused with the ground network embedding code vectors to obtain first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding code vectors. The plurality of satellite network embedding code vectors are respectively fused with the ground network embedding code vectors to obtain first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding code vectors. The plurality of satellite network embedding code vectors are respectively fused with the ground network embedding code vectors to obtain first satellite-ground network embedding integrated vectors corresponding to the plurality of satellite network embedding code vectors. performing the first vector integration on the satellite network embedding coding vector corresponding to the at least one group of first satellite side state data and the ground network embedding coding vector according to the first satellite-ground collaborative confidence coefficient, to obtain a first satellite-ground network embedding integrated vector corresponding to the at least one group of first satellite side state data; performing the first vector integration on the satellite network embedding coding vector corresponding to the second satellite side state data and the ground network embedding coding vector according to the second satellite-ground collaborative confidence coefficient, to obtain a first satellite-ground network embedding integrated vector corresponding to the second satellite side state data.

6. The method of claim 5, wherein, The method further comprises: determining a number of groups of the at least one group of first satellite side state data; determining a probability heat map frequent item feature and a probability heat map edge item feature according to the number of groups; configuring the first satellite-ground collaborative confidence coefficient for the at least one group of first satellite side state data according to a time sequence priority interval and a probability heat statistical rule, wherein the first satellite side state data located in the time sequence priority interval is configured with the probability heat map frequent item feature, and the first satellite side state data located at the boundary of the time sequence priority interval is configured with the probability heat map edge item feature.

7. The method of any one of claims 1 to 3, wherein, The satellite-ground collaborative network matrix comprises a plurality of network matrix blocks, wherein an xth network matrix block corresponds to an xth group of satellite side state data, and x is a positive integer; Before the mining of the satellite network embedding coding vectors corresponding to the plurality of groups of satellite side state data, the method further comprises:

8. The method of any one of claims 1 to 3, wherein, determining yth initial satellite side state data in the xth network matrix block as the xth group of satellite side state data, and y is a positive integer. The data construction situation discrimination on the satellite-ground collaborative network matrix according to the second satellite-ground network embedding integrated vector comprises: 9.A satellite-terrestrial collaborative data building system, characterized in that, inputting the second satellite-ground network embedding integrated vector into a pre-debugged data construction situation discrimination network to generate the construction situation decision information corresponding to the satellite-ground collaborative network matrix. The system comprises at least one processor and a memory; the memory stores computer execution instructions; the at least one processor executes the computer execution instructions stored in the memory, so that the at least one processor executes the method of any one of claims 1-8.

Images