CN117828292A - Satellite link saturation evolution analysis method based on space-time diagram convolution - Google Patents
Satellite link saturation evolution analysis method based on space-time diagram convolution Download PDFInfo
- Publication number
- CN117828292A CN117828292A CN202311765807.XA CN202311765807A CN117828292A CN 117828292 A CN117828292 A CN 117828292A CN 202311765807 A CN202311765807 A CN 202311765807A CN 117828292 A CN117828292 A CN 117828292A
- Authority
- CN
- China
- Prior art keywords
- satellite
- space
- time
- saturation
- link
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010586 diagram Methods 0.000 title claims abstract description 79
- 238000004458 analytical method Methods 0.000 title claims abstract description 43
- 238000004891 communication Methods 0.000 claims abstract description 40
- 238000000034 method Methods 0.000 claims abstract description 36
- 238000012549 training Methods 0.000 claims abstract description 18
- 230000003068 static effect Effects 0.000 claims abstract description 9
- 239000011159 matrix material Substances 0.000 claims description 14
- 238000011835 investigation Methods 0.000 abstract description 3
- 238000005457 optimization Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 13
- 238000004590 computer program Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 10
- 238000013135 deep learning Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 8
- 230000015654 memory Effects 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 238000009826 distribution Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 230000002123 temporal effect Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Radio Relay Systems (AREA)
Abstract
The invention relates to a satellite link saturation evolution analysis method based on space-time diagram convolution, which comprises the following steps: step S1, satellite data are processed and cleaned, and satellite attribute information, service information and satellite link information are reserved as characteristic information of satellite nodes; s2, carrying out static topology modeling on a satellite link network according to different time slices, and sending each two adjacent frames of the space-time diagram after modeling into a space-time diagram convolution network for training as a training sample; s3, sending the convolved characteristics into a pre-measuring head, and outputting the saturation probability of each frequency band until model training converges; and S4, predicting the saturation of the satellite link by using the trained model. The method and the system can effectively capture the space-time dependence of the satellite link, enable the prediction and saturation analysis of the link load to be more accurate and reliable, can better adapt to complex and changeable communication environments, and provide more powerful support for network planning, resource optimization and fault investigation.
Description
Technical Field
The invention relates to the technical field of satellite links and space-time diagram convolution nerves, in particular to a satellite link saturation evolution analysis method based on space-time diagram convolution.
Background
With the rapid development of communication technology, satellite links play a vital role in modern communications and information transmission. The satellite link is not only used for international satellite communication, but also widely applied to a plurality of fields such as satellite internet, satellite broadcasting, satellite navigation, earth remote sensing and the like. In such diverse application scenarios, load balancing and performance analysis of satellite links become important issues for guaranteeing communication quality and service reliability. The invention defines the saturation of the satellite link according to the satellite link characteristics, designs an effective satellite link saturation analysis method, can help communication operators, government institutions and enterprises to plan and manage satellite link resources better, improves communication coverage, meets the increasing communication requirements, is an important task in the satellite communication field, and has wide value for various applications and fields.
Deep learning is also increasingly used in the satellite field. For example, deep learning has an important role in satellite image processing. Large-scale images acquired through satellite remote sensing data require efficient processing and analysis to extract useful geographic information. The deep learning method is excellent in image classification, target detection, change detection and the like, can help accurately identify information such as city planning, land utilization, resource distribution and the like, and provides powerful support for the fields such as city planning, agricultural management, natural resource protection and the like. Secondly, deep learning has wide application prospect in the field of satellite communication. With the continued expansion and upgrade of satellite communication networks, more intelligent network management and spectrum allocation methods are needed to cope with the increasing communication demands of users. The deep learning space-time diagram convolution technology can be used for optimizing satellite link resources, improving communication quality, improving network performance and reducing communication cost.
Therefore, a satellite link saturation evolution analysis method based on space-time diagram convolution is urgently needed.
Disclosure of Invention
In order to solve the technical problems in the prior art, the invention aims to provide a satellite link saturation evolution analysis method based on space-time diagram convolution, which can effectively capture space-time dependence of a satellite link and enable the prediction and saturation analysis of the link load to be more accurate and reliable.
In order to achieve the above object, the present invention provides a satellite link saturation evolution analysis method based on space-time diagram convolution, comprising the following steps:
step S1, satellite data are processed and cleaned, and satellite attribute information, service information and satellite link information are reserved as characteristic information of satellite nodes;
s2, carrying out static topology modeling on a satellite link network according to different time slices, and sending each two adjacent frames of the space-time diagram after modeling into a space-time diagram convolution network for training as a training sample;
s3, sending the convolved characteristics into a pre-measuring head, and outputting the saturation probability of each frequency band until model training converges;
and S4, predicting the saturation of the satellite link by using the trained model.
According to one aspect of the present invention, in the step S1, the satellite attribute information includes at least a satellite identifier, a satellite communication frequency band, a satellite orbit parameter, a satellite state parameter and a satellite quality; the service information at least comprises a service identifier, service arrival time, service data volume and a service destination station; the satellite link information includes at least a link frequency band, a link state and a link channel capacity.
According to one technical scheme of the invention, in the step S2, the space-time diagram convolution network is constructed by adopting a space-time convolution module formed by connecting eight layers of diagram convolution modules and a time convolution module residual.
According to one technical scheme of the invention, in the step S2, a time link is established between each node of the topological graph at the time t and the time t-1 by the time space graph convolution network, and when graph convolution is carried out each time, the graph convolution is carried out on the topological graph at the time t and the time t-1 as a whole;
wherein t-1 is the last time of t.
According to one embodiment of the present invention, in the step S2, the formula of the space-time diagram convolutional network is:
wherein A is n Is the adjacency matrix of the input graph, the node pair has a time link at the last moment, v of the two nodes 1 ,v 2 The adjacency matrix A of the graph of (2) isThen A n Is->The newly added rank is a time link, Λ n And the same is true.
According to one embodiment of the present invention, in the step S3, a loss function is a mean square error regression loss function, and a mean square error average is performed between the predicted value of the network and the true saturation of the next frame as the loss function of the network.
According to one technical scheme of the invention, the formula of the loss function is as follows:
wherein Y is the real satellite frequency band occupation condition at a certain moment, the value of Y is a K1 dimension matrix, the Y is single thermal coding, F (x) is the load probability condition of each frequency band in K frequency bands predicted by a network, and the load probability condition is 0-1.
According to one embodiment of the present invention, in the step S3, the output of the space-time diagram convolutional network is:
F(x)＝[p 1 ,p 2 ,…,p k ]
wherein p is i The load probability of each frequency band in K frequency bands predicted by the network is a value of 0-1.
According to one aspect of the present invention, the satellite link saturation is defined as:
compared with the prior art, the invention has the following beneficial effects:
the invention provides a satellite link saturation evolution analysis method based on space-time diagram convolution, which provides a more accurate, self-adaptive, efficient and practical solution in load analysis. The introduction of the space-time diagram convolution enhances the modeling of complex space-time relations, improves the accuracy of prediction and ensures that the analysis method has more practicability. Compared with the traditional method requiring manual parameter adjustment, the method has higher self-adaption and generalization performance, is beneficial to remarkably improving efficiency and quality in satellite communication operation, and brings technical progress and operation improvement to the industry.
Compared with a routing algorithm which only considers the static space-time weighted graph, the invention obviously improves the generalization capability of the algorithm. The traditional method cannot cope with real-time changes of satellite links, and an algorithm based on a space-time diagram convolutional neural network can capture dynamic topology information more intelligently. In addition, the scheme considers the link saturation, thereby being beneficial to reducing congestion and improving communication reliability. The scheme provides a more efficient and reliable solution for satellite link routing.
Furthermore, the application of deep learning brings higher intelligent and automatic level for the satellite field, and further improves the performance and benefits of satellite remote sensing, communication, navigation and the like. Meanwhile, the satellite technology can be promoted to develop towards more intelligent, reliable and efficient directions.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments will be briefly described below. It is apparent that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained from these drawings without inventive effort for a person of ordinary skill in the art.
FIG. 1 schematically shows a general flow chart of a satellite link saturation evolution analysis method based on space-time diagram convolution according to the invention;
FIG. 2 schematically illustrates a basic convolution module configuration diagram according to one embodiment of the present disclosure;
FIG. 3 schematically illustrates a graph wrapping process according to one embodiment of the invention;
fig. 4 schematically shows a flow chart of a satellite link saturation evolution analysis method based on space-time diagram convolution according to an embodiment of the present invention.
Detailed Description
The description of the embodiments of this specification should be taken in conjunction with the accompanying drawings, which are a complete description of the embodiments. In the drawings, the shape or thickness of the embodiments may be enlarged and indicated simply or conveniently. Furthermore, portions of the structures in the drawings will be described in terms of separate descriptions, and it should be noted that elements not shown or described in the drawings are in a form known to those of ordinary skill in the art.
Any references to directions and orientations in the description of the embodiments herein are for convenience only and should not be construed as limiting the scope of the invention in any way. The following description of the preferred embodiments will refer to combinations of features, which may be present alone or in combination, and the invention is not particularly limited to the preferred embodiments. The scope of the invention is defined by the claims.
As shown in fig. 1 to 4, the satellite link saturation evolution analysis method based on space-time diagram convolution of the invention comprises the following steps:
s1, configuring equipment components of an equipment system and configuring internal components of the equipment system;
step S2, generating and setting operation of an internal level of the system, and generating and setting operation of the system level;
s3, generating and configuring mission tasks of the equipment system;
s4, generating and setting an internal level working mode of the system, and generating and setting a system level working mode;
step S5, constructing and configuring a task mode of the equipment system.
In this embodiment, first, complex satellite data is processed and cleaned, and satellite attribute information, service information and satellite link information are reserved as characteristic information of the satellite node. And then, according to different time slice divisions, carrying out static topology modeling on a satellite link to form a satellite link space-time diagram, sending the space-time diagram matrix at the moment as training data into a space-time diagram convolution network for training, and calculating an optimized network model through special space-time diagram convolution and space-time loss. And finally, defining the saturation of the satellite link, and carrying out satellite link network evolution analysis according to the saturation.
In particular, the space-time diagram convolution allows the satellite link saturation analysis model to conduct information transfer and feature learning in the space-time dimension, thereby better understanding the topology of the communication network and the interrelation between nodes. By means of node embedding on the learning space-time diagram, the deep learning method can effectively capture space-time dependence of a satellite link, and prediction and saturation analysis of the link load are more accurate and reliable. The comprehensive modeling method enables the satellite link saturation analysis to be better suitable for complex and changeable communication environments, and provides more powerful support for network planning, resource optimization and fault investigation.
In one embodiment of the present invention, preferably, the space-time diagram convolution-based satellite link saturation evolution analysis method of the present invention includes a satellite link space-time diagram modeling method, a space-time diagram convolution neural network algorithm, and a satellite link saturation definition and analysis method.
1. Satellite data processing and space-time diagram convolutional network design
In the early stages of data acquisition, it may be affected by various sources of interference, such as sensor errors, communication channel interference, or environmental noise. Therefore, it is a primary task to clean and denoise raw data to eliminate outliers and bad data points. Helping to ensure that the satellite data used is of high quality, providing reliable characteristic information. Then, satellite attribute information, service information, and satellite link information are extracted as characteristic information of the satellite node. The satellite attribute information comprises a satellite identifier Sid, a satellite communication frequency band Srange, a satellite orbit parameter Sop, a satellite state parameter Ssp and a satellite quality Sq; the service information comprises a service identifier Bid, a service arrival time Bstart, a service data volume Bdata and a service destination station Bdes; the satellite link information includes a link band Lrange, a link state Lstatus, and a link channel capacity Lcap.
The above procedure is called a data preprocessing stage, and next, detailed static topology modeling will be performed on the satellite link network based on different time slices. The key objective of this modeling process is to capture the complexity of the space-time relationships in the satellite link network. Space-time convolutional networks will be trained to understand these relationships to better predict the saturation of satellite links. The data conversion between different time slices is utilized, so that the network can adapt to the change of the link topology, and the accuracy is maintained. The data for each adjacent two time slices is then converted into a training small lot. This training small batch will be fed as input into the space-time diagram convolutional network for training. Meanwhile, a part of data in the space-time diagram is reserved as test data for testing the performance of the space-time diagram convolutional network model.
The space-time diagram convolution network adopts a highly complex architecture, and is formed by connecting eight layers of diagram convolution modules and a residual error of the time convolution module. The core idea of this design is to combine spatial information and temporal information to more fully understand the evolution process of the satellite link network. Specifically, each graph convolution module is responsible for extracting important features from the spatial graph, while the temporal convolution module captures the temporal dependencies. The information of the two aspects is organically integrated together, and the integrity and consistency of the information are ensured through residual connection transmission. Finally, after passing through these complex modules, the features are fed into the pre-header, which is the last part of the network. The task of the pre-measurement head is to translate these features into saturation probabilities for the individual frequency bands. Once the model has been trained and converged, it will be able to predict the saturation of the satellite link with a high degree of accuracy, making the analysis of the satellite link saturation more efficient and accurate. Not only helps to identify the performance bottlenecks of the network in time, but also provides powerful data support for decision-makers to better plan and manage satellite communication networks. The space-time diagram convolution network provides an advanced tool for monitoring and predicting the saturation of the satellite link.
The key principle of the satellite link space-time diagram modeling method is to extract key features of the satellite link and conduct static topology modeling so as to better understand the dynamic property and complexity of the link network. Firstly, in the data preprocessing stage, satellite data are cleaned and denoised, so that the quality and reliability of the data are ensured. Then, satellite attributes, service information, and satellite link information are extracted from the satellite data, which will be used as features to describe the state and performance of the satellite link network. The data in the different time slices is then used to construct a static topology model of the satellite link. This modeling process aims at capturing the structure and evolution over time of the satellite link network. By modeling this topology, the performance and saturation of the satellite link network can be better understood. The importance of this method is that it provides an efficient way to analyze the saturation of satellite links, by cleaning the data, extracting features and building a topology model, it is possible to better understand the performance of the satellite link network and take corresponding measures to improve its performance and reliability.
2. Space-time diagram convolution detailed solution and loss function design
As shown in fig. 2, the basic convolution module structure is relatively simple but is the core of the overall graph convolution network module. First, the input features undergo a two-dimensional convolution operation 1*1 that helps learn the nonlinear mapping of the features. The einstein summing convention is then used to calculate the sum of the convolved feature matrices, which facilitates the overall processing of the features. Subsequently, a portion of neurons are randomly turned off by a DropOut layer, thereby preventing the model from being too dependent on a particular neuron, and improving generalization ability. The features are then passed to a time convolution network module, which is an important component of the overall convolution network. In the time convolution module, features are firstly sent to the BN (Batch Normalization) layer to perform normalization operation of the features, which is helpful for accelerating the training process. Then, nonlinear mapping is performed through the ReLU activation function, and the complexity of the features is further extracted. The features are then subjected to a two-dimensional convolution operation 9*1 to facilitate the correlation of the captured timing information with the frequency domain information. And finally, carrying out feature normalization and DropOut layer through the BN layer again to improve the generalization capability of the model and reduce the risk of overfitting.
The core part of the whole graph convolution network is formed by repeatedly connecting 8 times by the basic convolution module so as to fully extract and integrate the characteristic information. Finally, the characteristics are sent to a prediction head for outputting the load probability of each frequency band, and the saturation prediction task of the satellite link is completed. The structure and the hierarchical distribution of the model are beneficial to effectively capturing the space-time information and improving the performance of the model.
The formula of the conventional graph convolution is as follows:
Λ ii ＝∑ j (A ij +I ij )
wherein A is the adjacent matrix of the input diagram, I is the identity matrix, and A+I is the self-loop added to the input diagram to ensure the effectiveness of data transmission. W is composed of weight vectors of multiple output channels, f in Is a feature map of the input, f out Is the output signature. Λ type ii Is the diagonal element of the adjacency matrix, and the calculation process of the value is to sum the elements of the corresponding positions of the input graph and the identity matrix.
The formula of the graph convolution of the invention is as follows:
wherein A is n Still being an adjacency matrix of the input graph, but its nodes have a time link to their own moment, which is equivalent to the graph expansion. With v having two nodes 1 ,v 2 For example, the adjacency matrix A isThen A n Is thatThe newly added rank is a time link, Λ n And the same is true.
The loss function adopts a mean square error loss, and a specific function formula is as follows:
wherein Y is the real satellite frequency band occupation condition at a certain moment, the value of Y is a K1-dimensional matrix, and the Y is one-time thermal coding. F (x) is the load probability of each frequency band in K frequency bands predicted by the network, and the load probability is a value of 0-1, and the square sum of the difference between the two values is the mean square error, namely the loss function of the invention.
Space-time convolutional neural networks play a key role in satellite link saturation analysis. The core principle is that the complexity of space-time data is combined, and the space-time relationship is encoded into the neural network, so that the network can better understand the evolution and load distribution of satellite links. The deep learning method for fusing the space-time information has the advantages compared with the traditional method, and on one hand, the space-time diagram convolution can capture the space-time dependence. The performance of the satellite link is related to time and space, and the network can effectively capture hidden space-time dependency through the space-time diagram learning module, which helps to more accurately predict the saturation of the satellite link. On the other hand, the space-time diagram convolutional neural network provides a more accurate saturation analysis result through the strong feature extraction and space-time modeling capability of the deep learning model, and is favorable for taking proper adjustment measures in advance to ensure the normal operation of the network. In addition, the space-time convolutional neural network has adaptability, and the adaptability enables the management of satellite links to be more flexible and reliable. The significance of the space-time diagram convolutional neural network in satellite link saturation analysis is that the space-time relation-based deep learning model can analyze and predict the load state of the satellite link more comprehensively, accurately and adaptively, and is beneficial to improving the performance and reliability of the satellite communication network.
3. Satellite link saturation calculation and analysis
F(x)＝[p 1 ,p 2 ,…,p k ]
The last output of the convolution network of the space-time diagram is p i The load probability of each frequency band in K frequency bands predicted by the network is a value of 0-1. And the saturation of the satellite link is defined as:
high saturation indicates that the satellite link network is in a very busy state, meaning that many satellite communications bands have approached or exceeded their load capacity. In this case, the network performance may be significantly affected, such as an increase in packet loss rate, an increase in delay, and even an increase in risk of communication interruption. This is a serious problem for satellite communication networks, as these networks are commonly used for transmitting important data and communications, such as in the fields of emergency rescue, remote monitoring and communications. Conversely, low saturation means that the satellite link network is relatively lightly loaded. This may mean that the network has sufficient remaining capacity to handle additional communication traffic, providing faster response times and more reliable communication services. However, too low saturation may also indicate that resources are underutilized, resulting in inefficient network performance. Thus, maintaining moderate saturation is critical to the efficient operation of the satellite link network.
The core principle of the satellite link saturation analysis method is to define the saturation of the satellite link by predicting and accumulating the load probabilities of K satellite communication frequency bands. The saturation reflects the load condition of the satellite link network, and a high saturation indicates that the network is heavy in load, and a low saturation indicates that the network is relatively light in load. The satellite communication system can be monitored and managed in real time, and continuity of data transmission and communication service is ensured. Accurate saturation analysis helps take action in advance to prevent network congestion or resource waste, thereby maintaining performance and reliability of satellite communications. Therefore, the satellite link saturation analysis method is a key tool for ensuring the smooth operation of the satellite communication system, has important practical application significance,
the state of the satellite link can be better understood by network operators through the prediction and monitoring of the saturation of the satellite link by the space-time diagram convolution network, measures are timely taken to optimize the network performance and allocate resources, and the high quality and reliability of communication are ensured, which is important for improving the efficiency of the satellite link network in different application scenes.
According to an aspect of the present invention, there is provided an electronic apparatus including: one or more processors, one or more memories, and one or more computer programs; wherein the processor is connected to the memory, and the one or more computer programs are stored in the memory, and when the electronic device runs, the processor executes the one or more computer programs stored in the memory, so that the electronic device executes a satellite link saturation evolution analysis method based on space-time diagram convolution according to any one of the above technical solutions.
According to an aspect of the present invention, there is provided a computer readable storage medium for storing computer instructions which, when executed by a processor, implement a method for analysis of satellite link saturation evolution based on space-time diagram convolution according to any one of the above technical solutions.
The invention discloses a satellite link saturation evolution analysis method based on space-time diagram convolution, which comprises the following steps: step S1, satellite data are processed and cleaned, and satellite attribute information, service information and satellite link information are reserved as characteristic information of satellite nodes; s2, carrying out static topology modeling on a satellite link network according to different time slices, and sending each two adjacent frames of the space-time diagram after modeling into a space-time diagram convolution network for training as a training sample; s3, sending the convolved characteristics into a pre-measuring head, and outputting the saturation probability of each frequency band until model training converges; s4, predicting the saturation of the satellite link by using the trained model; space-time diagram convolution allows the satellite link saturation analysis model to conduct information transfer and feature learning in the space-time dimension, so that the topological structure of the communication network and the interrelation between nodes are better understood. By means of node embedding on the learning space-time diagram, the deep learning method can effectively capture space-time dependence of the satellite link, so that prediction and saturation analysis of the link load are more accurate and reliable, the satellite link saturation analysis can be better adapted to complex and changeable communication environments, and more powerful support is provided for network planning, resource optimization and fault investigation.
Furthermore, it should be noted that the present invention can be provided as a method, an apparatus, or a computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the invention may take the form of a computer program product on one or more computer-usable storage media having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It should also be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
It is finally pointed out that the above description of the preferred embodiments of the invention, it being understood that although preferred embodiments of the invention have been described, it will be obvious to those skilled in the art that, once the basic inventive concepts of the invention are known, several modifications and adaptations can be made without departing from the principles of the invention, and these modifications and adaptations are intended to be within the scope of the invention. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Claims (9)
1. The satellite link saturation evolution analysis method based on space-time diagram convolution is characterized by comprising the following steps of:
step S1, satellite data are processed and cleaned, and satellite attribute information, service information and satellite link information are reserved as characteristic information of satellite nodes;
s2, carrying out static topology modeling on a satellite link network according to different time slices, and sending each two adjacent frames of the space-time diagram after modeling into a space-time diagram convolution network for training as a training sample;
s3, sending the convolved characteristics into a pre-measuring head, and outputting the saturation probability of each frequency band until model training converges;
and S4, predicting the saturation of the satellite link by using the trained model.
2. The method for analyzing satellite link saturation evolution based on space-time diagram convolution according to claim 1, wherein in the step S1, the satellite attribute information includes at least a satellite identifier, a satellite communication frequency band, a satellite orbit parameter, a satellite state parameter and a satellite quality; the service information at least comprises a service identifier, service arrival time, service data volume and a service destination station; the satellite link information includes at least a link frequency band, a link state and a link channel capacity.
3. The method for analyzing satellite link saturation evolution based on space-time diagram convolution according to claim 1, wherein in the step S2, the space-time diagram convolution network is constructed by adopting a space-time convolution module formed by connecting eight layers of diagram convolution modules with a residual error of the time convolution module.
4. The satellite link saturation evolution analysis method based on space-time diagram convolution according to claim 3, wherein in the step S2, the space-time diagram convolution network establishes a time link between each node of the topological graph at the time t and the time t-1, and each time the graph convolution is performed, the graph convolution is performed on the topological graph at the time t and the time t-1 as a whole;
wherein t-1 is the last time of t.
5. The method for analyzing satellite link saturation evolution based on space-time diagram convolution according to claim 4, wherein in the step S2, the formula of the space-time diagram convolution network is:
wherein A is n Is the adjacency matrix of the input graph, the node pair has a time link at the last moment, v of the two nodes 1 ,v 2 The adjacency matrix A of the graph of (2) isThen A n Is->The newly added rank is a time link, Λ n And the same is true.
6. The method for evolution analysis of satellite link saturation based on space-time diagram convolution according to claim 5, wherein in step S3, a loss function is a mean square error regression loss function, and a mean square error is averaged between a network predicted value and the true saturation of the next frame as the loss function of the network.
7. The space-time diagram convolution-based satellite link saturation evolution analysis method according to claim 6, wherein the formula of the loss function is:
wherein Y is the real satellite frequency band occupation condition at a certain moment, the value of Y is a K1 dimension matrix, the Y is single thermal coding, F (x) is the load probability condition of each frequency band in K frequency bands predicted by a network, and the load probability condition is 0-1.
8. The method for analyzing satellite link saturation evolution based on space-time diagram convolution according to claim 1, wherein in the step S3, the output of the space-time diagram convolution network is:
F(x)＝[p 1 ,p 2 ,…,p k ]
wherein p is i The load probability of each frequency band in K frequency bands predicted by the network is a value of 0-1.
9. The space-time diagram convolution-based satellite link saturation evolution analysis method according to claim 8, wherein the satellite link saturation is defined as:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311765807.XA CN117828292A (en) | 2023-12-20 | 2023-12-20 | Satellite link saturation evolution analysis method based on space-time diagram convolution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311765807.XA CN117828292A (en) | 2023-12-20 | 2023-12-20 | Satellite link saturation evolution analysis method based on space-time diagram convolution |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117828292A true CN117828292A (en) | 2024-04-05 |
Family
ID=90514572
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311765807.XA Pending CN117828292A (en) | 2023-12-20 | 2023-12-20 | Satellite link saturation evolution analysis method based on space-time diagram convolution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117828292A (en) |
-
2023
- 2023-12-20 CN CN202311765807.XA patent/CN117828292A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jiang | Cellular traffic prediction with machine learning: A survey | |
| GB2593180A (en) | A distributed network traffic data decomposition method | |
| Tian | Network traffic prediction method based on wavelet transform and multiple models fusion | |
| CN111245673A (en) | SDN time delay sensing method based on graph neural network | |
| CN111294812A (en) | Method and system for resource capacity expansion planning | |
| CN117201410B (en) | Flow management method and system for Internet of things | |
| CN112116078A (en) | Information security baseline learning method based on artificial intelligence | |
| CN115065992A (en) | Edge calculation-based cooperative sensing and resource allocation method | |
| CN114265913A (en) | Space-time prediction algorithm based on federal learning on industrial Internet of things edge equipment | |
| CN112115830A (en) | Target distributed fusion recognition method based on bit domain feature extraction | |
| CN113541986B (en) | Fault prediction method and device for 5G slice and computing equipment | |
| CN117828292A (en) | Satellite link saturation evolution analysis method based on space-time diagram convolution | |
| CN116545871A (en) | Multi-mode network traffic prediction method, device and medium | |
| CN114401063B (en) | Edge equipment cooperative spectrum intelligent monitoring method and system based on lightweight model | |
| CN115758643A (en) | Network flow prediction method and device based on temporal-spatial feature fusion and storage medium | |
| CN115277789A (en) | Safety protection system and method for cascade hydropower station | |
| CN113780371A (en) | Insulator state edge recognition method based on edge calculation and deep learning | |
| CN114615693A (en) | Network capacity prediction method, device, electronic equipment and computer storage medium | |
| CN113822135A (en) | Video processing method, device and equipment based on artificial intelligence and storage medium | |
| CN113949633A (en) | 5G network slice disaster recovery pool resource management method and device based on machine learning | |
| LIU et al. | A case study on intelligent operation system for wireless networks | |
| CN113098707B (en) | Virtual network function demand prediction method in edge network | |
| CN117440135B (en) | Image transmission method and system based on Beidou satellite communication | |
| CN117692026B (en) | Link sensing method and device for power line communication | |
| CN116366133B (en) | Unloading method and device based on low-orbit satellite edge calculation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |