CN117674961A - Low orbit satellite network time delay prediction method based on space-time feature learning - Google Patents
Low orbit satellite network time delay prediction method based on space-time feature learning Download PDFInfo
- Publication number
- CN117674961A CN117674961A CN202311542525.3A CN202311542525A CN117674961A CN 117674961 A CN117674961 A CN 117674961A CN 202311542525 A CN202311542525 A CN 202311542525A CN 117674961 A CN117674961 A CN 117674961A
- Authority
- CN
- China
- Prior art keywords
- time
- service
- delay
- network
- low
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 230000005540 biological transmission Effects 0.000 claims abstract description 40
- 238000012549 training Methods 0.000 claims abstract description 21
- 238000012360 testing method Methods 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 14
- 230000000737 periodic effect Effects 0.000 claims description 11
- 239000011159 matrix material Substances 0.000 claims description 10
- 239000013598 vector Substances 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000004891 communication Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 5
- 230000004069 differentiation Effects 0.000 claims description 5
- 238000005315 distribution function Methods 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 230000001934 delay Effects 0.000 claims description 2
- 238000005096 rolling process Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013468 resource allocation Methods 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)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention relates to a time delay prediction method of a low-orbit satellite network based on space-time feature learning, which constructs a network service time delay time sequence data set by carrying out flow modeling on three different services in the low-orbit satellite network and simulating a service flow transmission process under a multi-time slice topology, and counting the propagation delay and the on-board queue delay of corresponding services. And simultaneously, the spatial characteristics of the dynamic topology under each time slice are extracted by using a graph convolution network, a time delay data set is combined to be used as output, the output is input into a long-short-term memory network for training, the time delay condition under a certain time sequence at the next stage is predicted, and error analysis is performed on corresponding statistical data, so that the time delay accurate prediction of the dynamic topology of the low-orbit satellite network and the complex service environment is realized.
Description
Technical Field
The invention relates to the technical field of satellites, in particular to a low-orbit satellite network time delay prediction method based on space-time feature learning.
Background
In recent years, with the continued development of low-orbit satellite networks, the network scale and traffic volume have increased. Compared with a medium-high orbit satellite, the low orbit satellite network has the characteristics of high dynamic performance, wide coverage and massive access. The low-orbit satellite orbit is lower in height and smaller in inter-satellite distance, so that the inter-satellite communication transmission time is shorter, and the method can meet the increasingly diversified user demands of user services and the global real-time end-to-end coverage.
However, compared with the traditional ground network, on-board satellite resources are limited, as the number of services increases, the queuing time of the satellite node increases, so that the queuing processing time increases, and queuing time delay is generated. Meanwhile, because wireless communication is adopted between satellites, the transmission distance is farther than the ground, and the channel environment is worse, so that the propagation delay also exists in the inter-satellite communication, but the magnitude of the propagation delay is relatively fixed. The propagation delay and the queuing delay form the total delay of the network service, and as the user demand increases, the low-delay requirement is also put forward on the low-orbit satellite Internet.
The research at the present stage focuses on optimizing the on-board resource allocation in the network through a routing strategy or a load balancing strategy, so that the node congestion in the network is reduced, and the network time delay is further reduced, but the methods cannot be optimized for the time delay per se. Therefore, a method for realizing high-precision prediction of the whole network service delay based on delay prediction is needed.
Disclosure of Invention
In view of the above, the present invention aims to solve the above problems, and provides a low-orbit satellite network delay prediction method based on space-time feature learning.
The embodiment of the invention provides a low-orbit satellite network time delay prediction method based on space-time feature learning, which comprises the following steps:
s100, constructing a low-orbit satellite network, and forming a low-orbit satellite network topology matrix under multiple time slices;
s200, taking a topology matrix under each time slice as input, and extracting the spatial characteristics of the low orbit satellite network topology by using a graph convolution network;
s300, modeling burst service, periodic service and long-time transmission service in a low-orbit satellite network and respectively forming flow data sets;
s400, deploying traffic service under the current topology, and counting time delay conditions in the transmission process of each service, wherein the time delay consists of propagation delay and on-board queue delay, so as to form a service time delay time sequence data set;
s500, preprocessing the time delay data set, and dividing the time delay data set into a training set and a testing set. The time delay data and the space feature vector are taken as input together, training is carried out through a long-short-term memory network (LSTM) algorithm, the time delay change condition in a period of time is predicted, and the time delay change condition is compared with a test set.
In a preferred embodiment of the present invention, the S100 includes:
constructing M low-orbit satellite orbit planes, wherein each orbit plane consists of N low-orbit satellites, and a low-orbit satellite network consisting of M multiplied by N satellites is obtained;
outputting network topology and inter-satellite link distances in the low-orbit satellite network under K time slices, forming a network topology snapshot, and converting the network topology snapshot into a corresponding network adjacency matrix Map K ={Map 1 ,Map 2 ,...,Map k And each time slice T K ={T 1 ,T 2 ,...,T k The underlying network topology is fixed.
In a preferred embodiment of the present invention, the S200 includes:
topology of multiple time slicesThe matrices are integrated and as input the spatial feature vectors of the topology are extracted from the graph convolution network
In a preferred embodiment of the present invention, the S300 includes:
s310, defining three types of traffic services M= { M in the network 1 ,M 2 ,M 3 },M 1 For periodic traffic, M 2 For long-time transmission of class services, M 3 Is burst service;
s320, periodic service M 1 Characterizing the data length and the arrival rate distribution through a Poisson distribution probability density function;
s330, long-time transmission service M 2 Characterizing the data length and the transmission time interval by a gaussian distribution function;
s340, burst service M 3 Is characterized by an exponential distribution;
s350, according to M 1 、M 2 And M 3 Randomly generating a source and a sink according to the characteristic distribution of the network traffic data set; the total number of the business is N M And N M =N 1 +N 2 +N 3 Wherein N is 1 ,N 2 And N 3 Respectively M 1 、M 2 And M 3 Corresponding number of services.
In a preferred embodiment of the present invention, the S320 includes:
periodic traffic M 1 The method comprises the steps of inter-satellite daily communication and control instruction service, wherein arrival rate distribution of the service is represented by a Poisson distribution probability density function, and the Poisson distribution probability density is as follows:
where λ represents the average number of occurrences of the service, k represents the number of occurrences of the service, and e is a natural logarithmic constant.
In a preferred embodiment of the present invention, the S330 includes:
long-time transmission class service M 2 The method comprises the steps of representing a transmission time interval through a Gaussian distribution function, wherein the Gaussian probability density function is as follows:
wherein σ represents the standard deviation of the traffic transmission interval, e is a natural logarithmic constant, and μ represents the average value of the traffic transmission interval.
In a preferred embodiment of the present invention, the S340 includes:
burst class of service M 3 The method comprises the steps of temporarily generating end-to-end burst large stream data in a network, wherein the transmission time interval can be characterized by exponential distribution, and the probability density of an exponential function is as follows:
h(x)=λe -λx ,x≥0
where λ is a positive real number, and is a parameter of an exponential distribution, its value is the inverse of the expected value of the traffic transmission time interval distribution, e is a natural logarithmic constant, and x represents the traffic transmission time interval.
In a preferred embodiment of the present invention, the S400 includes:
s410, traffic service is deployed in the network topology, route planning is carried out for each service flow by adopting a shortest path method, and satellite queuing delay during service flow transmission is countedPropagation delay between satellites->
S420, the time delay of each service isAnd { t } is set in any time slice 1 ,t 2 ,...,t k }∈T k ;
S430, constructing a corresponding data set according to the time line according to all the counted traffic service delays of the time slices, wherein the data set comprises two parts of contents, one part is used for counting the time node of each time of starting traffic service as the starting time to form a corresponding time sequence, and the other part is used for counting the service delay of the corresponding service from the information source to the information sink.
In a preferred embodiment of the present invention, the S410 includes:
calculating inter-satellite link propagation speed threshold C from inter-satellite link distance i Simultaneously, a queuing queue is constructed on the satellite, and when the traffic flow is transmitted, the traffic flow is larger than a propagation speed threshold C i When the redundant part waits on the satellite, the queue waiting time is generated, thereby determining the delay of the queue on the satellite
Calculating transmission time as propagation delay according to inter-satellite distance
In a preferred embodiment of the present invention, the S500 includes:
s510, taking a part of the generated time delay data set as a training set, taking the rest as a test set, carrying out data preprocessing on the training set, wherein the preprocessing comprises normalization and differentiation, converting the normalization and differentiation into time characteristics of service flows, and then putting the time characteristics and space characteristic vectors into an LSTM multi-input single-output model as input for training.
S520, model training is completed by continuously adjusting the learning rate, the loss rate and the iteration times, service flow time delay results equal to the time sequence of the test set are predicted, the service flow time delay results are compared with actual results in the test set, and the relative error of the predicted results is calculated according to the cross entropy function.
According to the low-orbit satellite network time delay prediction method based on space-time feature learning, provided by the embodiment of the invention, the prediction of network service time delay under dynamic network topology and complex network service environment is realized by extracting space-time features in a network based on an LSTM prediction model.
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 needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a low-orbit satellite network time delay prediction method based on space-time feature learning according to an embodiment of the invention;
fig. 2 is a schematic diagram of a flow of a space-time feature extraction and training algorithm in a low-orbit satellite network time delay prediction method based on space-time feature learning according to an embodiment of the present invention;
fig. 3 is a schematic diagram comparing a time delay prediction result with an actual result of a low-orbit satellite network time delay prediction method based on space-time feature learning 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, the low-orbit satellite network time delay prediction method based on space-time feature learning in the embodiment of the invention comprises the following steps:
s100, constructing a low-orbit satellite network, and forming a low-orbit satellite network topology matrix under multiple time slices;
s200, taking a topology matrix under each time slice as input, and extracting the spatial features of the satellite network topology by using GCN;
s300, defining three traffic services in a network, namely burst service, periodic service and long-time transmission service, modeling the three traffic services and forming a traffic data set;
s400, deploying traffic service under the current topology, and counting time delay conditions in the transmission process of each service, wherein the time delay consists of propagation delay and on-board queue delay, so as to form a service time delay time sequence data set;
s500, preprocessing the time delay data set, and dividing the time delay data set into a training set and a testing set. The time delay data and the space feature vector are taken as input together, training is carried out through a long-short-term memory network (LSTM) algorithm, the time delay change condition in a period of time is predicted, and the time delay change condition is compared with a test set.
Each step is further described below according to this embodiment:
s100, constructing M low-orbit satellite orbit planes, wherein each orbit plane consists of N low-orbit satellites, and a low-orbit satellite network consisting of M multiplied by N satellites is obtained; outputting network topology and inter-satellite link distances in the low-orbit satellite network under K time slices, forming a network topology snapshot, and converting the network topology snapshot into a corresponding network adjacency matrix Map K ={Map 1 ,Map 2 ,...,Map k Default per time slice T K ={T 1 ,T 2 ,...,T k The underlying network topology is fixed.
S200, integrating the topology matrixes of the time slices, and extracting the space feature vector of the topology from a graph rolling network (GCN) as input
S300, defining three types of traffic services M= { M in a network 1 ,M 2 ,M 3 And the traffic is respectively periodic traffic, long-time continuous traffic and burst traffic. Wherein periodic traffic M 1 The method comprises the inter-satellite daily communication class and control instruction service, and the arrival rate distribution condition of the service can be represented through a Poisson distribution probability density function. The Poisson distribution probability density is:
where λ represents the average number of occurrences of the service, k represents the number of occurrences of the service, and e is a natural logarithmic constant.
Continuously transmitting traffic M for a long time 2 Then audio and video streaming traffic is included, the amount of data is large, and such traffic can characterize the transmission time interval with a gaussian distribution function. The Gaussian probability density function is:
wherein sigma represents standard deviation of service transmission interval, is an important parameter of Gaussian distribution, e is natural logarithmic constant, and mu represents average value of service transmission interval.
Burst class of service M 3 Then end-to-end bursty large stream data is included that is temporarily generated within the network and the transmission time interval of which may be characterized by an exponential distribution. The exponential function probability density is:
h(x)=λe -λx ,x≥0
where λ is a positive real number, and is a parameter of an exponential distribution, its value is the inverse of the expected value of the traffic transmission time interval distribution, e is a natural logarithmic constant, and x represents the traffic transmission time interval.
S400, the traffic service is deployed in the network, the routing planning is carried out for each service flow by adopting the shortest path method, and the satellite queuing delay during the transmission of the service flow is countedTransmission delay between satellites->And the service delay is the sum of the propagation delay and the queue delay, and a service delay data set under a plurality of time slices is constructed according to the sum.
S500, taking part of the generated time delay data set as a training set and the rest as a test set, wherein the ratio is (a: 1-a), 0< a <1. The training set is preprocessed by data including normalization and differentiation, the data are converted into time features of service flow, and then the time features and the space feature vectors are used as input together to be put into an LSTM multi-input single-output model for training.
Model training is completed by continuously adjusting parameters such as learning rate, loss rate, iteration number and the like, service flow time delay results with equal length of a time sequence of a test set are predicted, the service flow time delay results are compared with actual results in the test set, and relative errors of the prediction results are calculated according to a cross entropy function.
As shown in fig. 2, fig. 2 is a flow of a spatiotemporal feature extraction and training algorithm. The method comprises the steps of extracting satellite network topology space feature information based on time sequence by using GCN, wherein time delay data represent time features of traffic service, sequentially inputting the time and space features into an LSTM model according to time sequence by combining the time and space features as input, and predicting time delay at the next time.
As shown in fig. 3, fig. 3 is a relative error comparison chart of a prediction of time delay and a statistical value in a data set under a certain time sequence by adopting the low-orbit satellite network time delay prediction method based on topology space-time feature learning in the embodiment. As can be seen from fig. 3, the embodiment of the present invention can effectively implement better prediction for time delay under various service scenarios of the low-orbit satellite network.
The low orbit satellite network time delay prediction method based on space-time feature learning, provided by the embodiment of the invention, is based on an LSTM prediction model, and realizes the prediction of network service time delay under dynamic network topology and complex network service environment.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (10)
1. A low-orbit satellite network time delay prediction method based on space-time feature learning, which is characterized by comprising the following steps:
s100, constructing a low-orbit satellite network, and forming a low-orbit satellite network topology matrix under multiple time slices;
s200, taking a topology matrix under each time slice as input, and extracting the spatial characteristics of the low orbit satellite network topology by using a graph convolution network;
s300, modeling burst service, periodic service and long-time transmission service in a low-orbit satellite network and respectively forming flow data sets;
s400, deploying traffic service under the current network topology, and counting time delay conditions in the transmission process of each service, wherein the time delay consists of propagation delay and on-board queue delay, so as to form a service time delay time sequence data set;
s500, preprocessing a time delay data set, dividing the time delay data set into a training set and a testing set, taking the time delay data and the space feature vector as inputs, training through a long-short-term memory network algorithm, predicting time delay change conditions in a period of time, and comparing the time delay data set with the testing set.
2. The method for predicting low-orbit satellite network delay based on space-time feature learning according to claim 1, wherein S100 comprises:
constructing M low-orbit satellite orbit planes, wherein each orbit plane consists of N low-orbit satellites, and a low-orbit satellite network consisting of M multiplied by N satellites is obtained;
outputting network topology and inter-satellite link distances in the low-orbit satellite network under K time slices, forming a network topology snapshot, and converting the network topology snapshot into a corresponding network adjacency matrix Map K ={Map 1 ,Map 2 ,…,Map k Each time sliceT k ={T 1 ,T 2 ,…,T k The underlying network topology is fixed.
3. The method for predicting low-orbit satellite network delay based on space-time feature learning according to claim 2, wherein S200 comprises:
integrating the topology matrix of the time slices and extracting the space feature vector of the topology from the graph rolling network as input
4. The method for predicting low-orbit satellite network delay based on topology space-time feature learning according to claim 1, wherein S300 comprises:
s310, defining three types of traffic services M= { M in the network 1 ,M 2 ,M 3 },M 1 For periodic traffic, M 2 For long-time transmission of class services, M 3 Is burst service;
s320, periodic service M 1 Characterizing the data length and the arrival rate distribution through a Poisson distribution probability density function;
s330, long-time transmission service M 2 Characterizing the data length and the transmission time interval by a gaussian distribution function;
s340, burst service M 3 Is characterized by an exponential distribution;
s350, according to M 1 、M 2 And M 3 Randomly generating a source and a sink according to the characteristic distribution of the network traffic data set; the total number of the business is N M And N M =N 1 +N 2 +N 3 Wherein N is 1 ,N 2 And N 3 Respectively M 1 、M 2 And M 3 Corresponding number of services.
5. The method for predicting low-orbit satellite network delay based on space-time feature learning according to claim 4, wherein S320 comprises:
periodic traffic M 1 The method comprises the steps of inter-satellite daily communication and control instruction service, wherein arrival rate distribution of the service is represented by a Poisson distribution probability density function, and the Poisson distribution probability density is as follows:
where λ represents the average number of occurrences of the service, k represents the number of occurrences of the service, and e is a natural logarithmic constant.
6. The method for predicting low-orbit satellite network delay based on space-time feature learning according to claim 4, wherein S330 comprises:
long-time transmission class service M 2 The method comprises the steps of representing a transmission time interval through a Gaussian distribution function, wherein the Gaussian probability density function is as follows:
wherein σ represents the standard deviation of the traffic transmission interval, e is a natural logarithmic constant, and μ represents the average value of the traffic transmission interval.
7. The method for predicting low-orbit satellite network delay based on space-time feature learning according to claim 4, wherein S340 comprises:
burst class of service M 3 The method comprises the steps of temporarily generating end-to-end burst large stream data in a network, wherein the transmission time interval can be characterized by exponential distribution, and the probability density of an exponential function is as follows:
h(x)=λe -λx ,x≥0
where λ is a positive real number, and is a parameter of an exponential distribution, its value is the inverse of the expected value of the traffic transmission time interval distribution, e is a natural logarithmic constant, and x represents the traffic transmission time interval.
8. The method for predicting low-orbit satellite network delay based on space-time feature learning according to claim 1, wherein S400 comprises:
s410, traffic service is deployed in the network topology, route planning is carried out for each service flow by adopting a shortest path method, and satellite queuing delay during service flow transmission is countedPropagation delay between satellites->
S420, the time delay of each service isAnd { t } is set in any time slice 1 ,t 2 ,...,t k }∈T k ;
S430, constructing a corresponding data set according to the time line according to all the counted traffic service delays of the time slices, wherein the data set comprises two parts of contents, one part is used for counting the time node of each time of starting traffic service as the starting time to form a corresponding time sequence, and the other part is used for counting the service delay of the corresponding service from the information source to the information sink.
9. The method for predicting low-orbit satellite network delay based on space-time feature learning according to claim 8, wherein S410 comprises:
calculating inter-satellite link propagation speed threshold C from inter-satellite link distance i Simultaneously, a queuing queue is constructed on the satellite, and when the traffic flow is transmitted, the traffic flow is larger than a propagation speed threshold C i When the redundant part waits on the satellite, the queue waiting time is generated, thereby determining the delay of the queue on the satellite
Calculating transmission time as propagation delay according to inter-satellite distance
10. The method for predicting low-orbit satellite network delay based on space-time feature learning according to claim 1, wherein S500 comprises:
s510, taking a part of the generated time delay data set as a training set, taking the rest as a test set, carrying out data preprocessing on the training set, wherein the preprocessing comprises normalization and differentiation, converting the normalization and differentiation into time characteristics of service flow, and then putting the time characteristics and the space characteristic vector into a multi-input single-output model of a long-period memory network as input for training.
S520, model training is completed by continuously adjusting the learning rate, the loss rate and the iteration times, service flow time delay results equal to the time sequence of the test set are predicted, the service flow time delay results are compared with actual results in the test set, and the relative error of the predicted results is calculated according to the cross entropy function.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311542525.3A CN117674961B (en) | 2023-11-20 | 2023-11-20 | Low orbit satellite network time delay prediction method based on space-time feature learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311542525.3A CN117674961B (en) | 2023-11-20 | 2023-11-20 | Low orbit satellite network time delay prediction method based on space-time feature learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117674961A true CN117674961A (en) | 2024-03-08 |
CN117674961B CN117674961B (en) | 2024-05-28 |
Family
ID=90078069
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311542525.3A Active CN117674961B (en) | 2023-11-20 | 2023-11-20 | Low orbit satellite network time delay prediction method based on space-time feature learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117674961B (en) |
Citations (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2001015001A2 (en) * | 1999-08-23 | 2001-03-01 | General Electric Company | Apparatus and method for managing a fleet of mobile assets |
CN101299713A (en) * | 2008-03-21 | 2008-11-05 | 哈尔滨工业大学深圳研究生院 | Method for setting multilayer satellite network system route |
US20100090889A1 (en) * | 2006-09-29 | 2010-04-15 | Yoola Hwang | Precise orbit determination system and method using gps data and galileo data |
CN103716802A (en) * | 2013-12-17 | 2014-04-09 | 南京航空航天大学 | Multiple-input-multiple-output broadband satellite mobile communication channel modeling method based on dual-orthogonal-polarization antenna |
US20150143079A1 (en) * | 2013-11-15 | 2015-05-21 | Qualcomm Incorporated | VECTOR PROCESSING ENGINES (VPEs) EMPLOYING TAPPED-DELAY LINE(S) FOR PROVIDING PRECISION CORRELATION / COVARIANCE VECTOR PROCESSING OPERATIONS WITH REDUCED SAMPLE RE-FETCHING AND POWER CONSUMPTION, AND RELATED VECTOR PROCESSOR SYSTEMS AND METHODS |
CN107346988A (en) * | 2017-06-20 | 2017-11-14 | 大连大学 | A kind of appearance based on low-track satellite network late/hold circuit network route computing method |
CN108337032A (en) * | 2018-01-12 | 2018-07-27 | 西安交通大学 | A method of the latency measurement deviation quantization in SDSN and latency prediction |
CN108966352A (en) * | 2018-07-06 | 2018-12-07 | 北京邮电大学 | Dynamic beam dispatching method based on depth enhancing study |
CN109117856A (en) * | 2018-08-23 | 2019-01-01 | 中国联合网络通信集团有限公司 | People and object method for tracing, apparatus and system based on intelligent edge cloud |
CN110896556A (en) * | 2019-04-04 | 2020-03-20 | 中国电子科技集团公司第五十四研究所 | Time synchronization method and device for post-5G forward transmission network based on deep reinforcement learning |
CN111294108A (en) * | 2020-01-15 | 2020-06-16 | 南京航空航天大学 | Efficient routing method for orthogonal circular orbit configuration satellite constellation |
US20200259896A1 (en) * | 2019-02-13 | 2020-08-13 | Telefonaktiebolaget Lm Ericsson (Publ) | Industrial Automation with 5G and Beyond |
CN112034419A (en) * | 2020-09-08 | 2020-12-04 | 北京邮电大学 | Propagation delay estimation method and device for 5G positioning |
CN112187342A (en) * | 2020-09-30 | 2021-01-05 | 西安交通大学 | Satellite traffic routing method and system based on energy perception and load balancing |
CN112946698A (en) * | 2021-01-29 | 2021-06-11 | 合肥工业大学智能制造技术研究院 | Satellite signal cycle slip detection method based on reinforcement learning |
CN113163387A (en) * | 2021-05-21 | 2021-07-23 | 南通大学 | Emergency communication service sensing method |
CN114040016A (en) * | 2021-11-02 | 2022-02-11 | 清华大学 | Arranging method, system, medium and computing device for intelligent agent at satellite edge |
CN114629543A (en) * | 2022-01-28 | 2022-06-14 | 航天东方红卫星有限公司 | Satellite network adaptive traffic scheduling method based on deep supervised learning |
CN114745254A (en) * | 2022-06-10 | 2022-07-12 | 中国地质大学(武汉) | Time-varying inter-satellite link coloring Petri net modeling method based on function separation |
CN114884557A (en) * | 2022-03-25 | 2022-08-09 | 重庆邮电大学 | Satellite time-sensitive network path selection method based on network calculation |
US20220269258A1 (en) * | 2020-09-15 | 2022-08-25 | Zhejiang University | Method for anomaly classification of industrial control system communication network |
CN115967429A (en) * | 2022-03-12 | 2023-04-14 | 桂林电子科技大学 | Giant constellation-oriented on-satellite autonomous route generation and management method |
WO2023085984A1 (en) * | 2021-11-10 | 2023-05-19 | Telefonaktiebolaget Lm Ericsson (Publ) | Protecting a model against an adversary |
CN116170853A (en) * | 2022-12-29 | 2023-05-26 | 北京理工大学 | QoS and load balancing cooperative guarantee satellite routing algorithm based on machine learning |
-
2023
- 2023-11-20 CN CN202311542525.3A patent/CN117674961B/en active Active
Patent Citations (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2001015001A2 (en) * | 1999-08-23 | 2001-03-01 | General Electric Company | Apparatus and method for managing a fleet of mobile assets |
US20100090889A1 (en) * | 2006-09-29 | 2010-04-15 | Yoola Hwang | Precise orbit determination system and method using gps data and galileo data |
CN101299713A (en) * | 2008-03-21 | 2008-11-05 | 哈尔滨工业大学深圳研究生院 | Method for setting multilayer satellite network system route |
US20150143079A1 (en) * | 2013-11-15 | 2015-05-21 | Qualcomm Incorporated | VECTOR PROCESSING ENGINES (VPEs) EMPLOYING TAPPED-DELAY LINE(S) FOR PROVIDING PRECISION CORRELATION / COVARIANCE VECTOR PROCESSING OPERATIONS WITH REDUCED SAMPLE RE-FETCHING AND POWER CONSUMPTION, AND RELATED VECTOR PROCESSOR SYSTEMS AND METHODS |
CN103716802A (en) * | 2013-12-17 | 2014-04-09 | 南京航空航天大学 | Multiple-input-multiple-output broadband satellite mobile communication channel modeling method based on dual-orthogonal-polarization antenna |
CN107346988A (en) * | 2017-06-20 | 2017-11-14 | 大连大学 | A kind of appearance based on low-track satellite network late/hold circuit network route computing method |
CN108337032A (en) * | 2018-01-12 | 2018-07-27 | 西安交通大学 | A method of the latency measurement deviation quantization in SDSN and latency prediction |
CN108966352A (en) * | 2018-07-06 | 2018-12-07 | 北京邮电大学 | Dynamic beam dispatching method based on depth enhancing study |
CN109117856A (en) * | 2018-08-23 | 2019-01-01 | 中国联合网络通信集团有限公司 | People and object method for tracing, apparatus and system based on intelligent edge cloud |
US20200259896A1 (en) * | 2019-02-13 | 2020-08-13 | Telefonaktiebolaget Lm Ericsson (Publ) | Industrial Automation with 5G and Beyond |
CN110896556A (en) * | 2019-04-04 | 2020-03-20 | 中国电子科技集团公司第五十四研究所 | Time synchronization method and device for post-5G forward transmission network based on deep reinforcement learning |
CN111294108A (en) * | 2020-01-15 | 2020-06-16 | 南京航空航天大学 | Efficient routing method for orthogonal circular orbit configuration satellite constellation |
CN112034419A (en) * | 2020-09-08 | 2020-12-04 | 北京邮电大学 | Propagation delay estimation method and device for 5G positioning |
US20220269258A1 (en) * | 2020-09-15 | 2022-08-25 | Zhejiang University | Method for anomaly classification of industrial control system communication network |
CN112187342A (en) * | 2020-09-30 | 2021-01-05 | 西安交通大学 | Satellite traffic routing method and system based on energy perception and load balancing |
CN112946698A (en) * | 2021-01-29 | 2021-06-11 | 合肥工业大学智能制造技术研究院 | Satellite signal cycle slip detection method based on reinforcement learning |
CN113163387A (en) * | 2021-05-21 | 2021-07-23 | 南通大学 | Emergency communication service sensing method |
CN114040016A (en) * | 2021-11-02 | 2022-02-11 | 清华大学 | Arranging method, system, medium and computing device for intelligent agent at satellite edge |
WO2023085984A1 (en) * | 2021-11-10 | 2023-05-19 | Telefonaktiebolaget Lm Ericsson (Publ) | Protecting a model against an adversary |
CN114629543A (en) * | 2022-01-28 | 2022-06-14 | 航天东方红卫星有限公司 | Satellite network adaptive traffic scheduling method based on deep supervised learning |
CN115967429A (en) * | 2022-03-12 | 2023-04-14 | 桂林电子科技大学 | Giant constellation-oriented on-satellite autonomous route generation and management method |
CN114884557A (en) * | 2022-03-25 | 2022-08-09 | 重庆邮电大学 | Satellite time-sensitive network path selection method based on network calculation |
CN114745254A (en) * | 2022-06-10 | 2022-07-12 | 中国地质大学(武汉) | Time-varying inter-satellite link coloring Petri net modeling method based on function separation |
CN116170853A (en) * | 2022-12-29 | 2023-05-26 | 北京理工大学 | QoS and load balancing cooperative guarantee satellite routing algorithm based on machine learning |
Non-Patent Citations (3)
Title |
---|
刘庆利;刘楠楠;石怀峰;: "基于地理位置多业务LEO卫星网络路由算法", 计算机仿真, no. 09, 15 September 2015 (2015-09-15) * |
左宗;陈明剑;李滢;姚翔;: "极区北斗卫星数据质量特征分析", 测绘通报, no. 06, 25 June 2018 (2018-06-25) * |
田英国;郝金明;于合理;刘伟平;谢建涛;张康;: "LEO卫星单频精密定轨电离层模型改进算法", 测绘学报, no. 07, 15 July 2016 (2016-07-15) * |
Also Published As
Publication number | Publication date |
---|---|
CN117674961B (en) | 2024-05-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Nie et al. | Network traffic prediction based on deep belief network in wireless mesh backbone networks | |
CN110601777B (en) | Method for estimating satellite-ground downlink co-channel interference under low-orbit mobile satellite constellation | |
CN112272381B (en) | Satellite network task deployment method and system | |
CN104463351A (en) | Communication bandwidth prediction method and device based on power business requirements | |
CN112187386B (en) | Network performance evaluation method and system for heaven-earth integrated network | |
CN114760669B (en) | Flow prediction-based route decision method and system | |
CN113469425B (en) | Deep traffic jam prediction method | |
CN114071528A (en) | Service demand prediction-based multi-beam satellite beam resource adaptation method | |
CN101013955A (en) | Fast simulated annealing for traffic matrix estimation | |
CN117395188B (en) | Deep reinforcement learning-based heaven-earth integrated load balancing routing method | |
Fan et al. | Network selection based on evolutionary game and deep reinforcement learning in space-air-ground integrated network | |
CN114528987A (en) | Neural network edge-cloud collaborative computing segmentation deployment method | |
CN117674961B (en) | Low orbit satellite network time delay prediction method based on space-time feature learning | |
Hao et al. | A multi-aspect expanded hypergraph enabled cross-domain resource management in satellite networks | |
CN113365229A (en) | Network time delay optimization method of multi-union chain consensus algorithm | |
CN115499875A (en) | Satellite internet task unloading method and system and readable storage medium | |
CN114615693A (en) | Network capacity prediction method, device, electronic equipment and computer storage medium | |
CN117014355A (en) | TSSDN dynamic route decision method based on DDPG deep reinforcement learning algorithm | |
Volos et al. | Reladec: Reliable latency decision algorithm for connected vehicle applications | |
Du et al. | Traffic prediction based resource configuration in space-based systems | |
CN114567587B (en) | Routing decision method for multi-user access edge computing terminal of power distribution Internet of things | |
CN115276765B (en) | ATDM satellite communication scheduling method for service priority | |
CN114422011B (en) | Low orbit satellite constellation network capacity measuring and calculating method | |
Chen et al. | Queuing theory application in imaging service analysis for small Earth observation satellites | |
Xu et al. | Joint Optimization of Network Topology and Link Capacity Expansion Based on a Greedy-Mutation Genetic Algorithm |
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 | ||
GR01 | Patent grant |