CN114580279A - Low-orbit satellite communication self-adaptive coding method based on LSTM - Google Patents
Low-orbit satellite communication self-adaptive coding method based on LSTM Download PDFInfo
- Publication number
- CN114580279A CN114580279A CN202210199385.3A CN202210199385A CN114580279A CN 114580279 A CN114580279 A CN 114580279A CN 202210199385 A CN202210199385 A CN 202210199385A CN 114580279 A CN114580279 A CN 114580279A
- Authority
- CN
- China
- Prior art keywords
- lstm
- satellite
- information
- weather
- orbit
- 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
- 238000004891 communication Methods 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012549 training Methods 0.000 claims description 14
- 230000003044 adaptive effect Effects 0.000 claims description 12
- 238000012795 verification Methods 0.000 claims description 8
- 238000012937 correction Methods 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 description 10
- 230000015654 memory Effects 0.000 description 6
- 210000002569 neuron Anatomy 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000009499 grossing Methods 0.000 description 4
- 230000004913 activation Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 210000004027 cell Anatomy 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 208000037004 Myoclonic-astatic epilepsy Diseases 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000000874 microwave-assisted extraction Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
-
- 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
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radio Relay Systems (AREA)
Abstract
The invention discloses a low orbit satellite communication self-adaptive coding method based on LSTM, which is characterized by comprising the following steps: step 1, acquiring local real-time weather and longitude and latitude by ground equipment; step 2, generating position information of the satellite according to the satellite orbit information stored in the ground equipment, and expressing the position information by a three-dimensional coordinate; step 3, establishing a prediction SNR model based on LSTM by utilizing real-time weather, longitude and latitude and position information; and 4, selecting an optimal coding scheme according to a table look-up method. The invention has the advantages that: the LSTM is used for predicting the SNR of the channel quality including the weather information, the SNR of the previous moments is considered, and the satellite operation rule is fully utilized; the LSTM can learn the information such as longitude and latitude and further learn the rain attenuation values of different regions, so that a global universal model is obtained.
Description
Technical Field
The invention relates to the technical field of computer networks, in particular to a low orbit satellite communication Adaptive Coding (ACM) method based on a Long and Short Term Memory (LSTM) network.
Background
The low-orbit satellite is a satellite which is dozens to hundreds of kilometers away from the earth surface, and the channel capacity is improved compared with the medium and high orbit satellites due to the fact that the low-orbit satellite is close to the earth surface. The low-orbit satellite can be covered without dead angles, high-quality communication is provided for aerial and ground facilities, and strategic position is very important. Low earth orbit satellite communications face the dual challenges of extreme weather availability and rapid throughput improvement: the low-orbit satellite needs to maintain high availability in disaster relief, polar regions, rain and snow and other scenes, the video content is increased along with the increase of the network bandwidth, the total throughput of the network is further increased, and lower time delay and higher bandwidth are needed when the low-orbit satellite is applied to the fields of unmanned driving, video watching and the like.
Adaptive Coding (ACM) is an important method for improving the utilization rate of satellite frequency bands to improve throughput. In traditional satellite communication, a 6dB margin value is usually provided for rain attenuation, but due to the fact that the thicknesses of cloud layers are different, rain attenuation values are different in all parts of the world, the rain attenuation value exists only in a rainy day, and the rain attenuation value in a sunny day is 0, so that ACM is usually carried out according to the weather in practice to improve throughput. ACM in low earth orbit satellite communication has technical difficulties, such as: the low-orbit satellite has high movement speed and frequent switching and needs rapid real-time prediction; the rain attenuation values of all places in the world are different, and a global model needs to be established; the global weather dynamic changes, and the rapid capture and synchronization are difficult.
In the adaptive coding problem of low-orbit satellite communication, researchers use methods such as antenna Noise temperature identification, linear Regression, exponential Regression, Support Vector Regression (SVR) and the like to predict Signal-to-Noise Ratio (SNR), but none of them utilizes information that a satellite has strong motion regularity. The Long-short-term memory (LSTM) is a recurrent neural network with memorability, and the memory degree of historical information and the adoption degree of new information are controlled by a forgetting gate, an input gate and an output gate, and the identification of passing information of a satellite can be realized through a memory unit.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an LSTM-based low-orbit satellite communication adaptive coding method.
In order to realize the purpose, the technical scheme adopted by the invention is as follows:
an LSTM-based low-orbit satellite communication adaptive coding method comprises the following steps:
and 4, selecting an optimal coding scheme according to a table look-up method.
Further, the specific steps of step 1 are as follows:
step 1.1, the ground equipment acquires the longitude and latitude of the ground equipment;
step 1.2, the ground equipment requests a website to acquire real-time weather according to the longitude and latitude;
it should be noted that, because the single communication time of the low earth orbit satellite and the ground device is very short, usually several minutes, the weather at the communication start time is considered as the weather of the communication, so that only one weather condition is required when the communication starts;
further, the specific steps of step 2 are as follows:
step 2.1, the ground equipment needs to store the orbit information of each satellite and has the capability of finding out the satellite which can communicate in the appointed time; the orbit information is information for describing and predicting the satellite position through calculation, is represented by a form of two lines of orbit elements, adopts two lines of 80-character ASCII codes to store data, and occupies little space;
2.2, converting the position of the satellite into a three-dimensional coordinate from the orbit information;
further, the specific steps of step 3 are as follows:
step 3.1 Using real-time weather, latitude and longitude and location information to compose a sequence
(UElat,UElon,Satlat,SatLon,Satalt,Weather);
Wherein the UElat,UElonRespectively longitude and latitude, Sat of the ground equipmentlat,SatLon,SataltLongitude and latitude and height of the satellite, Weather is local real-time Weather;
step 3.2, inputting the sequence in the step 3.1 into an LSTM network for training, and using the MAE as an error function; MAE is the absolute value of the difference between the true SNR and the predicted SNR; the error function is used for back propagation and updating parameters in the network;
step 3.3, using the trained LSTM to predict and verify;
the verification step is to divide all data sets into a training set, a verification set and a test set in order to prove that the network is not over-fitted, and observe the condition on the verification set during training to determine the over-fitting condition of the model;
further, the specific steps of step 4 are as follows:
selecting a coding scheme according to a predicted SNR lookup table, and selecting a coding scheme with the lowest redundancy Rate, wherein the Packet Error Rate (PER) is less than 0.1;
wherein, the packet error rate is the probability of error code in one frame of data in communication, and normal communication can be realized when the packet error rate is generally considered to be less than 0.1;
the redundancy rate is a ratio of the number of bits used for error correction in forward error correction coding to the total number of bits, that is, the redundancy rate (the number of codes used for error correction/the total number of bits) × 100%.
Compared with the prior art, the invention has the advantages that:
1. the LSTM is used for predicting the SNR of the channel quality including the weather information, the SNR of the previous moments is considered, and the satellite operation rule is fully utilized;
2. the LSTM can learn the information such as longitude and latitude and further learn the rain attenuation values of different regions, so that a global universal model is obtained;
drawings
FIG. 1 is a diagram of an LSTM network architecture in an embodiment of the present invention;
FIG. 2 is a diagram of the structure of an LSTM neuron in an embodiment of the present invention;
FIG. 3 is a diagram of packet error rates for redundancy rates of different coding schemes according to an embodiment of the present invention;
FIG. 4 is a diagram of the MAE accumulation Distribution (CDF) of different methods in different areas of the world according to an embodiment of the present invention;
FIG. 5 is a CDF diagram of Optimal coding scheme matching rate (OMR) for different methods in different regions of the world according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings by way of examples.
As shown in fig. 1, the present embodiment discusses the application of an LSTM-based adaptive encoding method for low-earth satellite communication to satellite-terrestrial communication.
The present embodiment is based on an LSTM network, which is responsible for identifying a two-dimensional location of a User Equipment (UE), a three-dimensional location of a satellite, and a weather condition where the UE is located, and predicting an SNR at a next time according to communication quality at a past time. The embodiment of the low-orbit satellite communication self-adaptive coding method based on the LSTM is applied as follows:
step 1.1, the ground equipment acquires the latitude and longitude of itself, such as the real-time latitude and longitude of UE (UE)lat,UElon) The latitude and longitude information can be obtained by GPS satellite positioning or base station positioning, and if the information is lost, the position of the ground base station interacted with the UE last can be used as a default position;
step 1.2, the ground equipment requests a website to acquire real-time weather according to the longitude and latitude, and after the longitude and latitude information is acquired, the UE can request a weather API to acquire the real-time weather of the place;
the API is an Application Programming Interface, an abbreviation of an Application program Interface, a user sends a request to the API, and the API gives a corresponding reply;
it should be noted that only the local weather needs to be queried each time a link is started, since the connection time for low earth satellites is very short, typically a few minutes. Taking Starlink as an example, the communication diameter is 1100 kilometers, and the longest connection time is 170 seconds, so that querying weather only when the connection is started does not cause excessive burden on the server, and excessive communication resources are not additionally occupied.
step 2.1 the ground equipment needs to store the orbit information of each satellite and has the ability to find out the satellites that can communicate within a specified time:
the information of the satellite orbit is stored in two lines of orbit information, the storage mode is global and saves storage space, and the storage space only needs 800KB space even if the orbit information of all the satellites in orbit at present is stored.
TABLE 1 satellite CALSPHERE 1 orbital information
Table 1 shows orbital parameter information for satellite CALSPHERE 1 transmitted on day 1, month 4, 2022:
the first row and the first column represent the NASA unique number thereof, and U represents a non-confidential satellite;
the first row and the second column represent the international number of the aircraft;
the first row and the third column are emission times, such as 22 means 2022, and 004.58268302 is its emission on day 4.58268302 of 2022;
the second column of the second row is an orbital inclination;
the second row and the third column are ascending node longitudes;
the second row and the fourth column are track eccentricity;
the fifth column in the second row is the angular distance of the near point;
the sixth row and the sixth column are parallel near point angles;
the second row and the seventh column are horizontal movement rates;
the satellite and its specific position within the communication distance at the current time can be calculated by the above orbit information.
2.2, converting the position of the satellite into a three-dimensional coordinate from the orbit information, converting the orbit information into the three-dimensional coordinate information by adopting methods such as Lagrange polynomial interpolation, Newton polynomial interpolation, Nervier successive linear interpolation and the like, and acquiring the position of the satellite at different moments by using satellite simulation software STK (System Toolkit, STK) in a simulation scene;
step 3.1 composing a sequence using the information in step 1, step 2
Combining the information obtained in the step 1 and the step 2, representing weather information in an One-Hot coding mode, practically using (Sun, cloud, Rainy) to represent weather, wherein the weather is coded as (1, 0, 0) in Sunny days and (0, 1, 0) in Cloudy days, considering that the weather is Rainy weather and snowy weather, uniformly considering the weather as rainfall weather, and the code is (0, 0, 1);
thus far, the octave of the input LSTM network is constructed
xt=(UElat,UElon,Satlat,SatLon,SataltSunny, Cloudy, Rainy), wherein the UElat,UElonRespectively longitude and latitude, Sat of the ground equipmentlat,SatLon,SataltLongitude, latitude and altitude of the satellite, (Sunny, Cloudy, Rainy) represents the local weather;
during simulation, each moment, namely every 1 second, has a piece of corresponding information.
Step 3.2, inputting the sequence in the last step into an LSTM network for training, and using the MAE as an error function;
the structure of LSTM network is shown in fig. 1, and an Embedding Layer (Embedding Layer) is introduced to process weather and location information respectively. There are 3 neurons responsible for processing One-Hot encoded weather information and 5 neurons responsible for processing location information. The embedded layer is followed by 1 LSTM layer, which contains 100 neurons and is responsible for memorizing past information; the back of the LSTM layer is a full connection layer containing 100 neurons and is responsible for integrating information of the L STM layer; finally, the predicted SNR is output for an output layer containing a neuron.
The LSTM network solves the problem of long-term dependence by means of three gate functions, and can remember long-term information by means of the gate functions, wherein the three gate functions are a forgetting gate, an input gate and an output gate respectively, and are shown in figure 2;
(1) forget the door: when the network receives new input, if the network needs to forget old information, the network can be completed by forgetting the door. The forgetting gate is an important component of the LSTM network, can control which information needs to be reserved and which information needs to be forgotten, and avoids the problems of gradient explosion and gradient disappearance caused when the gradient reversely propagates along with the time. Forget gate to determine network state C of LSTM network from last momentt-1In which information is forgotten. Forgetting gate to read last time output value ht-1And at this time, the input value xtThen mapped to a value between 0 and 1 by the sigmoid activation function, and finally the value is compared with the network state Ct-1Multiply to determine Ct-1What information should be discarded. When the forgetting gate value is 1, the network state C is completely reservedt-1When the value is 0, it means that the last network state C is completely discardedt-1The information of (1). The expression of the forget gate is:
ft=σ(Wf[ht-1,xt]+bf),
(2) an input gate: it is determined which part of the new input information is kept in the network state. The input gate is used for controlling the current input data x of the networktHow much of the input information x flows into the memory celltCan be saved to the current network state CtIn (1). The input gate comprises two parts, a first part: control signal i between 0 and 1, consisting of a sigmoid activation function, generatedtFor controlling the network state at that timeThe degree of input; the second part is the candidate network state at the current time generated by the tanh layerThis value will be represented by itThe degree of addition to the network state is decided. The state expression for the input gate is:
it=σ(Wi[ht-1,xt]+bi),
after the states of the forget gate and the input gate are obtained, the state of the network can be updated:
wherein C istIs the finally determined network state at the moment.
(3) An output gate: the output value is based on the network state at the current moment, but the filtering and screening processes are carried out. The output gate also includes two-part operation: the first part is a control signal o between 0 and 1 generated by a sigmoid activation function compositiont(ii) a The second part is output information tanh (C) generated after the tanh function is carried outt) And a control signal otMultiplying to obtain final output value h at the momentt. Output gate control memory cell CtFor the current output value htI.e. which parts of the memory cell are to be output at this moment.
In the process of error back propagation, MAE (mean Absolute error) is used as an error metric, because MAE tends to continue to reduce the error when the error is small, as compared to MSE (mean Squared error), which appears to be gradiently vanished due to its Squared nature.
The time-sharing weather information of all parts of the world can be obtained from the network during training,
step 3.3, the trained LSTM is used for prediction and verification;
since neural networks have a tendency to overfit, the whole data set was as follows 7: 2: the proportion of 1 is divided into a training set, a verification set and a test set, and the data of the three data sets are not overlapped and are disordered. The validation set was predicted every 10 epochs during training to see if the network was overfitting.
And sampling the test set after training is finished to obtain a final training result.
The learning rate during training is selected to be 0.001, the batch size is set to 128, and the associated time is set to 5.
After training, the MAEs for the different methods were compared as in table 2:
TABLE 2 MAE of different methods
Method | Linear smoothing | Exponential smoothing | SVR | LSTM |
MAE | 0.114 | 0.062 | 0.032 | 0.017 |
It can be seen that the error of the LSTM from the true value is the smallest of these.
In fig. 3, the horizontal axis represents MAE and the vertical axis represents the Cumulative Distribution Function (CDF) of different locations around the world. It can be seen from the figure that LSTM performs better than other methods at different locations around the world.
And 4, selecting an optimal coding scheme according to a table look-up method.
According to the convention of the Space advisory system council (Committee for Space Data Systems, CCSDS), the communication code of the Low earth orbit satellite is AR4JA code (AR 4JA) in the pseudorandom Low Density Check code (Quas i-cyclic Low Density Parity Check, QC-LDPC). CCSDS uses the AR4JA code and specifies 3 different data rates, 50%, 66%, 80%, respectively. As shown in fig. 4, at a lower signal-to-noise ratio, the communication environment is poor, and a coding scheme with a lower data rate and a higher redundancy rate should be selected to ensure the communication quality; when the communication quality is better, a coding scheme that can transmit more data should be selected to improve the overall throughput.
In practice, the coding scheme with the lowest redundancy rate, with a packet error rate PER of less than 0.1, is selected as the optimal coding scheme, which is also often used in practical low-earth-orbit satellite communication scenarios. And the overall performance of the adaptive coding system is expressed by an Optimal Matching Rate (OMR). The OMR is the proportion of the optimal coding scheme number to the total test number under different prediction means.
TABLE 3 OMR of different methods
Method | Linear smoothing | Exponential smoothing | SVR | LSTM |
OMR | 97.9% | 98.8% | 99.5% | 99.8% |
It can be seen in fig. 5 that the OMR of LSTM is highest compared to other methods, and that the performance variance is smaller at different locations around the world than other methods.
It will be appreciated by those of ordinary skill in the art that the examples described herein are intended to assist the reader in understanding the manner in which the invention is practiced, and it is to be understood that the scope of the invention is not limited to such specifically recited statements and examples. Those skilled in the art can make various other specific changes and combinations based on the teachings of the present invention without departing from the spirit of the invention, and these changes and combinations are within the scope of the invention.
Claims (5)
1. An LSTM-based low-orbit satellite communication adaptive coding method is characterized by comprising the following steps:
step 1, acquiring local real-time weather and longitude and latitude by ground equipment;
step 2, generating position information of the satellite according to the satellite orbit information stored in the ground equipment, and expressing the position information by a three-dimensional coordinate;
step 3, establishing a prediction SNR model based on LSTM by utilizing real-time weather, longitude and latitude and position information;
and 4, selecting an optimal coding scheme according to a table look-up method.
2. The LSTM-based low-earth-orbit satellite communication adaptive coding method according to claim 1, wherein: the specific steps of step 1 are as follows:
step 1.1, the ground equipment acquires the longitude and latitude of the ground equipment;
step 1.2, the ground equipment requests a website to acquire real-time weather according to the longitude and latitude;
it should be noted that, since the time for a single communication between the low earth orbit satellite and the ground device is very short, usually several minutes, the weather at the communication start time is considered as the weather of the communication, and therefore, only one weather condition needs to be requested when the communication starts.
3. The LSTM-based low-earth-orbit satellite communication adaptive coding method according to claim 1, wherein: the specific steps of step 2 are as follows:
step 2.1, the ground equipment needs to store the orbit information of each satellite and has the capability of finding out the satellite which can communicate in the appointed time; the orbit information is information for describing and predicting the satellite position through calculation, is represented by a form of two lines of orbit elements, adopts two lines of 80-character ASCII codes to store data, and occupies little space;
and 2.2, converting the position of the satellite into a three-dimensional coordinate from the orbit information.
4. The LSTM-based low-earth-orbit satellite communication adaptive coding method according to claim 1, wherein: the specific steps of step 3 are as follows:
step 3.1 Using real-time weather, latitude and longitude and location information to compose a sequence
(UElat,UElon,Satlat,SatLon,Satalt,Weather);
Wherein the UElat,UElonRespectively longitude and latitude, Sat of the ground equipmentlat,SatLon,SataltLongitude and latitude and height of the satellite, Weather is local real-time Weather;
step 3.2, inputting the sequence in the step 3.1 into an LSTM network for training, and using the MAE as an error function; MAE is the absolute value of the difference between the true SNR and the predicted SNR; the error function is used for back propagation and updating parameters in the network;
step 3.3, using the trained LSTM to predict and verify;
the verification step is to prove that no overfitting exists in the network, divide all data sets into a training set, a verification set and a test set, and observe the conditions on the verification set during training to determine the overfitting conditions of the model.
5. The LSTM-based low-earth-orbit satellite communication adaptive coding method according to claim 1, wherein: the specific steps of step 4 are as follows:
selecting a coding scheme according to the predicted SNR lookup table, and selecting the coding scheme with the lowest redundancy rate and the packet error rate of less than 0.1;
wherein, the packet error rate is the probability of error code in one frame of data in communication, and normal communication can be realized when the packet error rate is generally considered to be less than 0.1;
the redundancy rate is a ratio of the number of bits used for error correction in forward error correction coding to the total number of bits, that is, the redundancy rate (the number of codes used for error correction/the total number of bits) × 100%.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210199385.3A CN114580279B (en) | 2022-03-02 | Low-orbit satellite communication self-adaptive coding method based on LSTM |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210199385.3A CN114580279B (en) | 2022-03-02 | Low-orbit satellite communication self-adaptive coding method based on LSTM |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114580279A true CN114580279A (en) | 2022-06-03 |
CN114580279B CN114580279B (en) | 2024-05-31 |
Family
ID=
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140266867A1 (en) * | 2013-03-13 | 2014-09-18 | Northrop Grumman Systems Corporation | Adaptive coded modulation in low earth orbit satellite communication system |
WO2018194456A1 (en) * | 2017-04-20 | 2018-10-25 | Universiteit Van Amsterdam | Optical music recognition omr : converting sheet music to a digital format |
CN111259724A (en) * | 2018-11-30 | 2020-06-09 | 塔塔顾问服务有限公司 | Method and system for extracting relevant information from image and computer program product |
CN112364119A (en) * | 2020-12-01 | 2021-02-12 | 国家海洋信息中心 | Ocean buoy track prediction method based on LSTM coding and decoding model |
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140266867A1 (en) * | 2013-03-13 | 2014-09-18 | Northrop Grumman Systems Corporation | Adaptive coded modulation in low earth orbit satellite communication system |
WO2018194456A1 (en) * | 2017-04-20 | 2018-10-25 | Universiteit Van Amsterdam | Optical music recognition omr : converting sheet music to a digital format |
CN111259724A (en) * | 2018-11-30 | 2020-06-09 | 塔塔顾问服务有限公司 | Method and system for extracting relevant information from image and computer program product |
CN112364119A (en) * | 2020-12-01 | 2021-02-12 | 国家海洋信息中心 | Ocean buoy track prediction method based on LSTM coding and decoding model |
Non-Patent Citations (2)
Title |
---|
SHIQI ZHANG等: "LSTM-Based Adaptive Modulation and Coding for Satellite-to-Ground Communications", 《JOURNAL OF BEIJING INSTITUTE OF TECHNOLOGY》, vol. 31, no. 5, 31 October 2022 (2022-10-31), pages 473 - 482 * |
胡晓月等: "基于LSTM的LEO卫星链路自适应算法", 《系统工程与电子技术》, vol. 43, no. 01, 7 August 2020 (2020-08-07), pages 237 - 243 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114362810B (en) | Low orbit satellite beam jump optimization method based on migration depth reinforcement learning | |
EP4078899A1 (en) | Systems and methods for enhanced feedback for cascaded federated machine learning | |
CN112926789B (en) | Satellite cloud image prediction method, prediction device and readable storage medium | |
Han et al. | A prediction-based resource matching scheme for rentable LEO satellite communication network | |
CN112632311A (en) | Cloud layer change trend prediction method based on deep learning | |
Peng et al. | A predictive on-demand placement of UAV base stations using echo state network | |
CN113259950B (en) | Low-orbit satellite spot beam closing method based on service prediction | |
CN102970510B (en) | Method for transmitting human face video | |
CN113613301A (en) | Air-space-ground integrated network intelligent switching method based on DQN | |
CN113919483A (en) | Method and system for constructing and positioning radio map in wireless communication network | |
CN110555378A (en) | Live video-based weather prediction method and system and weather prediction device | |
Aparna et al. | Steering angle prediction for autonomous driving using federated learning: The impact of vehicle-to-everything communication | |
CN115278698A (en) | Unmanned aerial vehicle base station dynamic deployment method and device based on dynamic user distribution prediction | |
CN112261609B (en) | Millimeter wave internet of vehicles resource allocation method and device, electronic equipment and storage medium | |
CN114580279A (en) | Low-orbit satellite communication self-adaptive coding method based on LSTM | |
CN114580279B (en) | Low-orbit satellite communication self-adaptive coding method based on LSTM | |
CN117579126A (en) | Satellite mobile edge calculation unloading decision method based on deep reinforcement learning | |
CN114758282B (en) | Video prediction method based on time sequence correction convolution | |
CN116360883A (en) | Combined optimization method for unloading of Internet of vehicles computing tasks | |
US11901958B2 (en) | Quantum satellite-based global networks | |
CN113919188A (en) | Relay unmanned aerial vehicle path planning method based on context-MAB | |
CN115276771B (en) | Non-stationary orbit satellite gateway station service range simulation method and device | |
CN115361049B (en) | Routing method and device based on direction constraint and electronic equipment | |
US20230334680A1 (en) | Average depth estimation with residual fine-tuning | |
Tian et al. | Emergency Caching: Coded Caching-based Reliable Map Transmission in Emergency Networks |
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 |