CN110708129A - Wireless channel state information acquisition method - Google Patents
Wireless channel state information acquisition method Download PDFInfo
- Publication number
- CN110708129A CN110708129A CN201910817115.2A CN201910817115A CN110708129A CN 110708129 A CN110708129 A CN 110708129A CN 201910817115 A CN201910817115 A CN 201910817115A CN 110708129 A CN110708129 A CN 110708129A
- Authority
- CN
- China
- Prior art keywords
- prediction
- channel
- algorithm
- channel state
- precision
- 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
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/373—Predicting channel quality or other radio frequency [RF] parameters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/0202—Channel estimation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Power Engineering (AREA)
- Quality & Reliability (AREA)
- Electromagnetism (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
A wireless channel state information acquisition method belongs to the technical field of wireless and mobile communication, and is characterized by comprising the following steps: carrying out uniform format processing on a large number of channel information data samples obtained by channel estimation to form training samples; inputting the sample data into a prediction neural network to obtain a group of optimal parameters aiming at the data set; judging to obtain the actual state of the current prediction precision by the aid of the prediction effect, and training further optimization parameters if the prediction precision does not meet the requirement; the channel state information obtained by predicting the network will be applied to the originating side. The second characteristic is that the proposed Adaptive Structure Extreme Learning Machine (ASELM) algorithm can adaptively adjust the Structure of the network to match the changing characteristics of the channel data.
Description
Technical Field
The invention relates to a wireless channel state information acquisition method, belonging to the technical field of wireless and mobile communication.
Background
Taking an unmanned aerial vehicle communication scene as an example, in recent years, unmanned aerial vehicles are widely applied to many fields of social production and life. In some joint detection tasks, in order to enable the unmanned aerial vehicles to cooperate with each other more effectively, some data need to be shared among the unmanned aerial vehicles, and therefore, the unmanned aerial vehicles have more communication requirements. Channel state information is a key parameter for communication transmission between drones. However, the channel state changes rapidly in the mobile state, and due to the feedback delay in the channel estimation process, the obtained channel state information has an overdue problem, which may cause interruption of communication. Thus, high dynamics pose a great challenge to channel estimation, and channel prediction offers potential possibilities to solve such problems. In order to solve the above problems, researchers put a lot of effort in channel prediction. In a conventional transmission scheme, channel prediction based on historical channel state information obtained by channel estimation has the effect of improving system performance.
Researchers design some channel prediction algorithms, and in a classical Autoregressive (AR) algorithm, channel impulse response is represented by linear combination of historical channel state information, but the AR algorithm cannot adapt to the obvious reduction of channel prediction effect caused by high doppler frequency offset in the unmanned aerial vehicle communication process. In addition, Echo State Networks (ESNs) are also used to predict the state changes of the rayleigh channel. However, from the simulation effect, the performance of the ESN algorithm in the unmanned aerial vehicle communication scene still has difficulty in ensuring the reliability of communication. Different from AR and ESN algorithms, the ASELM algorithm can adaptively adjust the self-prediction network structure in the prediction process to solve the problem that the channel state is fast and changeable in the unmanned aerial vehicle communication process.
Disclosure of Invention
In view of this, the invention considers the structure of the adaptive adjustment prediction neural network, and proposes an ASELM channel prediction algorithm based on an ELM algorithm. The method is characterized in that the algorithm comprises the following processes, and the improved ELM algorithm for acquiring the channel information mainly comprises the following two steps:
training process: setting demand prediction precision, setting the length of a prediction window, obtaining a good prediction effect by an algorithm through setting the prediction window with a certain length and the sliding step length of the prediction window, and adaptively adjusting the internal neural network structure of the algorithm according to the predicted precision demand after training so as to realize combined optimization of the prediction precision and the prediction time, wherein the prediction errors are similar in a certain interval due to the number of hidden neurons meeting the demand; therefore, it can be considered that, in a certain hidden neuron number interval, the prediction result at any hidden neuron number value can be approximately considered as the corresponding prediction effect of all hidden neuron number values of the whole group; therefore, the rapid search of the optimal number of the hidden neurons can be carried out by adopting a grouping search mode (the number values of adjacent hidden neurons in a certain interval are set as a group, and the number value of any hidden neuron in each group is taken as a typical value), and whether the prediction effect of the typical value reaches the required prediction precision or not is judged by comparing NMSE values of test sets corresponding to the typical values of different numbers of hidden neurons; and if the requirement prediction precision is reached, immediately stopping searching new parameters, namely finding the minimum typical value meeting the precision requirement, and the experiment proves that the time required by the training process is less than 1 second.
And (3) prediction process: setting the parameter number of the algorithm as a typical value obtained in the training process, and outputting a channel prediction result of a communication scene between unmanned aerial vehicle platforms; it is verified that the predicted time of each node is generally less than 0.1 second.
According to the method, channel state information data is obtained by performing channel estimation on channel insertion pilot frequency and is used as a training set of the algorithm, and relevant characteristics of the channel state data are extracted. In the actual communication process, the channel prediction and the channel estimation can be effectively combined together by combining the pilot frequency sending mode, so that the control of a sender on the channel state is ensured in the data sending process. Meanwhile, as the historical data of the channel state information is fully utilized, the problem caused by the overdue channel state can be avoided, and the difficulty that the channel state is rapid and changeable caused by the change of the communication environment is effectively overcome.
Drawings
FIG. 1 is a flowchart of the algorithm of the method for acquiring the status information of the wireless channel
Fig. 2 is a scene model diagram of drone communication, showing the situation where the channel of communication between drone platforms is changing as the drone moves.
Fig. 3 is a basic algorithm block diagram of the ELM algorithm.
Fig. 4 is a frame structure diagram of channel prediction. The connection between channel prediction and channel estimation during data transmission is shown. The channel estimation is carried out in a pilot frequency insertion mode, historical channel state information is obtained, the data are used for training a prediction neural network, the trained channel prediction network provides a prediction value of the channel state information in the data transmission process, therefore, the channel state is guaranteed to be known by a sending end when the sending end sends the data, and the problems of packet loss and error code and even communication interruption caused by channel state change in the data transmission process are effectively reduced.
FIG. 5 is a comparison graph of the ASELM algorithm versus the AR and ESN algorithm prediction accuracy MATLAB simulation.
Fig. 6 is a flowchart of the whole working process of the method for acquiring the status information of the wireless channel in the communication process.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the accompanying drawings.
Theoretically, the propagation path and time of radio waves determine channel state information, if a transmitting end and a receiving end are both in a mobile state, a large Doppler frequency offset can be generated, the channel state changes rapidly, the data volume generated by the change of the channel state information is large, the prediction of the data is essentially a mathematical regression problem, when the ELM processes the regression problem, the ELM has the characteristics of strong generalization capability and rapid learning, and a prediction network can extract data characteristics from the channel state data through the learning of the channel state data, so that the data change trend can be rapidly and effectively predicted. Based on the prior knowledge, the ELM algorithm is optimized in a targeted manner, and a wireless channel state information acquisition method is provided.
The following illustrates the application of the present invention in channel prediction of drone platform communication, using a complex sinusoidal channel model to equivalent a wideband wireless channel model, where the drone at the transmitting and receiving ends is in 3D relative motion, and its channel state information can be expressed as
Where t is time, α represents the complex channel gain, fdIndicating the Doppler shift, the change of the Doppler shift depending on the changes of the transmitting end and the receiving end, aRIs a normalized reception matrix navigation vector, aTIs a normalized transmit matrix navigation vector. (theta)R,φR) Is the angle of arrival, (θ)T,φT) Representing the launch angle, (theta, phi) representing the beam pointing direction, and the navigation vector a (theta, phi) may be expressed as
Wherein omegay＝kdysin(θ)sin(φ)，Ωx＝kdxsin (θ) cos (φ), wave number k 2 π/λ,
the kronecker product is shown. N is a radical ofyAnd NxThe numbers of antenna elements in the y-direction and x-direction on a 2 x 2 MIMO planar array are shown, respectively. dx＝dyAnd λ/2, which indicates the spacing of the antenna elements in the y-direction and x-direction, respectively, where the spacing is set to a half wavelength. GantShown is the radiation pattern,
where theta and phi may be expressed as
By inserting pilots, we can obtain the above channel state information h (t). By arranging the data format of the channel state information, a training set and a test set required by algorithm training and testing are obtained.
The ASELM algorithm is an improved algorithm based on the ELM algorithm, wherein the ELM algorithm is a single-layer feedforward neural network (SLFNs), the structure of the algorithm is clear, the hierarchy is clear, and the basic ELM algorithm process is as follows:
suppose a training set is given
An excitation function g (x), the number N of hidden neurons,
step 1: arbitrary input weight wiAnd deviation bi，
Step 2: and calculating a hidden layer output matrix H.
And step 3: calculating the output weight beta:
wherein, the hidden layer neural network output matrix is
The output weight is
The output matrix is
Based on the ELM algorithm, an ASELM algorithm is provided. The method is characterized in that the algorithm comprises the following processes, and the improved ELM algorithm for acquiring the channel information mainly comprises the following two steps:
training process: the method comprises the steps of setting demand prediction precision, setting the length of a prediction window, obtaining a good prediction effect by an algorithm through setting the prediction window with a certain length and the sliding step length of the prediction window, and adaptively adjusting the neural network structure in the algorithm according to the predicted precision demand after training, so that the combined optimization of the prediction precision and the prediction time is realized, and the prediction errors are similar in the number of hidden neurons meeting the demand in a certain interval. Thus, it can be considered that within a certain interval of hidden neuron numbers, the prediction result at any hidden neuron number value can approximate the corresponding prediction effect considered as all the hidden neuron number values of the whole group. Therefore, the fast search of the optimal number of the hidden neurons can be carried out by adopting a grouping search mode (the number values of adjacent hidden neurons in a certain interval are set as a group, and the number value of any hidden neuron in each group is taken as a typical value), and whether the prediction effect of the typical value reaches the required prediction precision or not is judged by comparing NMSE values of test sets corresponding to different typical values of the number of the hidden neurons. And if the requirement prediction precision is reached, immediately stopping searching new parameters, namely finding the minimum typical value meeting the precision requirement, and the experiment proves that the time required by the training process is less than 1 second.
And (3) prediction process: and setting the parameter number of the algorithm as the typical value obtained in the training process, and outputting a channel prediction result of the communication scene between the unmanned aerial vehicle platforms. It is verified that the predicted time of each node is generally less than 0.1 second.
(1) The specific implementation process of the ASELM algorithm is shown in FIG. 1.
(2) Referring to fig. 2, in the drone communication model, two in-flight drones including the transceiving end are equivalent to the end-to-end channel model; the dashed lines with arrows indicate the direction and trajectory of the drone movement and the bar-shaped connection indicates that the drone's beam is aligned to the established channel.
The goal of our design is to ensure efficient prediction of channel conditions during communication between drone platforms. Because the unmanned aerial vehicle is doing high-speed 3D motion, so the channel state change is very fast, it is difficult to effectively predict the channel state information, which not only requires the algorithm to have good generalization ability, but also requires the algorithm to have the characteristic of fast learning, and the excessively complex prediction algorithm cannot adapt to the requirement of the unmanned aerial vehicle communication scene.
(3) Referring to fig. 3, in the method, the ELM is a lightweight single-layer neural network as can be seen in the ELM algorithm block diagram. The ASELM algorithm improved based on the ELM algorithm is also a lightweight algorithm, so that the consumption of a large amount of calculation and training time can be well avoided, calculation resources and calculation time are saved, and a foundation is laid for adapting to the rapid and variable channel states among the unmanned aerial vehicles.
(4) Referring to fig. 4, in the unmanned aerial vehicle communication system, after a communication link is established, the communication system performs channel estimation by inserting a pilot, and feeds back a channel state to a transmitting end after estimating the channel state, but the channel state fed back is out of date due to the characteristic of rapid and variable channel, and therefore the channel state cannot be directly applied to current data transmission, and at this time, the current channel state needs to be predicted according to the estimated historical channel state information. The channel prediction frame structure is shown in fig. 4, where the channel estimation value is a training data set, and the predicted channel state information is used as the basis for the current originating sending signal.
(5) In order to demonstrate the practical performance of various mechanisms in the present invention, the applicant performed multiple simulation experiments. The results of the MATLAB simulation test are shown in fig. 5.
(6) Referring to fig. 6, in the drone communication system, fig. 6 shows a complete wireless channel state information acquisition process.
The above description is only exemplary of the present invention and should not be taken as limiting the invention, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. A wireless channel state information acquisition method comprises the following steps: an adaptive structure overrun learning machine (ASELM) algorithm; the method is characterized in that the number of hidden neurons in an Extreme Learning Machine (ELM) algorithm is optimized, so that the network structure of the ELM can effectively adapt to the change characteristics of channel state data, and a better prediction effect is obtained;
the improved ELM algorithm training and prediction process is as follows:
training process: setting demand prediction precision, setting the length of a prediction window, obtaining a good prediction effect by an algorithm through setting the prediction window with a certain length and the sliding step length of the prediction window, and adaptively adjusting the internal neural network structure of the algorithm according to the predicted precision demand after training so as to realize combined optimization of the prediction precision and the prediction time, wherein the prediction errors are similar in a certain interval due to the number of hidden neurons meeting the demand; therefore, it can be considered that, in a certain hidden neuron number interval, the prediction result at any hidden neuron number value can be approximately considered as the corresponding prediction effect of all hidden neuron number values of the whole group; therefore, the fast search of the optimal number of hidden neurons can be performed by adopting a grouping search mode (the number values of adjacent hidden neurons in a certain interval are set as a group, and the number value of any hidden neuron in each group is taken as a typical value), and whether the prediction effect of the typical value reaches the required prediction precision or not is judged by comparing Normalized Mean Square Error (NMSE) values of test sets corresponding to different typical values of the number of hidden neurons. If the requirement prediction precision is achieved, new parameters are immediately stopped to be searched, namely the minimum typical value meeting the precision requirement is found, and experiments prove that the time required by the training process is less than 1 second;
and (3) prediction process: setting the parameter number of the algorithm as the typical value obtained in the training process, and outputting a channel prediction result; it is verified that the predicted time of each node is generally less than 0.1 second.
2. The method of claim 1, wherein parameters are rapidly filtered and optimized by means of group search, so as to obtain a channel prediction result meeting a required prediction accuracy.
3. The method according to claim 1, wherein the algorithm is based on the idea that the number of hidden neurons in the ELM is adjusted and optimized, so as to adjust the network structure.
4. The method as claimed in claim 1, wherein the channel prediction method is characterized by calibrating the prediction accuracy by retraining.
5. The method of claim 1, wherein the method is used for any one of forward channel prediction, backward channel prediction, precoding, pre-equalization, antenna selection, originating scheduling, and adaptive transmission.
6. The method of claim 2, wherein the method is used for any one of forward channel prediction, backward channel prediction, precoding, pre-equalization, antenna selection, originating scheduling, and adaptive transmission.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910817115.2A CN110708129B (en) | 2019-08-30 | 2019-08-30 | Wireless channel state information acquisition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910817115.2A CN110708129B (en) | 2019-08-30 | 2019-08-30 | Wireless channel state information acquisition method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110708129A true CN110708129A (en) | 2020-01-17 |
| CN110708129B CN110708129B (en) | 2023-01-31 |
Family
ID=69194198
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910817115.2A Active CN110708129B (en) | 2019-08-30 | 2019-08-30 | Wireless channel state information acquisition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110708129B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112134816A (en) * | 2020-09-27 | 2020-12-25 | 杭州电子科技大学 | ELM-LS combined channel estimation method based on intelligent reflection surface |
| CN113285896A (en) * | 2021-04-30 | 2021-08-20 | 南京邮电大学 | Time-varying channel prediction method based on stack type ELM |
| CN113472412A (en) * | 2021-07-13 | 2021-10-01 | 西华大学 | Superposition CSI feedback method based on enhanced ELM |
| WO2022033456A1 (en) * | 2020-08-13 | 2022-02-17 | 华为技术有限公司 | Channel state information measurement feedback method, and related apparatus |
| WO2022184010A1 (en) * | 2021-03-05 | 2022-09-09 | 维沃移动通信有限公司 | Information reporting method and apparatus, first device, and second device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150170514A1 (en) * | 2013-12-13 | 2015-06-18 | Here Global B.V. | Systems and Methods for Detecting Road Congestion and Incidents in Real Time |
| US20160110643A1 (en) * | 2014-10-16 | 2016-04-21 | Nanyang Technological University | Systems and methods for classifying electrical signals |
| CN105761488A (en) * | 2016-03-30 | 2016-07-13 | 湖南大学 | Real-time limit learning machine short-time traffic flow prediction method based on fusion |
| CN107528650A (en) * | 2017-08-30 | 2017-12-29 | 华南师范大学 | A kind of Forecasting Methodology of the cognitive radio networks frequency spectrum based on GCV RBF neurals |
| CN109902259A (en) * | 2019-02-25 | 2019-06-18 | 中国科学院地理科学与资源研究所 | A kind of reconstructing method of the missing space-time data of lightweight |
-
2019
- 2019-08-30 CN CN201910817115.2A patent/CN110708129B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150170514A1 (en) * | 2013-12-13 | 2015-06-18 | Here Global B.V. | Systems and Methods for Detecting Road Congestion and Incidents in Real Time |
| US20160110643A1 (en) * | 2014-10-16 | 2016-04-21 | Nanyang Technological University | Systems and methods for classifying electrical signals |
| CN105761488A (en) * | 2016-03-30 | 2016-07-13 | 湖南大学 | Real-time limit learning machine short-time traffic flow prediction method based on fusion |
| CN107528650A (en) * | 2017-08-30 | 2017-12-29 | 华南师范大学 | A kind of Forecasting Methodology of the cognitive radio networks frequency spectrum based on GCV RBF neurals |
| CN109902259A (en) * | 2019-02-25 | 2019-06-18 | 中国科学院地理科学与资源研究所 | A kind of reconstructing method of the missing space-time data of lightweight |
Non-Patent Citations (3)
| Title |
|---|
| MAHESH GHATE 等: "Design and Fabrication of a Special Purpose Winding Machine for ELM Control Coils of JET", 《IEEE TRANSACTIONS ON APPLIED SUPERCONDUCTIVITY》 * |
| SHIVA NAVABI 等: "Predicting Wireless Channel Features Using Neural Networks", 《2018 IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC)》 * |
| 刘宁 等: "基于最优定界椭球-极限学习机算法自适应软测量建模的应用", 《科学技术与工程》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022033456A1 (en) * | 2020-08-13 | 2022-02-17 | 华为技术有限公司 | Channel state information measurement feedback method, and related apparatus |
| CN112134816A (en) * | 2020-09-27 | 2020-12-25 | 杭州电子科技大学 | ELM-LS combined channel estimation method based on intelligent reflection surface |
| CN112134816B (en) * | 2020-09-27 | 2022-06-10 | 杭州电子科技大学 | ELM-LS combined channel estimation method based on intelligent reflection surface |
| WO2022184010A1 (en) * | 2021-03-05 | 2022-09-09 | 维沃移动通信有限公司 | Information reporting method and apparatus, first device, and second device |
| CN113285896A (en) * | 2021-04-30 | 2021-08-20 | 南京邮电大学 | Time-varying channel prediction method based on stack type ELM |
| CN113472412A (en) * | 2021-07-13 | 2021-10-01 | 西华大学 | Superposition CSI feedback method based on enhanced ELM |
| CN113472412B (en) * | 2021-07-13 | 2023-10-24 | 西华大学 | Enhanced ELM-based superposition CSI feedback method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110708129B (en) | 2023-01-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110708129B (en) | Wireless channel state information acquisition method | |
| CN111954228B (en) | Beam management method, beam management device, electronic equipment and computer readable storage medium | |
| Zhang et al. | Time-division ISAC enabled connected automated vehicles cooperation algorithm design and performance evaluation | |
| Jiang et al. | A comparison of wireless channel predictors: Artificial Intelligence versus Kalman filter | |
| CN110708108B (en) | Wave beam control optimization method of multi-unmanned-aerial-vehicle millimeter wave network | |
| CN111277312B (en) | Fixed subarray space-based millimeter wave beam forming method based on deep complex network | |
| Jiang et al. | Multi-antenna fading channel prediction empowered by artificial intelligence | |
| CN106788631B (en) | Large-scale MIMO reciprocity calibration method based on local calibration | |
| Hussain et al. | Learning and adaptation for millimeter-wave beam tracking and training: A dual timescale variational framework | |
| Evmorfos et al. | Reinforcement learning for motion policies in mobile relaying networks | |
| Dehkordi et al. | Adaptive beam tracking based on recurrent neural networks for mmWave channels | |
| Mai et al. | A UAV air-to-ground channel estimation algorithm based on deep learning | |
| Mohammadi et al. | Location-aware beamforming for MIMO-enabled UAV communications: An unknown input observer approach | |
| Ye et al. | Deep reinforcement learning assisted beam tracking and data transmission for 5g v2x networks | |
| Zheng et al. | Covert federated learning via intelligent reflecting surfaces | |
| Long et al. | Broad learning based hybrid beamforming for mm-wave MIMO in time-varying environments | |
| Gu et al. | Tune: Transfer learning in unseen environments for v2x mmwave beam selection | |
| Yang et al. | Intelligent computation offloading for joint communication and sensing-based vehicular networks | |
| Wang et al. | Multi-agent double deep Q-Learning for beamforming in mmWave MIMO networks | |
| Evmorfos et al. | Deep actor-critic for continuous 3D motion control in mobile relay beamforming networks | |
| Zhang et al. | Training beam sequence design for mmWave tracking systems with and without environmental knowledge | |
| Deng et al. | Interference constrained beam alignment for time-varying channels via kernelized bandits | |
| Hashida et al. | IRS-aided communications without channel state information relying on deep reinforcement learning | |
| CN116760439A (en) | Robust wave beam forming method based on radar communication integrated system | |
| Iqbal et al. | Deep Reinforcement Learning-Based Resource Allocation for Secure RIS-aided UAV Communication |
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 | ||
| GR01 | Patent grant |