User Scheduling and Simulated Beam Selection Optimization Method Based on Machine Learning

An analog beam and machine learning technology, applied in the field of wireless communication, can solve the problems of large scheduling scale, NP difficulty, and high computational complexity, and achieve the effects of high compatibility, improved classification accuracy, and high prediction accuracy.

Active Publication Date: 2022-05-24
南方电网互联网服务有限公司
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In addition, there is a low-complexity solution based on the greedy method in the existing beam selection technology, but when the system scale is large, its computational complexity is also high
[0004] Furthermore, due to the duality of the beam allocation matrix, the allocation of beams by the base station to users is a non-convex NP-hard problem, and the global optimization scheme based on exhaustive search has exponential computational complexity, which is unacceptable when the scheduling scale is large of

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • User Scheduling and Simulated Beam Selection Optimization Method Based on Machine Learning
  • User Scheduling and Simulated Beam Selection Optimization Method Based on Machine Learning
  • User Scheduling and Simulated Beam Selection Optimization Method Based on Machine Learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0080] like figure 1 As shown, this embodiment provides a method for optimizing user scheduling and analog beam selection based on machine learning, and the method includes the following steps:

[0081] The user channel feature vector acquisition step: obtain each user channel feature vector from the user set to be processed in turn. The user channel feature vector includes the path loss, the 2L real-valued feature of the L complex path gain, and the L emission angles of the base station, where L is The number of channel channels between the base station and the user;

[0082] The analog beam matching step: sequentially input the channel feature vector of each user to the beam prediction model to determine the analog beam, specifically, using multiple beam classifiers to divide the downlink channel between the base station and the selected user into multiple different beam classes, Use the hyperplane to predict the category of the channel and select the best simulated beam fo...

Embodiment 2

[0142] This embodiment also provides a machine learning-based user scheduling and analog beam selection optimization system, including: a channel feature vector acquisition module, an analog beam matching module, and a channel scheduling module;

[0143] In this embodiment, the channel feature vector obtaining module is used to obtain the channel feature vector of each user;

[0144]In this embodiment, the analog beam matching module is used for analog beam matching, and the channel eigenvectors of each user are sequentially input to the beam prediction model to determine the analog beam, and multiple beam classifiers are used to divide the downlink channels between the base station and the selected user. For multiple different beam classes, use the hyperplane to predict the channel class and select the best analog beam for each user;

[0145] In this embodiment, the beam prediction model is trained by machine learning, and the training data includes a user channel feature vec...

Embodiment 3

[0187] This embodiment uses the optimization scheme of Embodiment 1 to compare the computational complexity. The comparison schemes specifically include a global optimal scheme (ES) based on exhaustive search, a local optimal scheme based on differential convex (D.c.) programming, and State-of-the-art low-complexity scheme based on greedy methods.

[0188] For the machine learning-based user scheduling and simulation beam selection optimization method in Embodiment 1:

[0189] The complexity of user classification is P(T uc KN RF N BS ), where T uc is the maximum number of iterations, T uc in express the base, Represents the set of support vectors for the svm classifier.

[0190] Therefore, the estimated complexity of this scheme is:

[0191]

[0192] Global optimal solution for exhaustive search:

[0193] Compute the feasible set of all user and beam pairs, the computational complexity is:

[0194]

[0195] in The general form is:

[0196] For locall...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a user scheduling and simulation beam selection optimization method based on machine learning. The method includes the following steps: obtaining each user channel feature vector; simulation beam matching: sequentially inputting each user channel feature vector to the beam prediction model Determine the analog beam, use multiple beam classifiers to divide the downlink channel between the base station and the selected user into multiple different beam classes, use the hyperplane to predict the channel class and select the best analog beam for each user; after After matching the analog beams, it is judged whether all users in the user set have been matched. When all users are matched, channel scheduling is performed according to the analog beam scheduling set, where the analog beam scheduling set is the analog beam set corresponding to the output user set. The invention reduces the computing power required by the applicable large-scale system, has high compatibility, reduces the cost of communication system construction, and reduces the time delay of user matching channels in the case of multiple users.

Description

technical field [0001] The present invention relates to the technical field of wireless communication, in particular to a method for optimizing user scheduling and analog beam selection based on machine learning. Background technique [0002] Millimeter-wave (mmWave) communication technology and massive multiple-input multiple-output (MIMO) systems are key technologies in fifth-generation (5G) mobile communication systems to cope with the explosive growth of data services. For massive MIMO-mmWave systems, the traditional all-digital beamforming method is almost inapplicable in practice, because in all-digital beamforming, each antenna is equipped with a radio frequency (RF) chain, and each RF chain occupies a Dedicated baseband processor, so all-digital beamforming makes the system complexity and power consumption unbearable when the number of antennas is large. Hybrid beamforming, which divides beamforming into a low-dimensional digital part and an RF analog part, is a low...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): H04B7/0413H04B7/0456G06K9/62G06N20/00H04B17/00H04B17/391H04W16/28
CPCH04B7/0413H04B17/0087H04B17/3913G06N20/00H04W16/28H04B7/0456G06F18/23213
Inventor 赵赛邹章晨唐冬黄高飞
Owner 南方电网互联网服务有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap