Machine learning based hybrid precoder with user scheduling technique for maximizing sum rate in downlink MU-MIMO system

Millimetre wave communication and massive MIMO are eminent candidates, to meet the requirements in a 5G network. The Combination of these techniques will impart an innovative and efficient transceiver signal processing techniques and radio resource management. The main idea of this paper is to propo...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:International journal of information technology (Singapore. Online) Ročník 14; číslo 5; s. 2399 - 2405
Hlavní autoři: Rajarajeswarie, B., Sandanalakshmi, R
Médium: Journal Article
Jazyk:angličtina
Vydáno: Singapore Springer Nature Singapore 01.08.2022
Springer Nature B.V
Témata:
5G mobile communication
Algorithms
Artificial Intelligence
Communications systems
Computer Imaging
Computer Science
Correlation coefficients
Downlinking
Electronic systems
Entropy (Information theory)
Image Processing and Computer Vision
Machine Learning
Millimeter waves
MIMO communication
Original Research
Pattern Recognition and Graphics
Radio signals
Resource management
Scheduling
Signal processing
Software Engineering
Vision
SystemVue
User scheduling
Hybrid precoder
Weight assigned cross entropy
Multi-user MIMO
Machine learning
ISSN:2511-2104, 2511-2112
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Millimetre wave communication and massive MIMO are eminent candidates, to meet the requirements in a 5G network. The Combination of these techniques will impart an innovative and efficient transceiver signal processing techniques and radio resource management. The main idea of this paper is to propose a joint framework for user scheduling and hybrid precoder using Machine Learning (ML) for downlink Multi-User Multiple Input and Multiple Output systems (MU-MIMO) to enhance the sum rate. The downlink Multi-User MIMO (MU-MIMO) system model has been implemented in the Keysight’s Electronic System Level (ESL) SystemVue software to generate a channel matrix. Using the simulated channel matrix, correlation among the users were calculated which represents the realistic 5G channel response. A joint framework has been implemented with the Correlation factor-based user scheduling and cross entropy-based hybrid precoder (JUSHP) to maximize the achievable sum rate. The paper also proposes a weight assigned Cross- Entropy based hybrid precoder (JUSWHP) for further improvement in achievable sum rate. In both the framework machine learning algorithm has been applied to choose the best precoder for users in the MU-MIMO system. The simulation results authenticate that the JUSWHP framework achieves better performance with respect to achievable Sum Rate when compared to the other techniques.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2511-2104
2511-2112
DOI:10.1007/s41870-022-00902-3