Machine Learning-driven Channel State Information Prediction and Resource Optimization Strategy for Millimeter-wave Communication Channels
Abstract
millimeter-wave communication systems. By analyzing the characteristics of millimeter-wave communication and the limitations of traditional methods, we demonstrate the advantages of machine learning in this field. We provide a detailed exploration of machine learning-based
channel state information prediction methods, including model construction, feature extraction, and optimization strategies. Additionally, we
investigate resource optimization strategies encompassing spectrum allocation, power control, and beam management. Finally, through practical case studies, we evaluate the effectiveness of machine learning-driven approaches and outline future development trends, offering theoretical support and practical guidance for the advancement of millimeter-wave communication technologies.
Keywords
Full Text:
PDFReferences
[1] Jin HW, Zhong WZ, Liu X, et al. Millimeter-wave beam prediction for vehicle-to-everything networks based on machine learning [J].
Mobile Communications, 2024, 48(12):39-45+75.
[2] Tang Rongshun. Research on Millimeter Wave Large-Scale MIMO Beamforming and Cancellation Prediction Based on Machine Learning [D]. Southeast University, 2024.
[3] Wang Shengyi. Research on Beamforming Technology in Non-Cellular Millimeter Wave Systems under Dynamic Scenarios [D]. Southeast University, 2024.
[4] Li Qinqing. Research on Multi-Beam Scheduling Technology Based on Millimeter Wave System Coverage Enhancement [D]. Beijing
University of Posts and Telecommunications, 2022.
[5] Zou ZC. Research on User Scheduling and Beam Selection Scheme for Millimeter Wave Beamforming Systems[D]. Guangzhou University, 2022.
DOI: http://dx.doi.org/10.70711/frim.v4i5.9374
Refbacks
- There are currently no refbacks.
