Online learning-based beam and blockage prediction for indoor millimeter-wave communications

다국어 초록 (Multilingual Abstract)

As the majority of data traffic is generated in indoor environments, millimeter-wave (mm-wave) communications are essential. However, owing to their high directivity and high penetration loss, indoor mm-wave communication is vulnerable to blockages caused by users’ bodies and ambient obstacles. In this study, we investigate an online learning-based method that achieves efficient beam and blockage prediction for indoor mm-wave. The proposed method takes advantage of the fact that the optimal beam index and blockage status depend on the user’s position and corresponding data traffic demand. Simulation results based on 3GPP’s new radio channel and blockage models revealed that the proposed scheme could predict mm-wave blockages with an accuracy exceeding 90%. These results confirmed the viability of the proposed deep neural network (DNN) model for predicting optimal mm-wave beam and spectral efficiencies.

참고문헌 (Reference)

1 "Technical specification group radio access network; study on channel model for frequencies from 0.5 to 100 GHz"

2 Y. Lu, "Positioningaided 3D beamforming for enhanced communications in mmwave mobile networks" 8 : 55513-55525, 2020

3 Qualcomm, "Mobile Mmwave Is Here — and Indoor Deployment Opportunities Abound"

4 R. MacCartney, "Millimeter-wave human blockage at 73 GHz with a simple double knife-edge diffraction model and extension for directional antennas" 2017

5 M. Marcus, "Millimeter wave propagation: spectrum management implications" 6 (6): 54-62, 2005

6 M. Xiao, "Millimeter wave communications for future mobile networks" 35 (35): 1909-1935, 2017

7 Muhammad Alrabeiah, "Millimeter wave base stations with cameras: Vision aided beam and block-age prediction" 2020

8 A. Alkhateeb, "Machine learning for reliable mmwave systems: Blockage prediction and proactive handoff" 1055-1059, 2018

9 V. Va, "Inverse multipath fingerprinting for millimeter wave V2I beam alignment" 67 (67): 4042-4058, 2018

10 T. Shuang, "Design and evaluation of LTE-advanced double codebook" 2011

11 J. Schmidhuber, "Deep learning in neural networks: An overview" 61 : 85-117, 2015

12 Muhammad Alrabeiah, "Deep learning for mmwav beam and blockage prediction using sub-6 GHz channels"

13 W. Pedrycz, "Deep Learning: Concepts and Architectures" Springer Nature 2019

14 I. Goodfellow, "Deep Learning" MIT Press 2016

15 Cisco, "Cisco vni forecast and trends"

16 M. Gapeyenko, "Analysis of human-body blockage in urban millimeter-wave cellular communications" 2016