Deep reinforcement learning-based beam training with energy and spectral efficiency maximisation for millimetre-wave channels
Abstract The millimetre-wave (mmWave) spectrum has been investigated for the fifth generation wireless system to provide greater bandwidths and faster data rates. The use of mmWave signals allows large-scale antenna arrays to concentrate the radiated power into narrow beams for directional transmiss...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2022-11-01
|
| Series: | EURASIP Journal on Wireless Communications and Networking |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13638-022-02191-7 |
| _version_ | 1828185159217709056 |
|---|---|
| author | Narengerile John Thompson Paul Patras Tharmalingam Ratnarajah |
| author_facet | Narengerile John Thompson Paul Patras Tharmalingam Ratnarajah |
| author_sort | Narengerile |
| collection | DOAJ |
| description | Abstract The millimetre-wave (mmWave) spectrum has been investigated for the fifth generation wireless system to provide greater bandwidths and faster data rates. The use of mmWave signals allows large-scale antenna arrays to concentrate the radiated power into narrow beams for directional transmission. The beam alignment at mmWave frequency bands requires periodic training because mmWave channels are sensitive to user mobility and environmental changes. To benefit from machine learning technologies that will be used to build the sixth generation (6G) communication systems, we propose a new beam training algorithm via deep reinforcement learning. The proposed algorithm can switch between different beam training techniques according to the changes in the wireless channel such that the overall beam training overhead is minimised while achieving good performance of energy efficiency or spectral efficiency. Further, we develop a beam training strategy which can maximise either energy efficiency or spectral efficiency by controlling the number of activated radio frequency chains based on the current channel conditions. Simulation results show that compared to baseline algorithms, the proposed approach can achieve higher energy efficiency or spectral efficiency with lower training overhead. |
| first_indexed | 2024-04-12T07:00:02Z |
| format | Article |
| id | doaj.art-15e1150ab0be4cdcaec17f9909f34814 |
| institution | Directory Open Access Journal |
| issn | 1687-1499 |
| language | English |
| last_indexed | 2024-04-12T07:00:02Z |
| publishDate | 2022-11-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | EURASIP Journal on Wireless Communications and Networking |
| spelling | doaj.art-15e1150ab0be4cdcaec17f9909f348142022-12-22T03:43:02ZengSpringerOpenEURASIP Journal on Wireless Communications and Networking1687-14992022-11-012022113110.1186/s13638-022-02191-7Deep reinforcement learning-based beam training with energy and spectral efficiency maximisation for millimetre-wave channelsNarengerile0John Thompson1Paul Patras2Tharmalingam Ratnarajah3School of Engineering, University of EdinburghSchool of Engineering, University of EdinburghSchool of Informatics, University of EdinburghSchool of Engineering, University of EdinburghAbstract The millimetre-wave (mmWave) spectrum has been investigated for the fifth generation wireless system to provide greater bandwidths and faster data rates. The use of mmWave signals allows large-scale antenna arrays to concentrate the radiated power into narrow beams for directional transmission. The beam alignment at mmWave frequency bands requires periodic training because mmWave channels are sensitive to user mobility and environmental changes. To benefit from machine learning technologies that will be used to build the sixth generation (6G) communication systems, we propose a new beam training algorithm via deep reinforcement learning. The proposed algorithm can switch between different beam training techniques according to the changes in the wireless channel such that the overall beam training overhead is minimised while achieving good performance of energy efficiency or spectral efficiency. Further, we develop a beam training strategy which can maximise either energy efficiency or spectral efficiency by controlling the number of activated radio frequency chains based on the current channel conditions. Simulation results show that compared to baseline algorithms, the proposed approach can achieve higher energy efficiency or spectral efficiency with lower training overhead.https://doi.org/10.1186/s13638-022-02191-76GMillimetre waveBeam trainingDeep reinforcement learningSpectral efficiencyEnergy efficiency |
| spellingShingle | Narengerile John Thompson Paul Patras Tharmalingam Ratnarajah Deep reinforcement learning-based beam training with energy and spectral efficiency maximisation for millimetre-wave channels EURASIP Journal on Wireless Communications and Networking 6G Millimetre wave Beam training Deep reinforcement learning Spectral efficiency Energy efficiency |
| title | Deep reinforcement learning-based beam training with energy and spectral efficiency maximisation for millimetre-wave channels |
| title_full | Deep reinforcement learning-based beam training with energy and spectral efficiency maximisation for millimetre-wave channels |
| title_fullStr | Deep reinforcement learning-based beam training with energy and spectral efficiency maximisation for millimetre-wave channels |
| title_full_unstemmed | Deep reinforcement learning-based beam training with energy and spectral efficiency maximisation for millimetre-wave channels |
| title_short | Deep reinforcement learning-based beam training with energy and spectral efficiency maximisation for millimetre-wave channels |
| title_sort | deep reinforcement learning based beam training with energy and spectral efficiency maximisation for millimetre wave channels |
| topic | 6G Millimetre wave Beam training Deep reinforcement learning Spectral efficiency Energy efficiency |
| url | https://doi.org/10.1186/s13638-022-02191-7 |
| work_keys_str_mv | AT narengerile deepreinforcementlearningbasedbeamtrainingwithenergyandspectralefficiencymaximisationformillimetrewavechannels AT johnthompson deepreinforcementlearningbasedbeamtrainingwithenergyandspectralefficiencymaximisationformillimetrewavechannels AT paulpatras deepreinforcementlearningbasedbeamtrainingwithenergyandspectralefficiencymaximisationformillimetrewavechannels AT tharmalingamratnarajah deepreinforcementlearningbasedbeamtrainingwithenergyandspectralefficiencymaximisationformillimetrewavechannels |