Abstract
In this paper, we investigate a deep learning based resource allocation scheme for massive multiple-input-multiple-output (MIMO) communication systems, where a base station (BS) with a large scale antenna array communicates with a user equipment (UE) using beamforming. In particular, we propose Deep Scanning, in which a near-optimal beamforming vector can be found based on deep Q-learning. Through simulations, we confirm that the optimal beam vector can be found with a high probability. We also show that the complexity required to find the optimum beam vector can be reduced significantly in comparison with conventional beam search schemes.
Highlights
In the generation wireless communication systems, massive multiple-input-multiple-output (MIMO), in which a large number of antennas are utilized, is considered one of the key technologies that will be used to increase system capacity [1]
We show that the Deep Scanning proposed here can significantly reduce the time and complexity involved in finding the optimal serving beam for user equipment (UE) through simulations
A novel beam searching scheme based on deep learning has been proposed
Summary
In the generation wireless communication systems, massive multiple-input-multiple-output (MIMO), in which a large number of antennas are utilized, is considered one of the key technologies that will be used to increase system capacity [1]. Beam selection algorithm based on deep learning with the aid of scene data was proposed in Reference [9] They used additional information of scene data which is composed of the geographical data, for example, position and length of cars, to increase accuracy of beam selection such that the scheme is only feasible if the scene data is available, whereas only the instantaneous channel data is used in our proposed scheme. We propose a deep reinforcement learning architecture, thereby producing a novel beamforming vector selection scheme in a massive MIMO system which can determine the proper beamforming vector using a single pilot signal.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
