A Deep Reinforcement Learning-Based Resource Scheduler for Massive MIMO Networks

The large number of antennas in massive MIMO systems allows the base station to communicate with multiple users at the same time and frequency resource with multi-user beamforming. However, highly correlated user channels could drastically impede the spectral efficiency that multi-user beamforming c...

Full description

Saved in:
Bibliographic Details
Main Authors: Qing An, Santiago Segarra, Chris Dick, Ashutosh Sabharwal, Rahman Doost-Mohammady
Format: Article
Language:English
Published: IEEE 2023-01-01
Series:IEEE Transactions on Machine Learning in Communications and Networking
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10247079/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850051522402975744
author Qing An
Santiago Segarra
Chris Dick
Ashutosh Sabharwal
Rahman Doost-Mohammady
author_facet Qing An
Santiago Segarra
Chris Dick
Ashutosh Sabharwal
Rahman Doost-Mohammady
author_sort Qing An
collection DOAJ
description The large number of antennas in massive MIMO systems allows the base station to communicate with multiple users at the same time and frequency resource with multi-user beamforming. However, highly correlated user channels could drastically impede the spectral efficiency that multi-user beamforming can achieve. As such, it is critical for the base station to schedule a suitable group of users in each time and frequency resource block to achieve maximum spectral efficiency while adhering to fairness constraints among the users. In this paper, we consider the resource scheduling problem for massive MIMO systems with its optimal solution known to be NP-hard. Inspired by recent achievements in deep reinforcement learning (DRL) to solve problems with large action sets, we propose SMART, a dynamic scheduler for massive MIMO based on the state-of-the-art Soft Actor-Critic (SAC) DRL model and the K-Nearest Neighbors (KNN) algorithm. Through comprehensive simulations using realistic massive MIMO channel models as well as real-world datasets from channel measurement experiments, we demonstrate the effectiveness of our proposed model in various channel conditions. Our results show that our proposed model performs very close to the optimal proportionally fair (Opt-PF) scheduler in terms of spectral efficiency and fairness with more than one order of magnitude lower computational complexity in medium network sizes where Opt-PF is computationally feasible. Our results also show the feasibility and high performance of our proposed scheduler in networks with a large number of users and resource blocks.
format Article
id doaj-art-a28b43502cb64b1da1646f37d6b88e1b
institution DOAJ
issn 2831-316X
language English
publishDate 2023-01-01
publisher IEEE
record_format Article
series IEEE Transactions on Machine Learning in Communications and Networking
spelling doaj-art-a28b43502cb64b1da1646f37d6b88e1b2025-08-20T02:53:07ZengIEEEIEEE Transactions on Machine Learning in Communications and Networking2831-316X2023-01-01124225710.1109/TMLCN.2023.331398810247079A Deep Reinforcement Learning-Based Resource Scheduler for Massive MIMO NetworksQing An0https://orcid.org/0009-0005-9070-9033Santiago Segarra1https://orcid.org/0000-0002-8408-9633Chris Dick2Ashutosh Sabharwal3https://orcid.org/0000-0003-1898-5787Rahman Doost-Mohammady4https://orcid.org/0000-0002-6963-033XDepartment of Electrical and Computer Engineering, Rice University, Houston, TX, USADepartment of Electrical and Computer Engineering, Rice University, Houston, TX, USANVIDIA, Santa Clara, CA, USADepartment of Electrical and Computer Engineering, Rice University, Houston, TX, USADepartment of Electrical and Computer Engineering, Rice University, Houston, TX, USAThe large number of antennas in massive MIMO systems allows the base station to communicate with multiple users at the same time and frequency resource with multi-user beamforming. However, highly correlated user channels could drastically impede the spectral efficiency that multi-user beamforming can achieve. As such, it is critical for the base station to schedule a suitable group of users in each time and frequency resource block to achieve maximum spectral efficiency while adhering to fairness constraints among the users. In this paper, we consider the resource scheduling problem for massive MIMO systems with its optimal solution known to be NP-hard. Inspired by recent achievements in deep reinforcement learning (DRL) to solve problems with large action sets, we propose SMART, a dynamic scheduler for massive MIMO based on the state-of-the-art Soft Actor-Critic (SAC) DRL model and the K-Nearest Neighbors (KNN) algorithm. Through comprehensive simulations using realistic massive MIMO channel models as well as real-world datasets from channel measurement experiments, we demonstrate the effectiveness of our proposed model in various channel conditions. Our results show that our proposed model performs very close to the optimal proportionally fair (Opt-PF) scheduler in terms of spectral efficiency and fairness with more than one order of magnitude lower computational complexity in medium network sizes where Opt-PF is computationally feasible. Our results also show the feasibility and high performance of our proposed scheduler in networks with a large number of users and resource blocks.https://ieeexplore.ieee.org/document/10247079/Massive MIMOresource schedulingdeep reinforcement learning
spellingShingle Qing An
Santiago Segarra
Chris Dick
Ashutosh Sabharwal
Rahman Doost-Mohammady
A Deep Reinforcement Learning-Based Resource Scheduler for Massive MIMO Networks
IEEE Transactions on Machine Learning in Communications and Networking
Massive MIMO
resource scheduling
deep reinforcement learning
title A Deep Reinforcement Learning-Based Resource Scheduler for Massive MIMO Networks
title_full A Deep Reinforcement Learning-Based Resource Scheduler for Massive MIMO Networks
title_fullStr A Deep Reinforcement Learning-Based Resource Scheduler for Massive MIMO Networks
title_full_unstemmed A Deep Reinforcement Learning-Based Resource Scheduler for Massive MIMO Networks
title_short A Deep Reinforcement Learning-Based Resource Scheduler for Massive MIMO Networks
title_sort deep reinforcement learning based resource scheduler for massive mimo networks
topic Massive MIMO
resource scheduling
deep reinforcement learning
url https://ieeexplore.ieee.org/document/10247079/
work_keys_str_mv AT qingan adeepreinforcementlearningbasedresourceschedulerformassivemimonetworks
AT santiagosegarra adeepreinforcementlearningbasedresourceschedulerformassivemimonetworks
AT chrisdick adeepreinforcementlearningbasedresourceschedulerformassivemimonetworks
AT ashutoshsabharwal adeepreinforcementlearningbasedresourceschedulerformassivemimonetworks
AT rahmandoostmohammady adeepreinforcementlearningbasedresourceschedulerformassivemimonetworks
AT qingan deepreinforcementlearningbasedresourceschedulerformassivemimonetworks
AT santiagosegarra deepreinforcementlearningbasedresourceschedulerformassivemimonetworks
AT chrisdick deepreinforcementlearningbasedresourceschedulerformassivemimonetworks
AT ashutoshsabharwal deepreinforcementlearningbasedresourceschedulerformassivemimonetworks
AT rahmandoostmohammady deepreinforcementlearningbasedresourceschedulerformassivemimonetworks