2019
DOI: 10.1109/tccn.2019.2949589
|View full text |Cite
|
Sign up to set email alerts
|

Distributed Power Control for Large Energy Harvesting Networks: A Multi-Agent Deep Reinforcement Learning Approach

Abstract: In this paper, we develop a multi-agent reinforcement learning (MARL) framework to obtain online power control policies for a large energy harvesting (EH) multiple access channel, when only causal information about the EH process and wireless channel is available. In the proposed framework, we model the online power control problem as a discrete-time mean-field game (MFG), and analytically show that the MFG has a unique stationary solution. Next, we leverage the fictitious play property of the mean-field games… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
23
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
6
2
1

Relationship

0
9

Authors

Journals

citations
Cited by 48 publications
(23 citation statements)
references
References 36 publications
0
23
0
Order By: Relevance
“…That said, such a combination of environment nonstationarity and partial observability makes learning extremely difficult, which is made even worse if we scale the number of users large as in the upcoming internet of things era. While there have been some recent works [91] that attempt to solve the issue and establish the convergence guarantee of mean-field multi-agent RL, more investigation in this direction is still desired.…”
Section: Multi-agent Consideration In Deep Rlmentioning
confidence: 99%
“…That said, such a combination of environment nonstationarity and partial observability makes learning extremely difficult, which is made even worse if we scale the number of users large as in the upcoming internet of things era. While there have been some recent works [91] that attempt to solve the issue and establish the convergence guarantee of mean-field multi-agent RL, more investigation in this direction is still desired.…”
Section: Multi-agent Consideration In Deep Rlmentioning
confidence: 99%
“…In [115], a multi-agent DRL-based framework was proposed for power control and maximization of throughput in energy-harvesting super IoT systems. Furthermore, a DNN based for distributed online power control is developed to study the policies in the system.…”
Section: Energy Consumption/harvestingmentioning
confidence: 99%
“…In [28], the authors proposed a novel energy management algorithm to maximize the packet rate for point-to-point communication based on the actor-critic RL framework. In [29], a distributed multi-agent RL algorithm is proposed based on an identical reward function for all nodes.…”
Section: A Related Work and Contributionmentioning
confidence: 99%
“…Update θ (i) t and η (i) t using ( 28) and (29) t+1 has an associated potential function R t+1 (the centralized RL algorithm reward function) based on the potential games framework [41], [42]. Therefore, the optimization of the individual reward functions leads to a local optimal of the potential function under certain conditions.…”
mentioning
confidence: 99%