Reinforcement Learning for NOMA-ALOHA under Fading

Youngwook Ko; Jinho Choi

doi:10.1109/TCOMM.2022.3198125

Reinforcement Learning for NOMA-ALOHA under Fading

Youngwook Ko, Jinho Choi

Electronic Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

We consider a non-orthogonal multiple access in a random-access ALOHA system, in which each user randomly accesses one out of different time slots and send uplink packets based on power differences. In the context of an asymmetric game, we propose a non-orthogonal multiple access ALOHA system based on multi-agent reinforcement learning tools that can help each user to find its best strategies of improving the rates of successful action choices. While taking into account not only collisions, but also fading, we analyze the mean rewards of actions under general settings and focus on the case that involves two different groups of users. To characterize the behaviors of accessing strategies, we apply multi-agent action value methods that consider either greedy or non-greedy actions, combined with an acceleration gradient descent. Our results show that in the proposed system, users employing the greedy action-based methods can be randomly divided into two groups of users and increase the rates of successful action choices. Interestingly, in relatively limited channels, such greedy methods turn many of users to be with a state of barring-access. In this case, the proposed acceleration, non-greedy action methods are shown to reduce such unfairness, at a loss of successful action rates.

Original language	English
Journal	IEEE Transactions on Communications
DOIs	https://doi.org/10.1109/TCOMM.2022.3198125
Publication status	Published - 11 Aug 2022

Access to Document

10.1109/TCOMM.2022.3198125

Final_versionAccepted author manuscript, 3.12 MB

Cite this

@article{06a2833fd6a14b8b9a5a077c2041eea9,

title = "Reinforcement Learning for NOMA-ALOHA under Fading",

abstract = "We consider a non-orthogonal multiple access in a random-access ALOHA system, in which each user randomly accesses one out of different time slots and send uplink packets based on power differences. In the context of an asymmetric game, we propose a non-orthogonal multiple access ALOHA system based on multi-agent reinforcement learning tools that can help each user to find its best strategies of improving the rates of successful action choices. While taking into account not only collisions, but also fading, we analyze the mean rewards of actions under general settings and focus on the case that involves two different groups of users. To characterize the behaviors of accessing strategies, we apply multi-agent action value methods that consider either greedy or non-greedy actions, combined with an acceleration gradient descent. Our results show that in the proposed system, users employing the greedy action-based methods can be randomly divided into two groups of users and increase the rates of successful action choices. Interestingly, in relatively limited channels, such greedy methods turn many of users to be with a state of barring-access. In this case, the proposed acceleration, non-greedy action methods are shown to reduce such unfairness, at a loss of successful action rates.",

author = "Youngwook Ko and Jinho Choi",

year = "2022",

month = aug,

day = "11",

doi = "10.1109/TCOMM.2022.3198125",

language = "English",

journal = "IEEE Transactions on Communications",

issn = "0090-6778",

publisher = "IEEE",

}

TY - JOUR

T1 - Reinforcement Learning for NOMA-ALOHA under Fading

AU - Ko, Youngwook

AU - Choi, Jinho

PY - 2022/8/11

Y1 - 2022/8/11

N2 - We consider a non-orthogonal multiple access in a random-access ALOHA system, in which each user randomly accesses one out of different time slots and send uplink packets based on power differences. In the context of an asymmetric game, we propose a non-orthogonal multiple access ALOHA system based on multi-agent reinforcement learning tools that can help each user to find its best strategies of improving the rates of successful action choices. While taking into account not only collisions, but also fading, we analyze the mean rewards of actions under general settings and focus on the case that involves two different groups of users. To characterize the behaviors of accessing strategies, we apply multi-agent action value methods that consider either greedy or non-greedy actions, combined with an acceleration gradient descent. Our results show that in the proposed system, users employing the greedy action-based methods can be randomly divided into two groups of users and increase the rates of successful action choices. Interestingly, in relatively limited channels, such greedy methods turn many of users to be with a state of barring-access. In this case, the proposed acceleration, non-greedy action methods are shown to reduce such unfairness, at a loss of successful action rates.

AB - We consider a non-orthogonal multiple access in a random-access ALOHA system, in which each user randomly accesses one out of different time slots and send uplink packets based on power differences. In the context of an asymmetric game, we propose a non-orthogonal multiple access ALOHA system based on multi-agent reinforcement learning tools that can help each user to find its best strategies of improving the rates of successful action choices. While taking into account not only collisions, but also fading, we analyze the mean rewards of actions under general settings and focus on the case that involves two different groups of users. To characterize the behaviors of accessing strategies, we apply multi-agent action value methods that consider either greedy or non-greedy actions, combined with an acceleration gradient descent. Our results show that in the proposed system, users employing the greedy action-based methods can be randomly divided into two groups of users and increase the rates of successful action choices. Interestingly, in relatively limited channels, such greedy methods turn many of users to be with a state of barring-access. In this case, the proposed acceleration, non-greedy action methods are shown to reduce such unfairness, at a loss of successful action rates.

U2 - 10.1109/TCOMM.2022.3198125

DO - 10.1109/TCOMM.2022.3198125

M3 - Article

SN - 0090-6778

JO - IEEE Transactions on Communications

JF - IEEE Transactions on Communications

ER -