Cooperative Deep Reinforcement Learning based Grant-Free NOMA Optimization for mURLLC

TY - CONF

T1 - Cooperative Deep Reinforcement Learning based Grant-Free NOMA Optimization for mURLLC

AU - Liu, Yan

AU - Deng, Yansha

AU - Elkashlan, Maged

AU - Nallanathan, Arumugam

N1 - Publisher Copyright: © 2022 IEEE.

PY - 2022

Y1 - 2022

N2 - Grant-free non-orthogonal multiple access (GF-NOMA) is a potential technique to support massive Ultra-Reliable and Low-Latency Communication (mURLLC) service. However, the dynamic resource configuration in GF-NOMA systems is challenging due to the random traffics and collisions, which are unknown at the base station (BS). Meanwhile, joint consideration of the latency and reliability requirements makes the resource configuration of GF-NOMA more complex. To address this problem, we develop a general learning framework for signature-based GF-NOMA in mURLLC service taking into account the multiple access signature collision, the user (UE) detection, as well as the data decoding procedures for the K-repetition GF-NOMA. The goal of our learning framework is to maximize the long-term average number of successfully served UEs under the latency constraint. We propose a Cooperative Multi-Agent Deep Neural Network based Q-learning (CMA-DQN) approach to optimize the configuration of both the repetition values and the contention-transmission unit (CTU) numbers. Our results show the superior performance of CMA-DQN over the LE-URC in heavy traffic and demonstrate its capability in dynamically configuring in long term for mURLLC service.

AB - Grant-free non-orthogonal multiple access (GF-NOMA) is a potential technique to support massive Ultra-Reliable and Low-Latency Communication (mURLLC) service. However, the dynamic resource configuration in GF-NOMA systems is challenging due to the random traffics and collisions, which are unknown at the base station (BS). Meanwhile, joint consideration of the latency and reliability requirements makes the resource configuration of GF-NOMA more complex. To address this problem, we develop a general learning framework for signature-based GF-NOMA in mURLLC service taking into account the multiple access signature collision, the user (UE) detection, as well as the data decoding procedures for the K-repetition GF-NOMA. The goal of our learning framework is to maximize the long-term average number of successfully served UEs under the latency constraint. We propose a Cooperative Multi-Agent Deep Neural Network based Q-learning (CMA-DQN) approach to optimize the configuration of both the repetition values and the contention-transmission unit (CTU) numbers. Our results show the superior performance of CMA-DQN over the LE-URC in heavy traffic and demonstrate its capability in dynamically configuring in long term for mURLLC service.

KW - deep reinforcement learning

KW - grant-free

KW - mURLLC

KW - NOMA

KW - resource configuration

UR - http://www.scopus.com/inward/record.url?scp=85139132231&partnerID=8YFLogxK

U2 - 10.1109/ICC45855.2022.9882276

DO - 10.1109/ICC45855.2022.9882276

M3 - Paper

AN - SCOPUS:85139132231

SP - 1

EP - 6

T2 - 2022 IEEE International Conference on Communications, ICC 2022

Y2 - 16 May 2022 through 20 May 2022

ER -