Reference : BLER-based Adaptive Q-learning for Efficient Random Access in NOMA-based mMTC Networks
Scientific congresses, symposiums and conference proceedings : Paper published in a book
Engineering, computing & technology : Electrical & electronics engineering
Security, Reliability and Trust
http://hdl.handle.net/10993/46628
BLER-based Adaptive Q-learning for Efficient Random Access in NOMA-based mMTC Networks
English
Tran, Duc Dung mailto [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SigCom >]
Sharma, Shree Krishna mailto [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SigCom >]
Chatzinotas, Symeon mailto [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SigCom >]
Apr-2021
Proceedings of 2021 IEEE 93rd Vehicular Technology Conference (VTC2021-Spring)
1-5
Yes
No
International
2021 IEEE 93rd Vehicular Technology Conference (VTC2021-Spring)
from 25-04-2021 to 28-04-2021
IEEE
Helsinki
Finland
[en] BLER ; MTC ; NOMA ; Q-Learning ; short-packet communications
[en] The ever-increasing number of machine-type communications (MTC) devices and the limited available radio resources are leading to a crucial issue of radio access network (RAN) congestion in upcoming 5G and beyond wireless networks. Thus, it is crucial to investigate novel techniques to minimize RAN congestion in massive MTC (mMTC) networks while taking the underlying short-packet communications (SPC) into account. In this paper, we propose an adaptive Q-learning (AQL) algorithm based on block error rate (BLER), an important metric in SPC, for a non-orthogonal multiple access (NOMA) based mMTC system. The proposed method aims to efficiently accommodate MTC devices to the available random access (RA) slots in order to significantly reduce the possible collisions, and subsequently to enhance the system throughput. Furthermore, in order to obtain more practical insights on the system design, the scenario of imperfect successive interference cancellation (ISIC) is considered as compared to the widely-used perfect SIC assumption. The performance of the proposed AQL method is compared with the recent Q-learning solutions in the literature in terms of system throughput over a range of parameters such as the number of devices, blocklength, and residual interference caused by ISIC, along with its convergence evaluation. Our simulation results illustrate the superiority of the proposed method over the existing techniques, in the scenarios where the number of devices is higher than the number of available RA time-slots.
Interdisciplinary Centre for Security, Reliability and Trust (SnT) > SIGCOM
FNR and ERC
FNR-funded project CORE 5G-Sky (Grant C19/IS/13713801)
Researchers ; Professionals ; Students ; General public ; Others
http://hdl.handle.net/10993/46628
FnR ; FNR13713801 > Bjorn Ottersten > 5G-Sky > Interconnecting The Sky In 5g And Beyond - A Joint Communication And Control Approach > 01/06/2020 > 31/05/2023 > 2019

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
FINAL VERSION.pdfAuthor preprint271.72 kBView/Open

Bookmark and Share SFX Query

All documents in ORBilu are protected by a user license.