KLI

Performance Analysis and Deep Learning Design of Underlay Cognitive NOMA-Based CDRT Networks With Imperfect SIC and Co-Channel Interference

Metadata Downloads
Abstract
In this paper, we investigate an underlay cognitive non-orthogonal multiple access (NOMA)-based coordinated direct and relay transmission network with imperfect successive interference cancellation, imperfect channel state information, and co-channel interference caused by a multi-antenna primary transmitter. In the secondary network, a source communicates with a near user via direct link and with a far user through the assistance of multiple relays subject to transmit power constraints. Four relay selection schemes are proposed to enhance the performance of NOMA users and the overall system throughput. In our analysis, exact closed-form expressions for the outage probability (OP) of NOMA users and for the overall system throughput are derived. To provide further insights, a performance floor analysis is carried out considering two power-setting scenarios: (i) the transmit powers at the secondary source and relays go to infinity and (ii) the peak interference constraint goes to infinity. Towards real-time configurations, we also design a deep learning (DL) framework for the OP and system throughput prediction. Our results show that the deep neural network exhibits the lowest run-time prediction and root-mean-square error among the proposed DL models. Furthermore, the predicted results based on DL framework match with those of the analysis and simulation.
Author(s)
부 타이 혹Toan-Van NguyenDaniel Benevides da Costa김성환
Issued Date
2021
Type
Article
Keyword
cognitive radio (CR)Coordinated direct and relay transmission (CDRT)Deep learningInterchannel interferenceNOMAnon-orthogonal multiple access (NOMA)outage probability (OP)relay selectionRelaysResource managementSilicon carbidesystem throughputThroughput
DOI
10.1109/TCOMM.2021.3110209
URI
https://oak.ulsan.ac.kr/handle/2021.oak/9101
https://ulsan-primo.hosted.exlibrisgroup.com/primo-explore/fulldisplay?docid=TN_cdi_crossref_primary_10_1109_TCOMM_2021_3110209&context=PC&vid=ULSAN&lang=ko_KR&search_scope=default_scope&adaptor=primo_central_multiple_fe&tab=default_tab&query=any,contains,Performance%20Analysis%20and%20Deep%20Learning%20Design%20of%20Underlay%20Cognitive%20NOMA-Based%20CDRT%20Networks%20With%20Imperfect%20SIC%20and%20Co-Channel%20Interference&offset=0&pcAvailability=true
Publisher
IEEE TRANSACTIONS ON COMMUNICATIONS
Location
미국
Language
영어
ISSN
0090-6778
Citation Volume
69
Citation Number
12
Citation Start Page
8159
Citation End Page
8174
Appears in Collections:
Engineering > IT Convergence
Authorize & License
  • Authorize공개
Files in This Item:
  • There are no files associated with this item.

Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.