TY - JOUR
T1 - Joint user pairing and power control for C-NOMA with full-duplex device-to-device relaying
AU - DInh, Phuc
AU - Arfaoui, Mohamed Amine
AU - Sharafeddine, Sanaa
AU - Assi, Chadi
AU - Ghrayeb, Ali
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019
Y1 - 2019
N2 - This paper investigates the performance of coop- erative non-orthogonal multiple access (C-NOMA) in cellulardownlink systems. The system model consists of a base station(BS) that needs to serve multiple users within a region of service.A subset of the users, especially those located close to thecell edge, undergo severe fading and suffer from poor channelquality and low achievable rates. To overcome this problem, C-NOMA is proposed as the system design methodology, in whichusers that have the capability of full-duplex (FD) communicationcan assist the transmissions between the BS and users withpoor channel quality through device-to-device (D2D) communi-cations. To harness both the multiplexing gain from NOMA andthe diversity gain from FD-D2D communications, we formulateand solve a novel optimization problem that jointly deter-mines D2D user pairing and power allocation. The formulatedproblem is a mixed-integer non-linear program (MINLP) withprohibitively high complexity. To overcome this issue, a two-steppolicy is proposed to solve the problem in polynomial time. Oursimulation results show that with reasonable assumptions, theproposed scheme always outperforms some existing schemesin the literature, and that, under undesirable conditions, e.g.,poor D2D channel conditions or imperfect self-interference (SI)cancellation, the proposed scheme is reduced to conventionalNOMA.
AB - This paper investigates the performance of coop- erative non-orthogonal multiple access (C-NOMA) in cellulardownlink systems. The system model consists of a base station(BS) that needs to serve multiple users within a region of service.A subset of the users, especially those located close to thecell edge, undergo severe fading and suffer from poor channelquality and low achievable rates. To overcome this problem, C-NOMA is proposed as the system design methodology, in whichusers that have the capability of full-duplex (FD) communicationcan assist the transmissions between the BS and users withpoor channel quality through device-to-device (D2D) communi-cations. To harness both the multiplexing gain from NOMA andthe diversity gain from FD-D2D communications, we formulateand solve a novel optimization problem that jointly deter-mines D2D user pairing and power allocation. The formulatedproblem is a mixed-integer non-linear program (MINLP) withprohibitively high complexity. To overcome this issue, a two-steppolicy is proposed to solve the problem in polynomial time. Oursimulation results show that with reasonable assumptions, theproposed scheme always outperforms some existing schemesin the literature, and that, under undesirable conditions, e.g.,poor D2D channel conditions or imperfect self-interference (SI)cancellation, the proposed scheme is reduced to conventionalNOMA.
KW - Cooperative NOMA
KW - Device-to-device communication
KW - Power control
KW - User pairing
UR - http://www.scopus.com/inward/record.url?scp=85081983743&partnerID=8YFLogxK
U2 - 10.1109/GLOBECOM38437.2019.9013180
DO - 10.1109/GLOBECOM38437.2019.9013180
M3 - Conference article
AN - SCOPUS:85081983743
SN - 2334-0983
JO - Proceedings - IEEE Global Communications Conference, GLOBECOM
JF - Proceedings - IEEE Global Communications Conference, GLOBECOM
M1 - 9013180
T2 - 2019 IEEE Global Communications Conference, GLOBECOM 2019
Y2 - 9 December 2019 through 13 December 2019
ER -