Modeling Cellular Networks with Full Duplex D2D Communication: A Stochastic Geometry Approach

Konpal S. Ali, Hesham Elsawy, Mohamed-Slim Alouini

Research output: Contribution to journalArticlepeer-review

65 Scopus citations


Full-duplex (FD) communication is optimistically promoted to double the spectral efficiency if sufficient self-interference cancellation (SIC) is achieved. However, this is not true when deploying FD-communication in a large-scale setup due to the induced mutual interference. Therefore, a large-scale study is necessary to draw legitimate conclusions about gains associated with FD-communication. This paper studies the FD operation for underlay device-to-device (D2D) communication sharing the uplink resources in cellular networks. We propose a disjoint fine-tuned selection criterion for the D2D and FD modes of operation. Then, we develop a tractable analytical paradigm, based on stochastic geometry, to calculate the outage probability and rate for cellular and D2D users. The results reveal that even in the case of perfect SIC, due to the increased interference injected to the network by FD-D2D communication, having all proximity UEs transmit in FD-D2D is not beneficial for the network. However, if the system parameters are carefully tuned, non-trivial network spectral-efficiency gains (64% shown) can be harvested. We also investigate the effects of imperfect SIC and D2D-link distance distribution on the harvested FD gains.
Original languageEnglish (US)
Pages (from-to)4409-4424
Number of pages16
JournalIEEE Transactions on Communications
Issue number10
StatePublished - Aug 24 2016

Bibliographical note

KAUST Repository Item: Exported on 2020-10-01
Acknowledgements: This work was supported by the KAUST Sensor Research Initiative sponsored by the KAUST Office of Competitive Research. The associate editor coordinating the review of this paper and approving it for publication was S. Durrani.


Dive into the research topics of 'Modeling Cellular Networks with Full Duplex D2D Communication: A Stochastic Geometry Approach'. Together they form a unique fingerprint.

Cite this