Optimal Transmit Antenna Selection for Massive MIMO Wiretap Channels

In this paper, we study the impacts of transmit antenna selection on the secrecy performance of massive MIMO systems. We consider a wiretap setting in which a fixed number of transmit antennas are selected and then confidential messages are transmitted over them to a multi-antenna legitimate receive...

Full description

Saved in:
Bibliographic Details
Published in:IEEE journal on selected areas in communications 2018-04, Vol.36 (4), p.817-828
Main Authors: Asaad, Saba, Bereyhi, Ali, Rabiei, Amir Masoud, Muller, Ralf R., Schaefer, Rafael F.
Format: Article
Language:English
Subjects:
Antennas
Approximation
Channel estimation
Computer simulation
Massive MIMO wiretap channel
Mathematical analysis
MIMO (control systems)
MIMO communication
physical layer security
Protocols
Receivers
transmit antenna selection
Transmitting antennas
Wiretapping
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we study the impacts of transmit antenna selection on the secrecy performance of massive MIMO systems. We consider a wiretap setting in which a fixed number of transmit antennas are selected and then confidential messages are transmitted over them to a multi-antenna legitimate receiver while being overheard by a multi-antenna eavesdropper. For this setup, we derive an accurate approximation of the instantaneous secrecy rate. Using this approximation, it is shown that in some wiretap settings under antenna selection the growth in the number of active antennas enhances the secrecy performance of the system up to some optimal number and degrades it when this optimal number is surpassed. This observation demonstrates that antenna selection in some massive MIMO settings not only reduces the RF-complexity, but also enhances the secrecy performance. We then consider various scenarios and derive the optimal number of active antennas analytically using our large-system approximation. Numerical investigations show an accurate match between simulations and the analytic results.
ISSN:0733-8716
1558-0008
DOI:10.1109/JSAC.2018.2825159