On Secrecy Performance of Antenna Selection Aided MIMO Systems Against Eavesdropping

TL;DR

This paper analyzes the secrecy performance of MIMO systems with antenna selection schemes, proposing optimal and suboptimal methods, and compares their effectiveness against traditional space-time transmission in terms of zero secrecy capacity probability.

Contribution

It introduces optimal and suboptimal antenna selection schemes for MIMO secrecy enhancement and derives their generalized secrecy diversity orders.

Findings

OAS scheme outperforms STT and SAS in zero secrecy capacity probability.

Secrecy diversity order equals the product of source and destination antennas.

Generalized secrecy diversity depends on antenna numbers, not SNR.

Abstract

In this paper, we consider a multiple-input multiple-output (MIMO) system consisting of one source, one destination and one eavesdropper, where each node is equipped with an arbitrary number of antennas. To improve the security of source-destination transmissions, we investigate the antenna selection at the source and propose the optimal antenna selection (OAS) and suboptimal antenna selection (SAS) schemes, depending on whether the source node has the global channel state information (CSI) of both the main link (from source to destination) and wiretap link (from source to eavesdropper). Also, the traditional space-time transmission (STT) is studied as a benchmark. We evaluate the secrecy performance of STT, SAS, and OAS schemes in terms of the probability of zero secrecy capacity. Furthermore, we examine the generalized secrecy diversity of STT, SAS, and OAS schemes through an asymptotic analysis of the probability of zero secrecy capacity, as the ratio between the average gains of the main and wiretap channels tends to infinity. This is different from the conventional secrecy diversity which assumes an infinite signal-to-noise ratio (SNR) received at the destination under the condition that the eavesdropper has a finite received SNR. It is shown that the generalized secrecy diversity orders of STT, SAS, and OAS schemes are the product of the number of antennas at source and destination. Additionally, numerical results show that the proposed OAS scheme strictly outperforms both the STT and SAS schemes in terms of the probability of zero secrecy capacity.