QoS-Constrained Optimization of Intelligent Reflecting Surface Aided Secure Energy-Efficient Transmission

Yuto Kawai,Shinya Sugiura

doi:10.1109/tvt.2021.3075685

Yuto Kawai, Shinya Sugiura

Open Access

https://doi.org/10.1109/tvt.2021.3075685

Copy DOI

Abstract

We propose an intelligent reflecting surface (IRS)-aided secure energy-efficient transmission that minimizes transmit power under constraints on the signal-to-noise ratios at the legitimate user and the eavesdroppers. This is achieved by jointly optimizing the beamforming weights at the transmitter and the phase shift coefficients at the IRS. To overcome the non-convexity of the problem, an alternating optimization algorithm is proposed. More specifically, by iteratively calculating the beamforming weights of a transmitter and the phase-shift coefficients at an IRS, the original optimization problem is separately approximated by a second-order cone programming problem. The results of the proposed algorithm are guaranteed to be the same as the solution of the original problem.

Highlights

I NTELLIGENT reflecting surfaces (IRS) have recently emerged as a key enabler for achieving high spectrum and energy efficiencies in future wireless communication systems, owing to their capability of controlling the wireless propagation environment in a low-cost architecture [1]–[3]
By iteratively calculating the beamforming weights of a transmitter and the phase-shift coefficients at the IRS, the original optimization problem is separately solved by second-order cone programming (SOCP), which achieves lower-complexity than semidefinite programing (SDP)
The IRS of Rose is composed of a uniform rectangular array (URA) with 5 rows and N/5 columns that is located in the y-z plane

Summary

INTRODUCTION

I NTELLIGENT reflecting surfaces (IRS) have recently emerged as a key enabler for achieving high spectrum and energy efficiencies in future wireless communication systems, owing to their capability of controlling the wireless propagation environment in a low-cost architecture [1]–[3]. The secrecy rate of CM − CW is achievable without relying on upper-layer cryptography, where CM and CW are the capacities of a legitimate channel and an eavesdropper’s channel, respectively. Assuming a transmitter, equipped with multiple antenna elements, communicates with a single legitimate receiver in the presence of multiple eavesdroppers, as well as an IRS, we propose an IRS-aided secure energyefficient transmission that minimizes transmit power under constraints on the signal-to-noise ratios (SNRs) at the legitimate user and the eavesdroppers, i.e., given constraints on the target secrecy rate. By iteratively calculating the beamforming weights of a transmitter and the phase-shift coefficients at the IRS, the original optimization problem is separately solved by second-order cone programming (SOCP), which achieves lower-complexity than semidefinite programing (SDP). Cm×n and Rm×n denote the spaces of m × n complex and real matrices, respectively

SYSTEM MODEL AND PROBLEM FORMULATION

Weight Calculations with Fixed Phase-Shift Matrix

Phase-Shift-Matrix Calculations with Fixed Weights

Proposed Iterative Algorithm

PERFORMANCE RESULTS

CONCLUSIONS