[Paper Review] Intelligent Reflecting Surface Enhanced Wireless Network via Joint Active and Passive Beamforming

Qingqing Wu, Rui Zhang|arXiv (Cornell University)|Oct 6, 2018

Advanced Wireless Communication Technologies32 references103 citations

TL;DR

The paper studies jointly optimizing AP transmit beamforming and IRS phase shifts to minimize AP transmit power under user SINR constraints, for single-user and multiuser settings, with SDR and alternating-optimization approaches and a derived N-scaling law.

ABSTRACT

Intelligent reflecting surface (IRS) is envisioned to be a new and revolutionizing technology for achieving spectrum and energy efficient wireless communication networks cost-effectively in the future. Specifically, an IRS consists of a large number of low-cost passive elements each reflecting the incident signal with a certain phase shift to collaboratively achieve beamforming and/or interference suppression at designated receivers. In this paper, we study an IRS-aided multiuser multiple-input single-output (MISO) wireless system where one IRS is deployed to assist in the communication from a multi-antenna access point (AP) to multiple single-antenna users. As such, each user receives the superposed signals from the AP as well as the IRS via its reflection. We aim to minimize the total transmit power at the AP by jointly optimizing the transmit beamforming by active antenna array at the AP and reflect beamforming by passive phase shifters at the IRS, subject to users' individual signal-to-interference-plus-noise ratio (SINR) constraints. However, the formulated problem is non-convex and difficult to be solved optimally.

Motivation & Objective

• Motivate spectrum and energy-efficient wireless networks using IRS to shape propagation channels.
• Formulate and solve a power-minimization problem with joint AP beamforming and IRS phase-shifting under per-user SINR constraints.
• Analyze the feasibility condition and provide practical algorithmic approaches for single-user and multiuser scenarios.
• Characterize the transmit power scaling and compare IRS-assisted performance to conventional systems.

Proposed method

• Model the AP-IRS-user downlink with a multi-antenna AP, an IRS of N elements, and K single-antenna users.
• Formulate problem (P1) to minimize total AP transmit power subject to SINR constraints and unit-modulus IRS phase shifts.
• Apply semidefinite relaxation (SDR) to the single-user problem (P2) to obtain a lower bound and near-optimal solutions; extract a rank-one solution via SDR with randomization.
• Develop an alternating optimization algorithm that updates the IRS phase shifts and AP beamforming direction in closed form, guaranteeing convergence.
• Extend the approach to the multiuser case with two suboptimal algorithms trading off performance and complexity.
• Derive analytical insights and a power scaling law for infinite IRS elements showing N^2 gain under optimal phase design.

Experimental results

Research questions

• RQ1 Can joint optimization of AP transmit beamforming and IRS phase shifts meet SINR targets while minimizing AP transmit power?
• RQ2 What algorithms efficiently solve the non-convex joint design problem for single-user and multiuser cases?
• RQ3 How does the IRS influence feasibility and what are the conditions ensuring problem feasibility?
• RQ4 What is the asymptotic power scaling with the number of IRS elements N under optimal phase design?
• RQ5 How does IRS-aided performance compare to traditional massive MIMO or AF relay benchmarks?

Key findings

• Jointly optimizing AP beamforming and IRS phases significantly reduces required AP transmit power compared to systems without IRS.
• For a single user near the IRS, AP transmit power scales with N^2 under large N when using optimal IRS phase shifts.
• SDR provides a computable bound and, with randomization, yields a guaranteed approximation quality for the single-user problem.
• Alternating optimization yields closed-form updates for phase shifts and beamforming that converge to a stationary solution.
• Numerical results indicate substantial performance gains of IRS-enhanced schemes over benchmarks like massive MIMO without IRS.