[Paper Review] Intelligent Reflecting Surface-Aided SWIPT: Joint Waveform, Active and Passive Beamforming Design.

Yang Zhao, Bruno Clerckx+1 more|arXiv (Cornell University)|Dec 10, 2020

Advanced Wireless Communication Technologies7 citations

TL;DR

This paper proposes an intelligent reflecting surface (IRS)-assisted simultaneous wireless information and power transfer (SWIPT) system that jointly optimizes transmit waveform, active beamforming at the AP, passive beamforming via IRS reflection coefficients, and user power splitting ratio. By leveraging geometric programming and successive convex approximation, the scheme significantly enhances the rate-energy tradeoff, even under practical frequency-flat IRS reflection constraints.

ABSTRACT

The performance of Simultaneous Wireless Information and Power Transfer (SWIPT) is severely restricted by the strength of the received Radio-Frequency (RF) signal. To tackle this problem, we introduce a low-power Intelligent Reflecting Surface (IRS) that compensates the propagation loss and boosts the transmission efficiency by a passive beamforming gain. This paper investigates an efficient IRS-aided SWIPT architecture where a multi-carrier multi-antenna Access Point (AP) transmits information and power simultaneously to a single-antenna user under the assist of an IRS. Considering energy harvester nonlinearity, we aim to maximize the Rate-Energy (R-E) tradeoff through a joint optimization of the transmit waveform and active beamforming at the AP, the reflection coefficients at the IRS, and the power splitting ratio at the user. Stationary solutions are achieved by the Alternating Optimization (AO) technique, where the optimal active beamforming is obtained in closed form, the passive beamforming is optimized by the Successive Convex Approximation (SCA) technique, and the waveform and splitting ratio are optimized by the Geometric Programming (GP) technique. Although practical IRS is limited to Frequency-Flat (FF) reflection, results demonstrate significant benefits of the proposed architecture based on a joint waveform and beamforming design to enlarge the R-E region of IRS-aided SWIPT.

Motivation & Objective

• Address the limited RF signal strength in SWIPT systems by introducing an intelligent reflecting surface (IRS) to enhance signal coverage and energy transfer.
• Overcome the performance bottleneck of conventional SWIPT due to path loss and weak signal reception at the user equipment.
• Maximize the rate-energy (R-E) tradeoff in a multi-carrier multi-antenna access point (AP) system assisted by IRS, considering user energy harvester nonlinearity.
• Simultaneously optimize transmit waveform, active beamforming at the AP, passive beamforming via IRS reflection coefficients, and user power splitting ratio.
• Achieve practical implementation by accounting for the frequency-flat reflection constraint typical of real-world IRS hardware.

Proposed method

• Employ Alternating Optimization (AO) to decompose the joint optimization problem into tractable subproblems.
• Derive closed-form solutions for optimal active beamforming at the multi-antenna AP using convex optimization techniques.
• Apply Successive Convex Approximation (SCA) to iteratively optimize the IRS reflection coefficients under frequency-flat constraints.
• Use Geometric Programming (GP) to jointly optimize the transmit waveform and user power splitting ratio for maximum rate-energy tradeoff.
• Model the nonlinearity of the user’s energy harvester to ensure realistic performance evaluation.
• Integrate all components into a unified framework that maximizes spectral efficiency and energy harvesting gain under practical IRS hardware limitations.

Experimental results

Research questions

• RQ1 How does joint optimization of transmit waveform, active beamforming, passive beamforming, and power splitting improve the rate-energy tradeoff in IRS-aided SWIPT?
• RQ2 To what extent can IRS compensate for path loss and enhance RF signal strength in SWIPT systems with nonlinear energy harvesting?
• RQ3 What performance gains are achievable when combining multi-carrier transmission with intelligent reflecting surfaces in SWIPT?
• RQ4 How do practical IRS constraints—particularly frequency-flat reflection—affect the system’s rate-energy region?
• RQ5 Can closed-form solutions for active beamforming and iterative optimization techniques for passive beamforming and waveform design yield effective and efficient solutions?

Key findings

• The proposed joint design achieves a significant expansion of the rate-energy (R-E) region compared to conventional SWIPT systems without IRS.
• Closed-form solutions for active beamforming are derived, enabling efficient computation and real-time implementation.
• Successive Convex Approximation (SCA) effectively optimizes the IRS reflection coefficients under frequency-flat constraints.
• Geometric Programming (GP) enables optimal joint optimization of the transmit waveform and user power splitting ratio.
• Despite the practical limitation of frequency-flat IRS reflection, the system still achieves substantial performance gains in both data rate and energy harvesting.
• Numerical results confirm that the proposed architecture outperforms benchmark schemes in terms of R-E tradeoff, demonstrating the effectiveness of intelligent reflecting surfaces in SWIPT.