Intelligent Reflecting Surface Aided Wireless Energy and Information Transmission: An Overview

Qingqing Wu; Xinrong Guan; Rui Zhang

Source-linked AI summary

Intelligent Reflecting Surface Aided Wireless Energy and Information Transmission: An Overview

Qingqing Wu, Xinrong Guan, Rui Zhang

arXiv:2106.07997v3Jun 15, 2021cs.ITcs.ARcs.NI

TL;DR

The paper addresses how to design IRS-aided wireless energy and joint information-and-energy transmission systems under their distinct objectives, receiver architectures, and hardware constraints. It provides a tutorial overview of reflection design, channel acquisition, and resource allocation, concluding that IRS can improve WPT efficiency and benefit SWIPT and WPCN performance.

Problem
IRS research has largely emphasized wireless information transmission, leaving its potential to improve radio-frequency wireless energy transmission comparatively open.
Method
The paper provides a communication- and signal-processing-oriented tutorial of IRS-aided WPT, SWIPT, and WPCN, covering reflection design, channel acquisition, and resource allocation.
Results
IRS can significantly improve WPT efficiency while benefiting the rate-energy tradeoff in SWIPT and wireless-powered communication performance in WPCN.
Takeaways & Limitations
IRS reflection, channel acquisition, and resource allocation should be designed around the distinct performance metrics and hardware constraints of energy users.

Abstract

from arXiv · show

Intelligent reflecting surface (IRS) is a promising technology for achieving spectrum and energy efficient wireless networks cost-effectively. Most existing works on IRS have focused on exploiting IRS to enhance the performance of wireless communication or wireless information transmission (WIT), while its potential for boosting the efficiency of radio-frequency (RF) wireless energy transmission (WET) still remains largely open. Although IRS-aided WET shares similar characteristics with IRS-aided WIT, they differ fundamentally in terms of design objective, receiver architecture, and practical constraints. In this paper, we provide a tutorial overview on how to efficiently design IRS-aided WET systems as well as IRS-aided systems with both WIT and WET, namely IRS-aided simultaneous wireless information and power transfer (SWIPT) and IRS-aided wireless powered communication network (WPCN), mainly from a communication and signal processing perspective. In particular, we present state-of-the-art solutions to tackle the unique challenges in operating these systems, such as IRS passive reflection optimization, channel estimation and deployment. In addition, we also propose new solution approaches and point out important directions for future research and investigation.

I. INTRODUCTION

RF wireless power transfer can support the growing IoT ecosystem, but channel impairments and practical deployment constraints limit its efficiency. The paper motivates IRS as a cost-conscious approach for improving WPT and integrated WIT/WPT systems.

30.9 billion IoT devices are projected globally by 2025, increasing demand for reliable wireless energy supply.
RF WPT powers devices without wires over far-field links, but path loss, shadowing, and multipath fading reduce energy efficiency, especially over long distances.
Increasing transmitter radiation power raises received power but cannot improve end-to-end efficiency and may violate safety-related radiation or absorption limits.
Energy receivers generally require several tens of dB more received signal power than conventional information receivers, making efficient WPT a crucial challenge.
SWIPT concurrently delivers information and power over the same downlink waveform, creating a rate-energy tradeoff, whereas WPCN separates downlink harvesting from uplink information transmission.
Hardware improvements and communication techniques improve WPT/WIPT performance, but can be costly and may not overcome severe losses or deployment burdens.
IRS is motivated as a sustainable-cost technology for boosting WPT efficiency and WIPT performance without relying solely on bulky, power-hungry active infrastructure.

B. IRS-aided WIT and WPT

IRSs extend wireless-system design beyond transceivers by reconfiguring propagation, supporting communication, energy transfer, and integrated WIPT applications. IRS-aided WPT and WIPT require dedicated designs because their objectives, receiver architectures, and practical constraints differ from WIT.

IRS-enabled WPT: IRSs use smart reflections and large apertures to create focused passive energy beams that can extend WPT coverage in hotspot areas.They compensate signal attenuation over long distances and establish enhanced wireless charging zones.
IRS-enabled WPT: IRSs can guide energy signals around obstacles when line-of-sight links connect the access point, IRS, and users, supporting non-LoS WPT environments.This addresses efficiency degradation caused by blockage in conventional WPT systems.
Design challenges: IRS-aided WPT combines active energy beamforming at the access point with passive beamforming at the IRS to coherently combine signals at energy receivers.Transmit and reflect beamforming must be jointly designed to adjust impinging-signal amplitude and phase in real time.
Design challenges: IRS channel estimation is more difficult than conventional-link estimation because IRSs lack transceiver RF chains and contain many reflecting elements.These constraints accompany broader IRS-specific issues involving reflection design and deployment.
WIT versus WPT: WIT-oriented IRS designs may not transfer efficiently to WPT because the systems have different design objectives, receiver architectures, and practical constraints.For example, high channel correlation can benefit WPT energy beamforming even though low-rank LoS channels can be undesirable for multiuser WIT.
Paper scope: The paper provides a tutorial overview of IRS-aided WPT, SWIPT, and WPCN, reviews state-of-the-art solutions, proposes approaches, and identifies future research directions.Topics include joint active and passive beamforming, communication and energy resource allocation, and channel estimation for low-cost energy receivers.

II. IRS-AIDED WPT

This section formulates IRS-aided WPT as joint active and passive energy-beamforming and develops alternating optimization for its non-convex design. Optimized IRS deployment improves received EU sum-power and extends WPT range under the same transmit power.

System model and optimization: IRS-aided WPT jointly designs AP energy precoders and IRS phase shifts to maximize weighted EU sum-power under transmit-power and phase-shift constraints.EU weights determine priority in transferring energy among devices.
System model and optimization: The shared IRS phase shifts create a non-convex tradeoff among EUs, so the design alternates between energy precoders and IRS phase shifts.The reduced subproblems are solved with eigenvector-based precoding and successive convex approximation.
Optimization approach: For fixed IRS phases, one common energy beam aligned with the principal eigenvector is optimal; a dominant EU weight steers the beam toward that EU.When one weight greatly exceeds the others, the weighted objective is approximately governed by that EU's channel.
Optimization approach: The alternating-optimization objective is non-decreasing and bounded, and the resulting algorithm is guaranteed to converge to a stationary solution.The precoder and phase-shift variables do not couple in the constraints.
Performance evaluation: Joint active and passive energy beamforming significantly outperforms optimized AP beamforming with random IRS phase shifts.The comparison demonstrates the importance of properly optimized IRS phase shifts for energy-beamforming gain.

B. Channel Acquisition

IRS-aided WPT requires channel-acquisition methods tailored to additional IRS-related coefficients and power-constrained EUs. The section reviews conventional approaches and presents customized feedback-based learning for separate-antenna systems.

Motivation and architectures: Conventional channel-estimation methods may be inefficient or infeasible because IRS-aided WPT adds many channel coefficients while EUs have limited power and processing capability.These constraints make receiver pilot transmission and signal processing costly for WPT devices.
Motivation and architectures: Shared-antenna WPT uses time-division energy harvesting and communication, whereas separate-antenna WPT operates both modules concurrently and independently.The architecture determines which channel-acquisition procedures are applicable.
WPT without IRS: Downlink training with uplink CSI feedback supports TDD and FDD but requires complex EU baseband processing, and training time grows with AP antenna count M.This can be unaffordable for low-cost devices such as RF tags.
WPT without IRS: Uplink training via channel reciprocity removes EU-side estimation and feedback and makes training time independent of M, but consumes harvested energy and transmission time.It also creates a training-energy tradeoff: insufficient training reduces beamforming gain, whereas excessive training leaves less time for energy transmission.
Separate-antenna WPT: Separate-antenna systems cannot obtain energy-harvesting-antenna CSI by estimating channels with communication antennas.The section therefore motivates a channel-learning approach using the communication module to report harvested energy levels.
Separate-antenna WPT: Quantized harvested-energy feedback lets the AP adjust later beamforming and refine channel estimates for the energy-harvesting antennas.The EU measures and encodes harvested energy over intervals into bits before feedback.

2) Channel Acquisition in IRS-aided WIT Systems:

IRS-aided WIT channel acquisition uses semi-passive sensing, cascaded-channel estimation, element grouping, common-channel exploitation, and codebook beam training. The approaches trade estimation overhead, receiver assumptions, feedback, and TDD/FDD applicability.

Channel acquisition approaches: Semi-passive IRSs use sensors to receive pilots and estimate AP/IU-to-IRS channels, exploiting spatial correlation to reconstruct channels at reflecting elements.Compressed sensing, interpolation, and machine learning can support reconstruction.
Channel acquisition approaches: Fully-passive IRSs estimate cascaded AP-IRS-IU channels through designed reflection patterns, because separately estimating AP-IRS and IRS-IU channels is infeasible.The required pilot overhead can become prohibitively high with many reflecting elements.
Overhead reduction: IRS element grouping reduces overhead by estimating one effective cascaded channel per sub-surface rather than one channel for every reflecting element.For multi-user systems, cascaded channels also share a common AP-IRS channel, enabling reference-user estimation and low-dimensional scaling factors.
Overhead reduction: Codebook beam training acquires channels implicitly by comparing received powers across IRS reflection patterns and feeding back the index of the maximum-gain pattern.This approach avoids explicit channel estimation and applies to TDD and FDD systems.
WPT-specific acquisition: The proposed WPT schemes use two-phase active/passive beam training with EU energy measurements for single-user and multi-EU settings.The passive-training approach requires only low-rate feedback, without EU pilot transmission or complex signal processing; training time is independent of AP antenna count.

C. Extensions and Future Work

The overview identifies open issues in IRS-aided WPT involving realistic energy-harvester models, broadband channel acquisition, reflection constraints, and deployment. It also highlights promising directions based on waveform design, OFDM redundancy, and cooperative IRS placement.

Deployment: IRS-aided WPT research remains at an early stage, with deployment strategies and cooperative beamforming among distant IRSs requiring further investigation.Hybrid deployment is expected to provide superior WPT performance when multiple IRSs cooperate.
Energy-harvester modeling: The proposed designs assume a diode linear energy-harvester model, whereas nonlinear models make output DC power depend on the received waveform as well as signal power.Waveform and IRS passive beamforming may therefore require joint optimization.
Practical IRS constraints: The effects of amplitude-dependent or discrete IRS phase shifts, and nonlinear harvesting models, on waveform, beamforming, and training remain open.Prior work assumed continuous IRS phase shifts.
Broadband channel acquisition: Frequency-selective WPT channel acquisition is harder because multipath creates more coefficients, while IRS reflection coefficients are time-selective but frequency-flat.OFDM subcarrier redundancy has enabled efficient multi-user estimation in WIT, but broadband WPT remains insufficiently studied.
Broadband channel acquisition: Training efficiency could improve by jointly exploiting redundancy across antennas and OFDM subcarriers, extending spatial-overhead reduction ideas beyond narrowband systems.The cited OFDM approach exploits subcarrier redundancy, while element grouping and common-channel methods exploit spatial redundancy.
Deployment: IRS deployment strongly affects end-to-end channels and performance, with LoS links preferred for WPT and hybrid placement near APs and EUs promising flexibility.Multi-IRS designs introduce placement, association, and multi-hop beamforming challenges.

III. IRS-AIDED WIPT

The paper extends IRS-aided wireless power transmission to systems combining wireless information and energy transfer, including SWIPT and WPCN.

Scope: The paper studies IRS-aided SWIPT and WPCN as extensions of IRS-aided WPT.These systems combine information and energy transmission within the broader IRS-aided wireless-power framework.

A. IRS-aided SWIPT

IRS-aided SWIPT jointly designs information and energy transmission under rate, power, and unit-modulus IRS constraints. The IRS enlarges the power-SINR tradeoff, while information beams alone can achieve the optimum under arbitrary user channels.

System objective: SWIPT must balance information rates and harvested energy by allocating communication resources such as power, time, and bandwidth.The section develops joint beamforming and channel-acquisition designs for this tradeoff.
Joint beamforming design: The joint optimization maximizes weighted harvested sum-power subject to AP transmit-power, IRS unit-modulus, and individual IU SINR constraints.IRS phase shifts couple with both information and energy precoders through the SINR constraints and harvested-power objective.
Joint beamforming design: Dedicated energy beams consume AP power and interfere with IUs, explaining their avoidance in the optimal design.This conclusion differs from the unresolved question posed for IRS-aided SWIPT before the arbitrary-channel result.
Joint beamforming design: Under arbitrary user channels, sending information beams is sufficient for optimality, eliminating the need for dedicated energy beams.The result simplifies AP precoding, especially when IRS-induced channel correlation makes independent-channel assumptions unsuitable.
Numerical illustration: The achievable SWIPT power-SINR region can be significantly enlarged by deploying an IRS, improving EU RF power without compromising IU SINR.A separate information/energy-beam design suffers considerable performance loss relative to joint beamforming.

2) Channel Acquisition:

The paper extends energy-measurement-based passive beam training from IRS-aided WPT to SWIPT, using a three-phase protocol with separate AP information and energy beams.

2) Channel Acquisition:: The proposed SWIPT channel acquisition and transmission protocol divides each channel coherence time into active beam training, passive beam training, and simultaneous transmission phases.The separate AP information/energy beam design is adopted for practical implementation.
2) Channel Acquisition:: During active beam training, IUs and the IRS controller transmit pilots so the AP can estimate AP-IU/controller and AP-IRS-controller channels.The AP then uses the estimated CSI to optimize active beamforming.
2) Channel Acquisition:: The AP transmits one information beam per IU and an energy beam to the IRS controller before passive beam training.This separates information-beam and energy-beam design while enabling the controller to coordinate IRS training.
2) Channel Acquisition:: During passive training, the IRS cycles through codebook beam patterns, and each EU selects the pattern yielding the largest harvested-energy measurement.Each EU feeds the selected beam-pattern index back to the IRS controller.
2) Channel Acquisition:: In the final phase, information and energy signals are transmitted simultaneously while the IRS time-shares users’ individually selected passive beam patterns.This extends the WPT scheme, which requires only energy measurements and low-rate EU feedback, to the more complex SWIPT setting.

3) Extensions and Future Work:

The section reviews IRS-aided SWIPT and WPCN extensions while identifying open problems in receiver types, security, deployment, and joint system design.

3) Extensions and Future Work:: Existing IRS-aided SWIPT studies mainly use power-splitting receivers and optimize objectives including energy efficiency, harvested DC power, or AP transmit power.One cited work jointly optimizes transmit waveform, active beamforming, and passive beamforming for SWIPT.
3) Extensions and Future Work:: The reviewed SWIPT work largely leaves multiuser and time-switching-receiver scenarios for further study, alongside IRS deployment for mixed receiver types.The section also calls for joint design of channel acquisition, beamforming, and information/energy waveforms.
3) Extensions and Future Work:: IRS passive reflections can improve the secrecy rate-energy region by increasing or reducing achievable rates for information and energy users.Energy beams are unnecessary under perfect CSI in one formulation, but may serve as artificial noise under secrecy, energy, or imperfect-CSI constraints.
3) Extensions and Future Work:: A remaining security problem is optimally assigning a fixed total number of reflecting elements among IRSs near the AP, information users, and energy users.This assignment is tied to achieving the optimal secrecy rate-energy region.
3) Extensions and Future Work:: In WPCN, downlink energy harvesting and uplink information transmission use opposite directions and share time-frequency resources, motivating joint passive beamforming and resource allocation.The section presents this design and discusses the associated CSI acquisition issue.

1) Joint Passive Beamforming and Resource Allocation:

For IRS-aided WPCN, the paper jointly optimizes downlink WPT and uplink WIT through IRS phase shifts, device powers, and time allocation, and solves the resulting formulation using semidefinite relaxation methods.

1) Joint Passive Beamforming and Resource Allocation:: The considered WPCN has an IRS-assisted downlink WPT phase followed by uplink WIT from K wireless-powered devices to a single-antenna AP.The AP broadcasts energy for duration τ0, while each device transmits during its own uplink duration τk.
1) Joint Passive Beamforming and Resource Allocation:: Optimal uplink IRS phase shifts align the reflected signal with the direct AP-device link to maximize received signal power.The IRS can reconfigure its phase-shift vector separately for each device’s uplink transmission.
1) Joint Passive Beamforming and Resource Allocation:: The objective is weighted sum-rate maximization over WPT and WIT time allocations, device transmit powers, and IRS phase shifts in both phases.The rate weights αk represent device priorities.
1) Joint Passive Beamforming and Resource Allocation:: Relaxing the rank-one constraint converts the reformulated optimization into a convex semidefinite program solvable by standard convex solvers.Gaussian randomization or iterative largest-singular-value penalties can produce feasible rank-one solutions.
1) Joint Passive Beamforming and Resource Allocation:: IRS deployment with joint beamforming and resource allocation significantly improves WPCN sum throughput over no IRS, whereas random IRS phases provide only marginal gains.The comparison uses simulations varying the number of IRS elements with K = 10 and equal rate weights.
1) Joint Passive Beamforming and Resource Allocation:: The optimized downlink WPT duration decreases as the number of IRS elements increases, reducing AP energy consumption and leaving more time for uplink WIT.The paper characterizes these as dual benefits: higher system throughput and lower AP energy consumption.
1) Joint Passive Beamforming and Resource Allocation:: IRS-aided WPCN therefore improves throughput while lowering AP energy consumption under the simulated joint design.The paper explicitly describes these as double benefits of exploiting IRS for WPCN.

2) Channel Acquisition:

The paper extends energy-feedback passive beam training to WPCN and discusses future directions involving multiple access, mobility, self-sustainable IRS operation, and energy-reflection tradeoffs.

2) Channel Acquisition:: The WPCN channel coherence interval is divided into active training, passive training, downlink energy harvesting, and uplink data transmission phases.During uplink transmission, users employ TDMA and the IRS selects each user’s optimal phase-shift vector.
2) Channel Acquisition:: The assumed TDMA-based uplink can be extended to SDMA or NOMA, which generally require different beamforming and resource-allocation designs.NOMA-based IRS-aided WPCN has been studied through joint time-allocation and IRS-phase-shift optimization.
2) Channel Acquisition:: NOMA-based IRS-aided WPCN found that the same IRS phase-shift vector should be used for downlink WPT and uplink WIT.This differs from the phase-shift treatment described for TDMA-based uplink WIT.
2) Channel Acquisition:: For many users over a wide area, deploying IRSs near user clusters can reduce the mobile charger’s need to approach each user closely.This can address excessive charger power consumption and complex path planning associated with user-by-user movement.
2) Channel Acquisition:: Mobile IRSs mounted on AGVs or UAVs can expand reflection coverage in angle and distance by adapting their facing or location to user locations.Mobility also introduces rapidly time-varying channels and complicates channel acquisition.
2) Channel Acquisition:: RF energy-harvesting modules can provide additional IRS energy supply and may enable self-sustainable operation, using switching or splitting architectures.Possible hardware designs include antenna/element switching, time switching, and power splitting.
2) Channel Acquisition:: Self-powered IRS designs face a fundamental tradeoff between harvested energy and time available for signal reflection.Too little harvesting time limits reflection performance, whereas too much reduces time aiding WPT, SWIPT, or WPCN.

C. IRS-aided Ambient Energy Harvesting and Backscatter Communications

IRS can improve ambient RF energy harvesting by focusing otherwise random signals at energy users, while also supporting backscatter communication through reflection design. The section highlights unresolved challenges in channel acquisition, adaptive reflection, passive-reflection path loss, and practical IRS deployment.

Ambient Energy Harvesting: IRS passive beamforming can focus ambient signals at energy users and significantly improve energy-harvesting efficiency.Ambient RF signals are random, making efficient harvesting difficult; IRS addresses this by focusing them at EUs.
Ambient Energy Harvesting: Energy-feedback beam training can optimize the passive beam pattern using EUs’ energy measurements without explicitly estimating CSI.This is relevant because the channels from unknown RF resources to EUs are usually unknown.
Ambient Energy Harvesting: Ambient RF sources may vary in signal strength and angle of arrival, requiring adaptive IRS reflection to track the strongest source.The approach targets changing ambient sources rather than a fixed dedicated transmitter.
Backscatter Communications: Ambient backscatter avoids a dedicated carrier source but faces receiver interference from unknown ambient carriers and random carrier messages.IRS may enhance the desired backscatter signal or suppress interference through proper signal reflection.
Backscatter Communications: Double passive reflections make IRS-aided backscatter reflection design and channel acquisition more involved than in conventional IRS-aided WPT/WPCN.The added complexity arises from the passive backscatter device and IRS elements both contributing reflections.
Future Directions: Passive IRS can require many reflecting elements because its product-distance path loss limits performance at low energy and hardware cost.Active IRS amplifies reflected signals using negative-resistance components but entails higher hardware and energy costs; optimal WPT/WIPT designs remain open.
Future Directions: Reflection-type metasurfaces serve transmitters and receivers on the same side, whereas refraction-type metasurfaces support opposite-side coverage enhancement.Transmissive-IRS-aided WPT and WIPT remain identified as an important future research topic.
Conclusions: Passive IRS improves WPT efficiency, the SWIPT rate-energy tradeoff, and wireless-powered communication performance in WPCN.The overview studies IRS reflection design, channel acquisition, and resource allocation under EU performance metrics and hardware constraints.