PaperlayerView exact revision on arXiv

Source-linked AI summary

Wireless Communications Through Reconfigurable Intelligent Surfaces

Ertugrul Basar, Marco Di Renzo, Julien de Rosny, Merouane Debbah, Mohamed-Slim Alouini, Rui Zhang

arXiv:1906.09490v2eess.SPcs.IT

TL;DR

Wireless propagation is normally treated as uncontrollable and fading-prone, motivating RISs as a way to reconfigure how radio waves interact with the environment. The article surveys RIS technology and open issues, develops analytical models for link budget and error performance, and reports LOS-like distance scaling with an N^2 power gain under ideal phase control.

  • Problem

    Natural propagation causes attenuation, fading, and interference, while accurate communication-theoretic and physics-compliant models for reconfigurable surfaces remain limited.

  • Method

    The article synthesizes RIS research and uses analytical models, error-performance analysis, and channel-phase optimization to study RIS-assisted wireless systems.

  • Results

    With independently optimized RIS phases, received power scales proportionally to N^2 and decays with the inverse square of distance; RIS-based schemes also achieve lower SEP than classical BPSK in AWGN at low SNR.

  • Takeaways & Limitations

    RISs can shape multipath propagation deterministically and may serve as low-complexity wireless transmitters while retaining nearly passive operation.

Abstract

from arXiv · show

The future of mobile communications looks exciting with the potential new use cases and challenging requirements of future 6th generation (6G) and beyond wireless networks. Since the beginning of the modern era of wireless communications, the propagation medium has been perceived as a randomly behaving entity between the transmitter and the receiver, which degrades the quality of the received signal due to the uncontrollable interactions of the transmitted radio waves with the surrounding objects. The recent advent of reconfigurable intelligent surfaces in wireless communications enables, on the other hand, network operators to control the scattering, reflection, and refraction characteristics of the radio waves, by overcoming the negative effects of natural wireless propagation. Recent results have revealed that reconfigurable intelligent surfaces can effectively control the wavefront, e.g., the phase, amplitude, frequency, and even polarization, of the impinging signals without the need of complex decoding, encoding, and radio frequency processing operations. Motivated by the potential of this emerging technology, the present article is aimed to provide the readers with a detailed overview and historical perspective on state-of-the-art solutions, and to elaborate on the fundamental differences with other technologies, the most important open research issues to tackle, and the reasons why the use of reconfigurable intelligent surfaces necessitates to rethink the communication-theoretic models currently employed in wireless networks. This article also explores theoretical performance limits of reconfigurable intelligent surface-assisted communication systems using mathematical techniques and elaborates on the potential use cases of intelligent surfaces in 6G and beyond wireless networks.

I. INTRODUCTION

RISs turn the wireless environment from an uncontrollable source of fading and interference into a reconfigurable space that can shape radio propagation. The article surveys RIS principles, differences from related technologies, analytical models, open challenges, and potential 6G uses.

  • RIS concept: RISs are electronically controlled electromagnetic surfaces that enable network operators to reconfigure the propagation environment and create smart radio environments.They can actively participate in transferring and processing information while adapting to environmental changes.
  • RIS concept: RISs can counteract destructive multipath fading by coherently combining reflected signals through controlled interactions with surrounding objects.This approach addresses attenuation, fading, and uncontrollable interference caused by natural propagation.
  • Distinguishing features: Unlike relays and conventional MIMO beamforming, RISs are nearly passive, require ideally no dedicated energy source, and can be deployed on diverse surfaces.They are also described as contiguous programmable surfaces with full-band response and no ideal receiver noise from ADCs, DACs, or power amplifiers.
  • Research challenges: Fixed-function metasurfaces have limited impact in dynamic wireless environments, making reconfigurability fundamental for adapting to changing channels.RIS research therefore combines wireless communications, communication theory, computer science, physics, electromagnetism, and mathematics.
  • Article scope: The article uses simple analytical models focused on link budget and error performance while reviewing RIS technologies, open issues, and potential applications.It emphasizes why RIS deployment may require rethinking communication-theoretic models.

A. The Conventional Two-Ray System Model

The conventional two-ray model combines a line-of-sight path with a ground-reflected path whose uncontrolled phase difference can substantially reduce received power. The RIS-assisted variant instead adjusts the reflected ray to combine coherently with the direct path.

  • Conventional model: The conventional model contains a line-of-sight ray and a ground-reflected ray whose propagation distances determine their phase delays.The model uses geometrical optics, a sufficiently large ground plane, and specular reflection assumptions.
  • Conventional model: The ground reflection coefficient depends on ground material, wave polarization, and incidence angle, while the path difference determines the relative delay and phase.The received signal is modeled as the sum of the LOS and reflected components.
  • RIS-assisted model: The conventional two-ray received power decays with the fourth power of distance, whereas phase-aligned RIS assistance can restore second-power distance scaling.The RIS-assisted scaling matches the LOS free-space behavior described in the model.
  • Conventional model: The conventional reflected path can cause destructive interference because its phase is misaligned with the LOS path.Even one uncontrollable ground reflection can substantially degrade received signal strength without mobility or random environmental effects.
  • RIS-assisted model: With a reconfigurable metasurface, the reflected ray’s direction and phase can be controlled according to generalized or conventional Snell’s-law operation.In the considered phase-optimization case, the reflected phase is adjusted so the LOS and reflected rays add coherently.

C. The Two-Ray System Model with An RIS Made of Many Reconfigurable Meta-Surfaces

An RIS composed of independently configurable metasurfaces aligns the reflected components with the LOS path, producing a controllable power gain that scales with the square of the number of phases.

  • RIS structure: An RIS contains N metasurfaces whose reflection angles and phases can be tuned independently.The formulation assigns an index i to each reconfigurable metasurface.
  • Phase optimization: Unoptimized reflection coefficients can make the received power fluctuate significantly across the RIS elements.The analysis therefore considers independently optimized coefficients for coherent alignment.
  • Phase optimization: The optimized RIS aligns each reflected component’s phase with the LOS path while approximating each reflected path length by the transmitter-receiver distance.The assumed coefficients are R_i = e^j∆φ_i for all i.
  • Scaling law: The received power is proportional to N^2 and decays with the inverse square of the transmitter-receiver distance.Thus, independently controllable phases provide a power gain proportional to the square of their number while retaining LOS-like distance scaling.
  • Physical scale: A metasurface capable of shaping reflection angle and phase may have a size on the order of 10λ×10λ.This scale generally permits treatment as a specular reflector under geometrical optics.

E. Intelligent Reflection vs. Relaying and Backscattering: Reflectors vs. Diffusers

RISs differ from relaying and backscatter communications in both received-power scaling and physical operation: large passive surfaces behave as specular reflectors, whereas smaller devices behave as diffusers.

  • Transparent relaying and backscatter communications follow a fourth-power distance decay, Pr ∝ Pt/d^4, under the midpoint assumption.The relay or scattering tag is assumed to be midway between transmitter and receiver, with line-of-sight ignored.
  • RIS received-power scaling is sharply different from the path-loss law used for relay-aided and backscatter communications.The paper identifies this contrast as evidence that RISs require a different propagation model.
  • RISs are large enough relative to wavelength to be modeled as specular reflectors, unlike relay antennas and backscatter tags that act as diffusers.The distinction follows from the RIS geometric size being much larger than the wavelength, while the other devices are smaller than or comparable to it.
  • RIS path loss also reflects passive operation because RISs are assumed not to store or process impinging signals.

III. RECONFIGURABLE INTELLIGENT SURFACES: HOW DO THEY WORK?

RISs are reconfigurable electromagnetic sheets whose electronically controlled elements modify impinging waves. Implementations include frequency-selective surfaces, tunable resonators, software-controlled metasurfaces, and liquid-crystal reflectors.

  • RISs use many low-cost passive elements to modify impinging radio waves in ways unavailable from naturally occurring materials.Programmable metasurfaces can act as reconfigurable reflectors.
  • Active frequency-selective surfaces use PIN diodes to switch between nearly transparent and highly reflective states.The diode state determines whether incoming energy passes through or is mostly reflected.
  • A 0.4 m^2 spatial microwave modulator demonstrated 102 controllable electromagnetic reflectors operating at 2.47 GHz.Two Arduino 54-channel digital controllers controlled the reflectors.
  • Varactor-tuned resonators electronically change patch resonance, producing tunable phase shifts through bias-voltage control of each reflector unit.Electronic tuning changes resonant frequency without changing resonator dimensions.
  • Software-controlled metasurface tiles can be programmed to steer waves, control polarization, and absorb waves.The tiles implement electromagnetic functions by configuring electronic switches.
  • Liquid-crystal metasurface reflectors adjust unit-cell dielectric constants through DC voltages, enabling real-time phase control and beam steering.

IV. CONTROLLING THE MULTIPATH THROUGH RECONFIGURABLE INTELLIGENT SURFACES

The paper models RIS-assisted SISO transmission and derives average SEP behavior under Rayleigh fading, showing that phase control improves reliability and that performance scales strongly with the number of reflecting metasurfaces.

  • System model: The RIS-assisted SISO model uses N independently phase-configured reflecting metasurfaces between a single-antenna source and destination under Rayleigh fading.The channel coefficients from source to RIS and RIS to destination are modeled as independent complex Gaussian variables.
  • Equivalent representation: The RIS channel can be represented with a diagonal phase-shift matrix, making the model resemble MIMO precoding performed over the propagation environment.RIS and relay fast-fading models are similar, but their path-loss models differ when the RIS is sufficiently large to behave as a reflector.
  • Phase control: The instantaneous SNR is maximized by setting each RIS phase to the sum of the corresponding channel phases, requiring channel-phase knowledge at the RIS.This aligns the reflected components coherently at the destination.
  • SNR analysis: The average received SNR is proportional to N^2 Es/N0, confirming the quadratic gain associated with independently controllable RIS phases.This agrees with the received-power scaling derived for the RIS-assisted propagation model.
  • SEP performance: For BPSK with N = 16 and N = 32, the theoretical RIS scheme achieves significantly better average SEP than classical BPSK in AWGN, including low-SNR operation.The analysis identifies waterfall and slowly decaying SEP regions, with the latter still improving as N increases.
  • SEP performance: Doubling N yields approximately 6 dB improvement, or a four-fold decrease, in the required SNR in the waterfall region, while the CLT approximation is accurate for N ≥32.These observations come from simulated BPSK performance compared with the theoretical formula in (17).

V. RECONFIGURABLE INTELLIGENT SURFACE AS A LOW-COMPLEXITY AND ENERGY-EFFICIENT TRANSMITTER

An RIS can operate as a low-complexity transmitter by encoding information in the reflection phases of independently controllable metasurfaces illuminated by an unmodulated carrier. The analysis shows reliable transmission, a 1 dB SNR gain over reflective use, and an SNR penalty for higher-order signaling that can be offset by increasing N.

  • System concept: An illuminated RIS encodes data onto the phases of signals reflected from independently reconfigurable metasurfaces, avoiding conventional transmitter processing.The RIS is driven by a feeder or unmodulated carrier, while its reflection phases carry the information.
  • Signal model: The received signal forms a virtual bi-dimensional M-ary constellation whose information phases can be selected according to classical M-PSK.The message-dependent phase term is added to the phases optimized for coherent reflection and received-SNR maximization.
  • Binary signaling: High reliability is achievable, and RIS transmission provides a 1 dB SNR gain over using the RIS as a reflector.The comparison is made between the transmitter result in (28) and the reflector result in (19).
  • Higher-order signaling: For higher-order signaling M ≥16, an SNR loss is expected, but increasing N can reduce SEP in the relatively low-SNR range of interest.The conclusion compares the higher-order signaling result with (22).

VI. HISTORICAL PERSPECTIVE AND STATE-OF-THE-ART SOLUTIONS

The paper situates RIS research among unconventional wireless paradigms that exploit propagation or configurable scattering to simplify transceivers, improve QoS, or increase spectral efficiency.

  • Related paradigms: Spatial modulation maps information bits onto transmit-antenna indices by exploiting different fading realizations in MIMO systems.It is presented as a prominent example of index modulation.

A. Origins

Early intelligent-surface research developed methods for deliberately manipulating electromagnetic waves, progressing from fixed or static structures toward reconfigurable, programmable, and contiguous surfaces.

  • Early intelligent walls: Intelligent-wall designs used active frequency-selective surfaces to control signal strength, coverage, interference, and QoS.PIN-diode states produced transparent or highly reflective behavior, with simulations indicating efficient coverage and QoS control.
  • Learning-based control: ANN-controlled intelligent walls selected binary wall states from sensory data and system-performance measures.The cognitive engine used machine-learning algorithms to configure the walls.
  • Programmable metasurfaces: Programmable surfaces demonstrated passive microwave-field shaping, binary phase coding, scattering-pattern control, polarization control, and beam focusing.These functions were implemented through tunable metasurfaces, coding sequences, or PIN-diode states.
  • Reflect-arrays and LIS: Reflect-arrays and large intelligent surfaces extended controllable propagation through phase-shifting elements and contiguous electromagnetically active surfaces.The LIS concept uses the whole contiguous surface for transmission and reception, unlike traditional massive MIMO.
  • HyperSurfaces: HyperSurfaces combined ultra-thin meta-atoms with centralized control to absorb, focus, and steer impinging waves, while later work addressed sensing and adaptive configuration.The programmable wireless-environment concept was also generalized to multi-user interference minimization, eavesdropping, and multipath mitigation.

B. State-of-The-Art Solutions

State-of-the-art RIS research spans analytical models, multi-user optimization, channel estimation, energy efficiency, security, sensing, and extensions to diverse communication bands and systems. Reported studies combine RIS phase control with transceiver design and learning-based or compressive-sensing methods.

  • Research scope: Recent RIS studies analyze theoretical SNR and SEP, channel estimation, signal transmission and reception, energy efficiency, and related communication functions.The reviewed literature uses multiple names for RISs and covers a broad range of system objectives.
  • Multi-user optimization: Multi-user downlink work maximizes sum-rate by jointly optimizing RIS phases and user powers through alternating maximization and majorization-minimization.The resulting non-convex optimization reports improved overall system throughput.
  • Energy efficiency: Even 1-bit RIS phase resolution can increase energy efficiency relative to conventional amplify-and-forward relaying.Other studies also examine finite-resolution phases and practical energy-efficiency and sum-rate settings.
  • Finite-resolution control: Discrete phase shifts preserve the same power-scaling law with N as continuous shifts, while introducing a constant performance loss.Erroneous reflector phases have also been studied for their effect on error performance.
  • Channel estimation: RIS channel-estimation research includes compressive sensing, deep learning, active sparse sensors, and cascaded-channel estimation for passive elements.These approaches address unknown channel knowledge and training overhead in passive-RIS systems.
  • Applications and extensions: The literature also covers physical-layer security, over-the-air computation, sensing, and RIS-assisted millimeter-wave, terahertz, visible-light, free-space-optical, and OFDM systems.These studies expand RIS applications across objectives and communication modalities.

VII. POTENTIAL USE CASES

RISs are proposed for improving coverage, efficiency, and sustainability across difficult propagation scenarios while reducing transmitter complexity and power consumption. Key applications include reconfigurable non-LOS links, EM-pollution reduction, energy-free IoT, and low-complexity massive transmitters.

  • RISs can enhance coverage probability, spectral efficiency, and energy efficiency while reducing implementation complexity and power consumption.
  • Overcoming non-LOS Scenarios: Reconfigurable reflectors can create non-LOS links when line-of-sight paths are blocked or too weak, particularly at high frequencies.Relevant settings include millimeter-wave, D-band above 100 GHz, and visible-light transmission.
  • Overcoming Localized Coverage Holes: RISs can address localized coverage holes in urban areas and harsh indoor environments such as factories and underground metro stations.
  • Reducing the EM Pollution: RISs recycle existing radio waves constructively rather than generating new signals, potentially lowering electromagnetic radiation levels in hospitals and airplanes.
  • Energy-Free Internet of Things: RISs combined with backscatter communications could let IoT devices piggyback sensed data onto reflected waveforms at no overhead and zero energy cost.
  • Low-Complexity and Energy-Efficient Massive Transmitters: RIS transmitters can realize very large antenna arrays with a few, possibly one, active RF chain, reducing massive-MIMO complexity and power consumption.Reported testbeds leverage spatial modulation, media-based modulation, and index-modulation principles.

VIII. OPEN RESEARCH ISSUES

The paper identifies modeling, optimization, experimentation, and implementation challenges that must be addressed to realize RIS potential. Central gaps concern physical accuracy, channel validation, communication-theoretic limits, system-level analysis, feedback, and practical integration.

  • Physics- and EM-Compliant Models: Current RIS studies lack accurate and tractable electromagnetic models reflecting incidence, reflection, refraction, polarization, materials, and metasurface size.Perfect-reflector assumptions can produce overly simplistic models and optimistic performance predictions.
  • Experimentally-Validated Channel Models: No experimentally validated channel models currently provide realistic RIS path-loss, shadowing, and fast-fading statistics.Geometrical-optics accuracy depends on geometry and materials, while spatial correlation may affect performance limits and scaling laws.
  • Information- and Communication-Theoretic Models: RISs make conventional information- and communication-theoretic models obsolete, including Shannon-capacity formulations that must account for programmable system states.
  • Communication-Theoretic Performance Limits: RIS research needs tractable performance frameworks that reveal achievable communication limits as a function of the many constituent system parameters.
  • Spatial Models for System-Level Analysis and Optimization: Large-scale RIS networks require new spatial models because multiple surfaces may be distributed according to complex spatial patterns.A framework based on line processes was introduced under an stated spatial assumption.
  • Channel Estimation and Feedback Overhead: Nearly passive RIS operation creates fundamental challenges for channel estimation and feedback needed to optimize reflection phases.The idealized architecture avoids power amplifiers, channel estimators, ADCs, and DACs.
  • Implementation, Testbeds, and Field Trials: Reported testbeds initially support practical viability, but limited experimental detail prevents judging RIS potential under realistic operating conditions.
  • Data-Driven Optimization: Data-driven methods such as deep learning, reinforcement learning, and transfer learning are proposed to manage RIS system complexity.Machine learning is presented as a way to simplify implementation and increase efficiency.

IX. CONCLUSION

The paper surveys RIS-empowered wireless networks, distinguishes their propagation-control and transmitter roles, and analyzes their performance and use cases. It concludes that RISs offer controllable multipath and low-complexity transmission while leaving substantial research challenges.

  • The article surveys RIS research, contrasts RISs with relaying and backscatter communications, and identifies open issues for exploiting their potential.
  • RISs can shape radio waves to control multipath deterministically or realize energy-efficient transmitters with a limited, ideally one, active RF chain.
  • RIS-assisted error probability exhibits a waterfall behavior as the number of reconfigurable elements and the SNR increase.
  • For sufficiently large geometric size relative to wavelength, RISs can act as specular reflectors whose received-power distance dependence is determined at first order by Fermat’s principle.
Loading 1906.09490v2…