PaperlayerView exact revision on arXiv

Source-linked AI summary

RIS-Enabled Smart Wireless Environments: Deployment Scenarios, Network Architecture, Bandwidth and Area of Influence

George C. Alexandropoulos, Dinh-Thuy Phan-Huy, Kostantinos D. Katsanos, Maurizio Crozzoli, Henk Wymeersch, Petar Popovski, Philippe Ratajczak, Yohann Bénédic, Marie-Helene Hamon, Sebastien Herraiz Gonzalez, Placido Mursia, Marco Rossanese, Vincenzo Sciancalepore, Jean-Baptiste Gros, Sergio Terranova, Gabriele Gradoni, Paolo Di Lorenzo, Moustafa Rahal, Benoit Denis, Raffaele D'Errico, Antonio Clemente, Emilio Calvanese Strinati

arXiv:2303.08505v1cs.ITcs.ETmath.IT

TL;DR

The paper addresses the open question of when RIS deployments improve performance over conventional networks. It presents deployment scenarios and defines the Bandwidth and Area of Influence metrics; its conclusion states that Area of Influence depends on RIS design and placement and can be dynamically optimized for performance objectives.

  • Problem

    The scenarios in which RISs substantially improve performance over conventional network architectures have not been fully identified.

    Source: environment offers dynamic radio wave… · p. 3

  • Method

    The paper presents RIS deployment scenarios and introduces mathematical definitions of the Bandwidth and Area of Influence metrics.

    Source: §VI · pp. 39–40

  • Results

    The Area of Influence depends on RIS hardware design and placement and can be dynamically optimized to meet performance objectives.

    Source: §VI · p. 40

  • Takeaways & Limitations

    The paper concludes that RIS technology requires careful deployment and planning.

    Source: §VI · pp. 39–40

Abstract

from arXiv · show

Reconfigurable Intelligent Surfaces (RISs) constitute the key enabler for programmable electromagnetic propagation environments, and are lately being considered as a candidate physical-layer technology for the demanding connectivity, reliability, localization, and sustainability requirements of next generation wireless networks. In this paper, we first present the deployment scenarios for RIS-enabled smart wireless environments that have been recently designed within the ongoing European Union Horizon 2020 RISE-6G project, as well as a network architecture integrating RISs with existing standardized interfaces. We identify various RIS deployment strategies and sketch the core architectural requirements in terms of RIS control and signaling, depending on the RIS hardware architectures and respective capabilities. Furthermore, we introduce and discuss, with the aid of simulations and reflectarray measurements, two novel metrics that emerge in the context of RIS-empowered wireless systems: the RIS bandwidth and area of influence. Their extensive investigation corroborates the need for careful deployment and planning of the RIS technology in future networks.

I. INTRODUCTION

RISs are proposed as programmable propagation elements for next-generation wireless networks, with potential benefits in coverage, efficiency, reliability, and security. The paper presents deployment scenarios and network architecture, and introduces bandwidth and area-of-influence metrics whose investigation supports careful RIS planning.

I. INTRODUCTION

  • RISs are proposed as programmable propagation elements that can be jointly optimized with conventional transceivers to improve coverage, spectral and energy efficiency, reliability, and security.The envisioned systems also satisfy regulated electromagnetic-field emissions.Source: §I · p. 2
  • Existing demonstrations have mainly shown simple RIS functions, including electronic beam steering and multi-beam scattering.Source: §I · p. 2

and the resulting challenges. Those architectures endow surfaces with either minimal reception

  • RIS hardware can support sensing or coverage extension, while relay-like amplification can mitigate severe multiplicative path loss.The cited architectures include RF chains or minimal power amplifiers.Source: and the resulting challenges. Those… · p. 3

environment offers dynamic radio wave control for trading off high-throughput communications,

  • The paper addresses the gap in identifying when RIS deployments substantially improve performance over conventional networks by presenting RISE-6G scenarios and RIS control and signaling requirements.The requirements depend on RIS hardware architectures and their capabilities.Source: environment offers dynamic radio wave… · p. 3

ical definitions and computer simulations as well as experimental investigations with RISE-6G’s

II. RIS DEPLOYMENT SCENARIOS

The section groups RIS deployment scenarios by application and performance objective, distinguishing performance gains for established systems from applications otherwise infeasible. For connectivity and reliability, it describes single- and multiple-RIS configurations and highlights their control needs and interference interactions.

II. RIS DEPLOYMENT SCENARIOS

  • RIS deployment scenarios address connectivity and reliability, localization and sensing, sustainability, and secrecy, and may boost established systems or enable otherwise infeasible applications.Source: §II · p. 4

A. Connectivity and Reliability

  • RIS configuration depends on localization feasibility and expected performance, while the control channel's rate and latency must meet system requirements.The scenarios described here assume reflecting-mode RISs.Source: A. Connectivity and Reliability · p. 4

1) Connectivity and Reliability Boosted by a Single RIS:

2) Connectivity and Reliability Boosted by Multiple Individually Controlled RISs:

  • With multiple individually controlled RISs, each base-station–user pair controls its assigned surface, but RIS operation can create multiplicative interference on unintended links.This differs from conventional additive wireless interference and requires careful management.Source: 2) Connectivity and Reliability… · pp. 5–6

3) Connection Reliability Enabled by Multiple RISs:

This section examines how RISs support connection reliability and connectivity, including blocked-link scenarios, shared multi-tenant surfaces, and MEC. It also covers RIS-assisted localization and sensing, including UE and radar localization and joint channel and direction estimation for ground-to-drone communications.

3) Connection Reliability Enabled by Multiple RISs:

  • When direct BS–UE links are blocked, connectivity can rely on one or more RISs, whose phase profiles may be optimized for target reliability, potentially requiring coordination among BSs.The proposed coordination may resemble coordinated beamforming in LTE-Advanced.Source: 3) Connection Reliability Enabled by… · p. 6

4) Connectivity and Reliability Boosted by a Single Multi-Tenant RIS:

  • A shared RIS can serve multiple BS–UE pairs through a commonly optimized phase profile or time-shared control, although its control channel requires further investigation.The RIS is dynamically controlled by the infrastructure and accessed by the BS–UE pairs.Source: 4) Connectivity and Reliability… · p. 6
  • Shared RISs placed at cell boundaries can dynamically extend serving-BS coverage as UEs cross cells, reducing frequent handovers.Frequent handovers may occur when mobile UEs cross the boundaries of cells served by two or more BSs.Source: 4) Connectivity and Reliability… · p. 6

5) RIS-Empowered Multi-Access Edge Computing:

  • RISs can create adaptive service-boosted areas for MEC, trading energy efficiency, latency, and reliability to meet temporary, location-dependent needs, and connect UEs to an edge server through a backhaul-connected BS.MEC moves computing and storage toward the network edge so UEs can offload tasks to nearby processing units or edge servers.Source: 5) RIS-Empowered Multi-Access Edge… · p. 7

B. Localization and Sensing

  • RIS localization is classified as enabled, boosted, or low-profile (using much lower resources); one RIS enables localization, two RISs permit narrowband UE-location estimation, and wavefront curvature supports direct positioning,,.Source: B. Localization and Sensing · pp. 8–9
  • Selective control of multiple RISs can provide users with different localization accuracy and GDoP, while RIS-controlled multipath-profile evolution over BS–UE links enables passive-object sensing –[38].In highly reverberant environments, multiple RISs can create configurational diversity.Source: B. Localization and Sensing · pp. 9–10
  • The cited passages identify RIS sensing/localization scenarios as mitigating rubble-induced shadowing through an ad-hoc process, localization without BSs using one or multiple RISs, and electromagnetic cartography from RIS-enhanced radio-signature features.Source: B. Localization and Sensing · p. 10
  • Reported configurations include RIS-lens localization via narrow-band received signals exploiting wavefront curvature, indoor/outdoor localization with simultaneously transmitting and reflecting RISs, and multiple-hybrid-RIS radar to localize or detect static or moving targets [10],, [47].Source: B. Localization and Sensing · pp. 10–11
  • RIS elements and reflectors provide the system with the capability to localize UEs.Source: B. Localization and Sensing · p. 11
  • RIS elements and reflectors also provide the capability to localize the radar itself.Source: B. Localization and Sensing · p. 11
  • RISs can be deployed for joint channel and direction estimation for ground-to-drone wireless communications.Source: B. Localization and Sensing · p. 11

C. Sustainability and Secrecy

RISs can shape propagation to improve energy efficiency, reduce exposure or interception, and support sustainability and secrecy objectives. The section describes orchestrated RIS deployment in small areas as one use case.

C. Sustainability and Secrecy

  • RISs can focus energy toward an intended UE and reduce unnecessary radiation toward non-intended UEs, including exposed UEs and eavesdroppers.The section identifies energy efficiency and electromagnetic-field exposure utility as metrics that may improve.Source: C. Sustainability and Secrecy · p. 11
  • Secrecy spectral efficiency (SSE) is defined as the target UE's data rate minus the rate intercepted by an eavesdropper, [50].Energy efficiency is also described as target data rate divided by BS transmit power.Source: C. Sustainability and Secrecy · p. 11
  • A train station is identified as a use case where orchestrated RISs can optimize these metrics in small areas close to the surfaces.Such areas could be advertised as having high energy efficiency, low EMF exposure, or increased security.Source: C. Sustainability and Secrecy · p. 11

increased secrecy.”

RISs can boost received power at an intended UE while weakening propagation toward a non-intended UE. The section describes three propagation settings and notes extensions to uplink and coordinated multi-BS scenarios.

increased secrecy.”

  • RISs can shape propagation to boost received power at the intended UE and reduce it at a non-intended UE, supporting a low-EMF or secured link.The non-intended UE may be an exposed UE or an eavesdropper.Source: increased secrecy.” · p. 11
  • With a line-of-sight BS-to-target link, RIS-added paths combine coherently at the target and non-coherently at the non-intended UE.The stated effect is increased target power and reduced received power at the non-intended UE.Source: increased secrecy.” · p. 12
  • When an obstacle blocks the BS-to-target link, an RIS can add a path in the near field; in the far field, several RISs may help reduce energy at the non-intended UE.The far-field case is described as similar to the near-field case, except that several RISs may become useful.Source: increased secrecy.” · p. 12
  • Without a non-intended UE, the described single-BS downlink beamforming setup boosts target power and therefore improves energy efficiency.The section also notes that uplink receive beamforming can improve UE-side energy efficiency and reduce a UE's self-exposure.Source: increased secrecy.” · pp. 11–12

scenarios can be generalized to multi-BS scenarios, where several synchronized and coordinated

III. RIS INTEGRATION IN STANDARDIZED NETWORKS

Reflective RISs re-radiate incident waves and have no explicit coverage area or transmit bandwidth, complicating their treatment in conventional network planning and spectrum regulation. The paper introduces area and bandwidth of influence and outlines RIS integration, control, and signaling within standardized network architectures.

III. RIS INTEGRATION IN STANDARDIZED NETWORKS

  • Reflective RISs re-radiate impinging waves rather than emit signals, so they have no explicit coverage area or transmit bandwidth and cannot be integrated into planning like conventional BSs.Their coverage influence depends on RIS capabilities and the sources illuminating the panel, while their frequency influence is bounded by the frequencies they can affect.Source: §III · p. 13
  • The section introduces the RIS area of influence (AoI) and bandwidth of influence (BoI) as concepts relevant to deployment planning.AoI bounds spatial influence on existing coverage; BoI bounds influence in the frequency domain.Source: §III · p. 13

A. Network Architecture

  • Extensions that add active RIS elements for reception or transmission are left for future investigation.Source: A. Network Architecture · p. 13
  • RIS controllers can configure panel states from external instructions or operate autonomously, with capabilities ranging from basic configuration to smart algorithms.The architecture also identifies up to eight L1/L2 signaling types, including control messages and pilot signals between the RIS, gNB, and UE.Source: A. Network Architecture · pp. 14–15
  • The proposed architecture connects RISs to the gNB and UE over new air interfaces and channels, with L1/L2 signaling intended for future NR standards.The example has dedicated controllers for RIS panels, with transmitter and receiver modules for in-band signaling.Source: A. Network Architecture · p. 14, Figure 6 · p. 14
  • A representative TDD procedure has the gNB sound channels across synchronized RIS beam configurations, jointly select BS/RIS beamforming, then configure the RISs for data transmission.The example assumes uplink-downlink channel reciprocity.Source: A. Network Architecture · p. 15

RAN architecture perspective, the figure depicts the near-RealTime (near-RT) RAN Intelligent

The proposed RIS-enabled architecture connects conventional network blocks with RIS-oriented entities through F1-x, R2, and Ro interfaces. It supports both operator-owned, directly controlled RISs and independent deployments integrated on request.

RAN architecture perspective, the figure depicts the near-RealTime (near-RT) RAN Intelligent

  • The near-RT RIC integrates applications through the standardized E2 interface, while the non-RT RIC exchanges messages with it over A1.The architecture envisions protocols that can be optimized for reflective, transmissive, and simultaneous reflecting-and-sensing RISs.Source: RAN architecture perspective, the… · p. 17
  • F1-x, R2, and Ro connect conventional network architecture blocks with new RIS-oriented entities.The architecture is presented as an open environment that can accommodate different RIS types.Source: RAN architecture perspective, the… · p. 17
  • A directly connected RIS lets the operator jointly optimize base-station transmission and RIS reflective beamforming.In this deployment, the network operator owns the RIS; the base station control plane can trigger phase configurations through the RIS actuator.Source: RAN architecture perspective, the… · p. 17
  • An independently deployed RIS can be integrated dynamically and used by operators upon request through a dedicated xApp and RIS-controller counterpart.The RIS controller configures the actuator and panel, and interacts with the near-RT RIC.Source: RAN architecture perspective, the… · p. 17

B. The RIS Bandwidth of Influence (BoI) Challenge

The section defines the RIS bandwidth of influence (BoI) using unit-cell contrast across frequency, and the area of influence (AoI) as a geographical coverage change. It notes that AoI boundaries depend on the performance metric and require planning tools to visualize.

B. The RIS Bandwidth of Influence (BoI) Challenge

  • For a uniform RIS, a unit cell’s BoI suffices to characterize the whole surface because its frequency range cannot exceed that of the cell.The BoI is the frequency range in which the cell has significant adjustable influence on incoming and outgoing waves; the threshold is set to Cmin = 1.Source: B. The RIS Bandwidth of Influence… · pp. 18, 20
  • The cell’s reflection and transmission contrasts compare its S-parameters across feasible states, with contrasts near zero indicating non-reconfigurable behavior.The paper defines maximum contrasts CR(f) and CT(f), and uses the applicable contrast to mark the BoI frequency range.Source: B. The RIS Bandwidth of Influence… · p. 18

C. The RIS Area of Influence (AoI) Challenge

  • An RIS’s AoI is the geographical area where it modifies base-station coverage for a service and target QoS, separating improved coverage from degraded coverage.The shape and location are not intuitive and depend partly on the performance metric; ray-tracing planning tools are needed to visualize AoI and help choose deployment locations.Source: C. The RIS Area of Influence (AoI)… · p. 20, O2 BoI · p. 21

D. Operators Co-Existence with RISs

The paper shows that an RIS’s bandwidth of influence can disrupt another operator when it overlaps that operator’s spectrum, producing channel-state mismatch and erroneous reception. It also characterizes several RIS unit cells, reporting a 6.7 GHz BoI for a transmissive design and a 1.3 GHz BoI for a varactor-based reflective design.

D. Operators Co-Existence with RISs

  • An RIS with a BoI overlapping another operator’s spectrum can cause unpredictable reflections, stale CSI-based link adaptation, a sudden SNR drop, and erroneous packet reception.The authors conclude that characterizing RIS BoI matters for equipment design and fabrication to support operator coexistence.Source: D. Operators Co-Existence with RISs · pp. 21–23

IV. BANDWIDTH-OF-INFLUENCE EXPERIMENTAL CHARACTERIZATION

  • The experimental BoI characterization covers a 1-bit transmissive RIS and varactor-, RF-switch-, and PIN-diode-based reflective RIS unit cells.For each implementation, the characterization examines frequency-dependent contrast and BoI at Cmin = 1.Source: §IV · p. 23

A. BoI of an 1-bit PIN-Diode-based Transmissive RIS

B. BoI of a Varactor-based Reflective RIS

  • 1.3 GHz is the BoI width for the varactor-based reflective RIS at Cmin = 1, over approximately [5.1, 6.4] GHz.The unit cell uses voltage-controlled varactors with capacitance values from 0.1 to 0.9 pF.Source: B. BoI of a Varactor-based Reflective… · p. 25

C. BoI of an RF-Switch-based Reflective RIS

The RF-switch design provides seven phase configurations near 5.3 GHz but a narrow BoI, while the 1-bit PIN-diode design maintains near-opposite phase states over a broader band.

C. BoI of an RF-Switch-based Reflective RIS

D. BoI of an 1-bit PIN-Diode-based Reflective RIS

E. BoI Comparison among Different RIS Implementations

Different RIS hardware designs produce different bandwidths of influence, so the hardware choice should match the deployment scenario and use case. The paper also characterizes how RIS deployment changes energy-efficiency and coverage across a building.

E. BoI Comparison among Different RIS Implementations

  • The varactor-based and PIN-diode-based RISs have the largest BoIs, while the RF-switch-based R-RIS has the smallest.The authors conclude that BoI should be optimized for the intended deployment and use case; a larger BoI is not always preferable.Source: E. BoI Comparison among Different RIS… · pp. 30–31
  • For RF-switch and varactor designs, BoI depends on the patch’s narrow-band resonance or on cell size, respectively.The comparison attributes their different contrast-versus-frequency behavior to their distinct reconfiguration mechanisms.Source: E. BoI Comparison among Different RIS… · p. 31

V. AREA-OF-INFLUENCE EXPERIMENTAL CHARACTERIZATION

  • RIS area-of-influence scenarios cover UE energy efficiency and exposure, localization, and spectral-efficiency boosting.The experimental characterization section examines reflective-RIS AoI for different QoS requirements.Source: §V · pp. 31–32

A. Example of an EE-/S-EMFEU-Optimized AoI

B. Example of an Uplink SE-Optimized AoI

The paper maps RIS influence for uplink spectral efficiency and localization, distinguishing regions where the RIS enables service, boosts performance, or leaves it unchanged.

B. Example of an Uplink SE-Optimized AoI

  • Where propagation is too weak to meet minimum voice performance, the UE remains out of coverage and does not transmit.The comparison uses heatmaps of uplink SE without and with jointly optimized BS receive and RIS reflective beamforming.Source: B. Example of an Uplink SE-Optimized AoI · p. 34
  • The SE-optimized AoI comprises the boosted area and the enabled area, where the RIS respectively improves SE or provides voice and potentially data coverage.The scenario considers voice plus best-effort data, with the UE transmitting at maximum power and adapting its data rate.Source: B. Example of an Uplink SE-Optimized AoI · pp. 34–35

C. Example of a Localization-Optimized AoI

  • The localization scenario adds one RIS-reflected near-field path to the direct path, but only when the BS nearest the RIS transmits.The other two BSs contribute only direct paths.Source: C. Example of a Localization-Optimized… · p. 36
  • RIS-boosted localization improves PEB by 3 dB or more in regions where the three-BS system already provides acceptable performance.The RIS can also enable localization where the conventional system fails the assumed 0.1 m target; areas with less than 2 dB improvement are labeled unchanged.Source: C. Example of a Localization-Optimized… · pp. 36–37

D. Example of an SSE-Optimized AoI

The SSE-optimized AoI example shows how RIS deployment changes secrecy-rate coverage around an eavesdropper, including regions where service is enabled, boosted, or unchanged. The conclusions also report that RIS influence metrics depend on hardware and placement and are characterized through preliminary simulations and measurements.

D. Example of an SSE-Optimized AoI

  • The example optimizes BS precoding and artificial-noise covariance, RX combining, RIS passive beamforming, and stream count for perfect or imperfect Eve-CSI.The illustrated setup uses 8 BS antennas, 4 antenna elements each at Eve and RX, and 400 RIS unit cells.Source: D. Example of an SSE-Optimized AoI · pp. 38–39
  • A 26 bits/sec/Hz minimum SSE threshold marks below-threshold locations as out of coverage, with SSE set to zero in the thresholded map.The resulting AoI distinguishes RIS-enabled service, RIS-boosted performance, and unchanged out-of-coverage areas.Source: D. Example of an SSE-Optimized AoI · p. 39
  • The SSE-optimized AoI separates RIS-enabled, boosted, and unchanged areas, distinguishing newly served locations from improved or still-uncovered locations.The enabled area meets the guaranteed minimum SSE; the unchanged area remains outside BS coverage.Source: D. Example of an SSE-Optimized AoI · p. 39
  • Without RIS, SSE falls to 16.69 bits/sec/Hz above Eve, while secrecy rate is largest near the BS.The reported low-SSE region is where yRX > yEve.Source: D. Example of an SSE-Optimized AoI · p. 39
  • With RIS, SSE rises by at least 2 bits/sec/Hz near Eve and at least 10 bits/sec/Hz near the RIS; the maximum difference, 12.58 bits/sec/Hz, occurs near the RIS.The improvement is reported in the upper sub-area above Eve.Source: D. Example of an SSE-Optimized AoI · p. 39

VI. CONCLUSIONS

  • RIS deployment scenarios are categorized by expected achievable performance and aim for substantial improvement over conventional network architectures.The conclusions describe these scenarios as identified within the ongoing EU H2020 RISE-6G project.Source: §VI · pp. 39–40
  • The paper introduces and mathematically defines BoI and AoI, then characterizes them through preliminary simulations and measurements.The conclusions identify hardware technology, design, and placement as relevant to RIS influence.Source: §VI · p. 40
  • AoI depends on RIS hardware design and placement and can be dynamically optimized for specified performance objectives.Source: §VI · p. 40
Loading 2303.08505v1…