PaperlayerView exact revision on arXiv

Source-linked AI summary

Towards Smart and Reconfigurable Environment: Intelligent Reflecting Surface Aided Wireless Network

Qingqing Wu, Rui Zhang

arXiv:1905.00152v5cs.ITcs.NI

TL;DR

5G networks still face high complexity, hardware cost, energy consumption, and largely uncontrollable signal propagation. This article overviews intelligent reflecting surface technology and reports that doubling the number of reflecting elements can provide about 6 dB gain in representative scenarios.

  • Problem

    5G networks still face high complexity, hardware cost, energy consumption, and largely uncontrollable wireless propagation despite adapting to changing environments.

  • Method

    The article overviews IRS applications, advantages, hardware architecture, signal models, and challenges in designing IRS-aided hybrid wireless networks.

  • Results

    Doubling the number of IRS reflecting elements from 150 to 300 reduces required base-station transmit power by about 6 dB for the same user SNR.

  • Takeaways & Limitations

    IRS provides a controllable wireless-environment degree of freedom and can enhance link performance through fine-grained three-dimensional passive beamforming.

Abstract

from arXiv · show

Although the fifth-generation (5G) technologies will significantly improve the spectrum and energy efficiency of today's wireless communication networks, their high complexity and hardware cost as well as increasingly more energy consumption are still crucial issues to be solved. Furthermore, despite that such technologies are generally capable of adapting to the space and time varying wireless environment, the signal propagation over it is essentially random and largely uncontrollable. Recently, intelligent reflecting surface (IRS) has been proposed as a revolutionizing solution to address this open issue, by smartly reconfiguring the wireless propagation environment with the use of massive low-cost, passive, reflective elements integrated on a planar surface. Specifically, different elements of an IRS can independently reflect the incident signal by controlling its amplitude and/or phase and thereby collaboratively achieve fine-grained three-dimensional (3D) passive beamforming for signal enhancement or cancellation. In this article, we provide an overview of the IRS technology, including its main applications in wireless communication, competitive advantages over existing technologies, hardware architecture as well as the corresponding new signal model. We focus on the key challenges in designing and implementing the new IRS-aided hybrid (with both active and passive components) wireless network, as compared to the traditional network comprising active components only. Furthermore, numerical results are provided to show the potential for significant performance enhancement with the use of IRS in typical wireless network scenarios.

I. INTRODUCTION

IRS is proposed to address the largely random and uncontrollable wireless propagation environment through software-controlled reflection by many low-cost passive elements. The article overviews IRS applications, advantages, architecture, signal modeling, and challenges in designing hybrid active–passive wireless networks.

  • Motivation: 5G technologies have largely achieved targeted capacity and connectivity goals, but their complexity, hardware cost, and energy consumption remain crucial issues.The cited enabling technologies include ultra-dense networks, massive MIMO, and millimeter-wave communication.
  • IRS Technology: IRS reconfigures the wireless propagation environment with software-controlled reflection from many low-cost passive elements.Each element independently changes the incident signal’s amplitude and/or phase, enabling collaborative fine-grained three-dimensional reflect beamforming.
  • Applications: IRS applications include bypassing blocked links to create virtual line-of-sight connections, supporting low-power device-to-device transmissions, and enabling wireless information and power transfer.These applications use intelligent reflection for coverage extension, interference mitigation, and compensation of long-distance power loss.
  • Implementation Advantages: IRSs offer low-profile, lightweight, conformal, and easily deployable hardware that can complement existing cellular or WiFi systems.Deployment on walls, ceilings, building facades, and advertisement panels does not require changing existing wireless standards.
  • Competitive Advantages: Unlike active relays, IRS uses no active transmit module and instead passively reflects received signals as an array.The article positions IRS among technologies related to active relays, backscatter communication, and active-surface massive MIMO.
  • Article Scope: The article examines IRS-aided networks that combine active BS, AP, and user-terminal components with passive IRS components, creating distinct design challenges.Its overview covers signal modeling, hardware architecture, passive beamforming, channel acquisition, and node deployment, along with potential solutions.

II. SIGNAL MODEL AND HARDWARE ARCHITECTURE

This section presents a general signal model for IRS reflection and examines the hardware implementation constraints that shape practical reflection coefficients.

  • It introduces a general signal model describing IRS reflection.
  • It discusses the hardware implementation of IRS.
  • It explains the resulting practical constraints on IRS reflection-coefficient design.

A. Signal Model

The IRS signal model represents each reflected path as a dyadic backscatter channel formed by concatenating the BS–IRS link, IRS reflection, and IRS–user link. Each IRS element independently controls reflection amplitude and phase, enabling spatial signal enhancement or cancellation.

  • A. Signal Model: Each IRS element’s composite BS-to-user path concatenates the BS–IRS link, IRS reflection, and IRS–user link, forming a dyadic backscatter channel unlike a direct channel.The channel resembles keyhole or pinhole propagation.
  • A. Signal Model: The nth element reflects incident signal x_n as y_n = β_ne^jθ_nx_n, where β_n ∈ [0, 1] controls amplitude and θ_n ∈ [0, 2π) controls phase.N denotes the total number of IRS elements.
  • A. Signal Model: Adjusting reflection coefficients spatially enables received-power maximization in dead zones or signal/interference cancellation.Maximum-power reflection uses β_n = 1 for all elements, whereas cancellation may use unequal amplitudes below the maximum.

B. Hardware Architecture

IRS hardware uses a digitally controllable metasurface: a planar array of subwavelength meta-atoms, integrated with layered signal-interaction, shielding, control, and smart-controller components. Element-level bias control independently configures reflection phase shifts, while the controller coordinates with other network components through separate wireless links.

  • Metasurface: IRS is built as a planar array of many electrically subwavelength meta-atoms made from digitally controllable 2D metamaterial.Meta-atoms can be designed through their geometry, size, and orientation.
  • Layered architecture: A typical IRS has three layers: metallic patches on a dielectric substrate, a copper plate preventing signal leakage, and an inner control circuit board.A smart controller is integrated with this layered architecture.
  • Smart controller: An FPGA can serve as the smart controller and gateway, coordinating with BSs, APs, and user terminals through separate wireless links for low-rate information exchange.The controller communicates with and coordinates other network components.
  • Element control: Each element can embed a PIN diode whose DC bias switches it between On and Off states, generating a phase-shift difference of π in rad.Independent element phase shifts are realized by setting the corresponding biasing voltages.

C. Discrete Amplitude and Phase-shift Model · III. MAIN DESIGN CHALLENGES

The section contrasts advantageous continuous IRS amplitude and phase tuning with its costly, potentially unscalable implementation, then identifies key signal-processing and communication challenges for IRS-aided networks.

  • C. Discrete Amplitude and Phase-shift Model: Continuous tuning of each IRS element’s reflection amplitude and phase shift is advantageous for communication applications.
  • C. Discrete Amplitude and Phase-shift Model: Implementing continuous tuning is costly because high-precision elements require sophisticated design and expensive hardware.
  • C. Discrete Amplitude and Phase-shift Model: Such high-precision implementation may not scale as the number of IRS elements becomes very large.
  • C. Discrete Amplitude and Phase-shift Model: 16 phase-shift levels require log2 16 = 4 PIN diodes integrated into each element.
  • III. MAIN DESIGN CHALLENGES: The design challenges extend beyond hardware to signal processing and communication aspects.
  • III. MAIN DESIGN CHALLENGES: Passive beamforming design is a central challenge in IRS-aided wireless networks.
  • III. MAIN DESIGN CHALLENGES: IRS channel acquisition is another key challenge in designing and implementing IRS-aided wireless networks.
  • III. MAIN DESIGN CHALLENGES: IRS deployment is also identified as a key design and implementation challenge.

A. Passive Beamforming Design

Passive beamforming design is challenging because discrete IRS amplitude and phase shifts make optimization NP-hard at large N. Practical methods relax and quantize these variables, while joint active-passive design can enhance desired signals and suppress interference, with received power scaling as O(N^2) in single-user systems.

  • Discrete passive beamforming: Discrete amplitude and phase-shift levels make IRS passive-beamforming optimization exponentially complex and NP-hard as the number of elements N grows.The difficulty arises from searching over discrete values for each IRS element.
  • Discrete passive beamforming: Relaxing amplitude and phase-shift constraints and quantizing continuous solutions reduces computation to polynomial orders of N but introduces quantization loss.Performance loss depends on the number of quantization levels and N.
  • Joint active-passive design: IRS reflection must generally be jointly designed with active transmit beamforming, such as BS beamforming, to optimize network performance.When the direct BS-user link is severely blocked, the BS should point toward the IRS to maximize reflected signal for the user.
  • Joint active-passive design: In multiuser systems, IRS beamforming enhances desired signals and suppresses multiuser interference, while frequency-selective channels further complicate joint optimization.Reflection coefficients must balance channels across different frequency subbands.
  • Performance scaling: O(N^2) received-power scaling occurs asymptotically in IRS-aided single-user systems as N →∞, corresponding to about 6 dB power gain for every doubling of N.This squared law combines the IRS’s O(N) power gain from reflected beamforming with coherent signal combination.

B. IRS Channel Acquisition

Accurate BS–IRS and IRS–user channel knowledge is generally required for IRS passive-beamforming gains. Channel acquisition depends on whether IRS elements have receive RF chains, with feedback-based reflection design available when they do not.

  • B. IRS Channel Acquisition: Accurate knowledge of channels between the IRS, BSs, and users is required to realize passive-beamforming performance gains.With the IRS in absorbing mode (β_n = 0 for all n), conventional estimation methods obtain BS–user-link CSI without the IRS.
  • Receive RF Chains: With low-power receive RF chains, IRS elements estimate BS/user-to-IRS channels from training signals.Under TDD, channel reciprocity can be leveraged for channel acquisition.
  • No Receive RF Chains: Without receive RF chains, the IRS cannot estimate channels with the involved BSs and users directly.This makes explicit IRS–BS/user channel estimation infeasible at the IRS.
  • No Receive RF Chains: When receive RF chains are absent, reflection coefficients can instead be designed from BS/user feedback on received IRS-reflected signals.A codebook-based method sweeps predefined reflection coefficients and selects the best beam using received-signal feedback.

C. IRS Deployment

IRS deployment in hybrid wireless networks requires different considerations from active-BS or relay placement because IRSs provide shorter-range, local coverage. Key challenges include selecting locations that support multiuser transmissions, obtaining deployment-related CSI efficiently, and characterizing the capacity gains of adding passive IRSs.

  • Deployment considerations: IRS deployment must account for its shorter operating range and local-coverage role, unlike deployment of active BSs or relays.The deployment problem concerns hybrid networks combining active BSs and passive IRSs.
  • Single-cell deployment: A clear BS–IRS LoS path can maximize received signal power, but a single LoS path may create low-rank MIMO channels for simultaneous transmissions.Thus, straightforward LoS-based placement may not work well when an IRS serves multiple users simultaneously.
  • Practical deployment optimization: In complicated environments with multiple associated BSs, heuristics may be ineffective, while exhaustive placement search requires global CSI at all candidate locations.Ray-tracing can estimate the needed CSI but is computationally costly and requires site-specific information such as building or floor layouts.
  • Multi-cell capacity: The multi-cell deployment problem includes determining the fundamental capacity limit of IRS-aided wireless networks.This question arises because increasing spatial density improves capacity in active-BS networks but can also increase interference.
  • Multi-cell capacity: Adding passive IRSs is expected to significantly improve network capacity by enhancing locally covered users without increasing network interference.Increasing passive-IRS density may therefore avoid the capacity degradation associated with interference growth in conventional active-only networks.

IV. NUMERICAL RESULTS AND DISCUSSION

The numerical results evaluate the proposed IRS for creating signal hotspots and interference-free zones. The scenario models a multi-antenna BS, an N-element IRS, and a single-antenna user, with a line-of-sight-dominated BS–IRS channel under proper deployment.

  • Evaluation objectives: The evaluation targets IRS-enabled creation of signal hotspots and interference-free zones.These objectives are illustrated in Fig. 1.
  • System setup: The considered setup contains a BS with M antennas, an IRS with N elements, and one single-antenna user.Their locations are specified in Fig. 4(a).
  • Channel and geometry: The BS–user horizontal separation is denoted by d meter (m), and proper deployment makes the BS–IRS channel line-of-sight dominated.The passage attributes this dominance to the LoS link.

A. Signal Power Enhancement and Scaling Law

The section evaluates IRS-assisted signal power enhancement when reflected and direct signals combine constructively, and shows that ideal continuous-phase operation yields an approximately O(N^2) reduction in required BS transmit power as reflecting elements increase.

  • Signal power enhancement: The evaluation considers a user served by the BS with IRS assistance, requiring the IRS-reflected signal to add constructively with the direct BS-user signal.Four schemes are compared under M = 5 and N = 40, including joint optimization of active BS transmit and passive IRS reflect beamforming.
  • Scaling law: For ideal continuous phase (b = ∞), BS transmit power scales down with the number of reflecting elements approximately as O(N^2).The experiment assumes constant reflection amplitude equal to one, uses a practical b-bit uniformly quantized phase shifter otherwise, and sets d = 50 m.

B. Interference Suppression · V. CONCLUSIONS

The section demonstrates IRS-based interference suppression and concludes that IRSs can create smart, reconfigurable wireless environments by dynamically adapting reflection coefficients to sensed conditions.

  • B. Interference Suppression: A neighboring transmitter at d = 50 m creates co-channel interference, which the IRS is deployed to help suppress at the user.The setup uses the BS in Fig. 4(a) as the interfering transmitter and corresponds to the scenario in Fig. 1(c).
  • B. Interference Suppression: The interference-suppression setup also represents a physical-layer security case in which the IRS helps cancel a legitimate transmitter’s signal at an eavesdropper.Here, the user is treated as the eavesdropper whose received signal from the legitimate BS needs to be canceled.
  • B. Interference Suppression: The interference experiment assumes M = 1 and a BS transmit power of 30 dBm.These assumptions are specified for the comparison of interference power across the evaluated schemes.
  • B. Interference Suppression: Interference power normalized by noise power is plotted against N for three schemes, including jointly optimized continuous amplitude and phase shifts.The comparison is presented in Fig. 4(d).
  • V. CONCLUSIONS: The article surveys IRS technology as a means of achieving smart and reconfigurable environments in future wireless networks.The conclusion characterizes IRS technology as promising while noting that IRS-aided networks remain largely unexplored.
  • V. CONCLUSIONS: IRSs can sense the wireless environment and dynamically adjust reflection coefficients to perform different functions using signal processing and machine learning.The conclusion identifies dynamic coefficient adjustment as a mechanism for adapting IRS behavior to the wireless environment.
Loading 1905.00152v5…