PaperlayerView exact revision on arXiv

Source-linked AI summary

Integrated Sensing and Communications with Reconfigurable Intelligent Surfaces

Sundeep Prabhakar Chepuri, Nir Shlezinger, Fan Liu, George C. Alexandropoulos, Stefano Buzzi, Yonina C. Eldar

arXiv:2211.01003v1eess.SP

TL;DR

High-frequency ISAC faces severe propagation loss, while RISs can modify harsh propagation environments. This article overviews RIS-aided sensing and shows that joint ISAC benefits increase with channel coupling, which RISs can control.

  • Problem

    High-frequency ISAC is constrained by severe propagation loss that can make non-line-of-sight paths too weak for practical communication and sensing.

  • Method

    The article provides an overview of RIS applications for sensing and examines joint ISAC designs through the coupling between communication and sensing channels.

  • Results

    Joint sensing and communications designs provide greater gains as communication and sensing channels become more coupled, while RISs provide means to control that coupling.

  • Takeaways & Limitations

    RISs can facilitate efficient control of beneficial channel coupling in RIS-aided ISAC systems.

Abstract

from arXiv · show

Integrated sensing and communications (ISAC) are envisioned to be an integral part of future wireless networks, especially when operating at the millimeter-wave (mmWave) and terahertz (THz) frequency bands. However, establishing wireless connections at these high frequencies is quite challenging, mainly due to the penetrating pathloss that prevents reliable communication and sensing. Another emerging technology for next-generation wireless systems is reconfigurable intelligent surfaces (RISs), which are capable of modifying harsh propagation environments. RISs are the focus of growing research and industrial attention, bringing forth the vision of smart and programmable signal propagation environments. In this article, we provide a tutorial-style overview of the applications and benefits of RISs for sensing functionalities in general, and for ISAC systems in particular. We highlight the potential advantages when fusing these two emerging technologies, and identify for the first time that: i) joint sensing and communications designs are most beneficial when the channels referring to these operations are coupled, and that ii) RISs offer means for controlling this beneficial coupling. The usefulness of RIS-aided ISAC goes beyond the individual obvious gains of each of these technologies in both performance and power efficiency. We also discuss the main signal processing challenges and future research directions which arise from the fusion of these two emerging technologies.

I. INTRODUCTION

RISs modify harsh wireless propagation environments, addressing severe high-frequency pathloss that can undermine communication and sensing. This article reviews RIS-aided sensing and ISAC, emphasizing channel coupling, controllable by RISs, as a source of joint-design gains.

  • RISs use programmable scattering and tunable reflection patterns to reshape propagation, create additional paths, and support sensing in harsh or NLoS environments.
  • Severe pathloss at mmWave and THz frequencies can prevent reliable communication and, in some cases, sensing.
  • The article surveys RIS applications for sensing and ISAC, including monostatic and bistatic configurations, beamforming, signal models, challenges, and research directions.
  • Joint sensing-and-communications designs are more beneficial when their channels are coupled, and RISs provide a means to control that coupling.
  • The article presents RIS-empowered ISAC designs targeting high-rate communications, accurate remote-target estimation, prescribed SINR, and sensing waveforms with good cross-correlation.

II. FUNDAMENTALS OF RISS

RISs are programmable metasurfaces whose electromagnetic responses can reshape propagation, with passive, active, and hybrid architectures offering different sensing and control capabilities. Hybrid designs add limited reception capability while retaining desired reflection control and low-power operation.

  • Hardware architectures: RISs use programmable electromagnetic elements to control scattering and reflection patterns, typically through externally configured meta-atoms.The article focuses primarily on passive RISs, whose elements do not amplify impinging signals.
  • Hardware architectures: Active metasurface antennas differ fundamentally from passive RISs because they include dedicated RF chains and signal amplifiers.The distinction separates reconfigurable scattering surfaces from active transceiver architectures.
  • Hardware architectures: Passive RISs require an external receiver or controller because their elements do not sense impinging signals.This dependency creates design challenges for sensing and configuration.
  • Hybrid architectures: Hybrid passive-active RISs combine response-reconfigurable meta-atoms, waveguide coupling, and reception RF chains to support reflection shaping and local AOA estimation.The reported architecture achieves desired reflection beampatterns while estimating impinging-signal angles of arrival.
  • Hardware architectures: RISs can use metasurfaces on dielectric substrates or alternative waveguide-based structures that permit limited coupling to the elements.Their planar construction can facilitate deployment and flexible form factors.
  • Control and power: RIS control circuitry can configure individual or grouped meta-atoms discretely with PIN diodes or continuously with varactors.The control circuitry consumes power, but the elements themselves typically operate at the microwatt scale.

Capabilities:

RISs can reshape wireless propagation for communications, localization, sensing, and other electromagnetic applications while operating with lower hardware complexity than active relays. Their passive operation also imposes dependence on external control and simplified modeling assumptions that may not capture practical element behavior.

  • Communications capabilities: RISs can operate as full-duplex passive relays, offering deployment flexibility and potentially lower cost than active-relay implementations.Their limited control circuitry and lack of RF chains contribute to this architecture-level advantage.
  • Control limitations: Passive RISs cannot sense incoming signals, so an external device must know relevant environmental or transmitter information and configure their reflection patterns.The controller may be a base station or dedicated device connected through a control link.
  • Modeling limitations: The common phase-shifter model can miss coupling between phase, attenuation, incident angle, frequency selectivity, and multipath reflections.The simplified model remains useful for signal-processing analysis despite these mismatches.
  • Other use cases: RISs support applications including wireless power transfer, interference mitigation, coverage enhancement, energy efficiency, and communication reliability.They can focus energy toward intended users while reducing radiation in undesired directions.
  • Sensing and localization: RIS-aided localization can improve accuracy, support NLoS operation, and enable localization with fewer transmitting terminals or no available access point.RIS-generated reflections provide additional paths for positioning.
  • ISAC use cases: RISs are also studied for joint sensing and communications, including spectrum-sharing and dual-function ISAC systems.The article focuses on RIS-empowered ISAC applications.

C. Signal Modeling

The signal model represents RIS-controlled sensing and communications channels using a configurable reflection vector and dual-function transmitted signals. RIS-aided sensing can create useful paths and improve target illumination in harsh or NLoS settings, but suffers severe pathloss when direct paths are absent.

  • Signal model: The model uses an RIS reflection vector to parameterize both stochastic sensing and communications channels for a common transmitted signal.The transmitted vector can carry a radar waveform, a communication message, or both.
  • Signal model: The communications channel is modeled with cascaded Tx-RIS and RIS-Rx links, a diagonal tunable-phase matrix, and additive white Gaussian noise.The simplified model commonly constrains RIS responses to unit-modulus phase variations.
  • RIS-aided sensing: RIS-generated phase profiles can improve target power and detection capability, especially in harsh and NLoS propagation settings.RISs can introduce virtual paths or intentional geometric delays when direct visibility is absent.
  • RIS-aided sensing: A forward RIS can make the direct and RIS-assisted paths comparable in strength for suitable placements, improving target illumination power.RIS-assisted paths are otherwise weaker because they experience double pathloss.
  • RIS-aided sensing: When direct paths are absent and sensing relies only on RISs, signals undergo inevitable quadruple pathloss, making transmit beamformer design critical.The beamformer must provide sufficient power to illuminate the target despite the additional propagation losses.
  • RIS-aided sensing configurations: RIS-aided sensing configurations include monostatic systems and bistatic systems using forward RISs, backward RISs, or only a forward RIS.These configurations create additional propagation paths for transmitted signals and target echoes.

A. Sensing Signal Model

The sensing model combines direct and RIS-mediated propagation with a precoded unit-power waveform, while jointly designing the transmit beamformer and RIS phases to maximize target illumination power. The RIS introduces additional propagation paths and gains, but the resulting design is non-convex and is solved by alternating optimization.

  • Signal and channel model: The model uses a unit-power sensing waveform precoded by a transmit beamformer, with target directions represented relative to the transmit and RIS arrays.The transmitted signal is x(t) = ws(t), where s(t) has unit power.
  • Signal and channel model: The received sensing signal includes direct and RIS-mediated channels, with the RIS path attenuation determined by the Tx–RIS and RIS–target distances.The RIS channel attenuation is proportional to the product of the squared Tx–RIS and RIS–target distances, whereas the direct path depends on the Tx–target distance.
  • Signal and channel model: The RIS-assisted sensing channel contains four propagation paths, adding channel gains and an extra measurement dimension through the target and RIS reflection angles.The paths include direct and RIS-reflected combinations, and estimating θ1 and θ2 supports target localization.
  • Beamforming design: Transmit beamforming and RIS phase shifts are jointly optimized to maximize target illumination power under a transmit-power constraint and unit-modulus RIS phases.The single-target formulation can be extended to multiple targets by maximizing worst-case illuminated power.
  • Beamforming design: Because the beamformer and RIS phases are coupled in a non-convex problem, the design is solved by alternating optimization until convergence.The procedure fixes one variable while optimizing the other, then iterates.
  • Beamforming design: Without a direct path, an RIS can provide an N2 beamforming gain, whereas the transmit array contributes an LT array gain.In the no-direct-path case, the transmit beamformer directs energy toward the RIS and the RIS phases align the reflected signal toward the target.

C. Target Detection and Parameter Estimation

RIS-aided sensing detects targets using matched filtering and estimates bearing through standard direction-finding methods, with performance governed by target illumination power. Increasing illumination power improves SNR and lowers the CRB, making RISs particularly useful for NLoS sensing.

  • Detection and estimation: The received sensing signal is processed with a matched filter, whose output SNR depends on the target illumination power.The matched filter is optimal in terms of SNR or detection probability when the received echoes do not depend on the RIS.
  • Target detection: Target presence is tested with a generalized likelihood ratio detector using a threshold selected for a constant false-alarm rate.For nonfluctuating target gain, the detection probability is expressed using the Marcum Q-function.
  • Parameter estimation: Standard direction-finding methods can estimate the target bearing angle, while the CRB provides a lower bound on the variance of unbiased bearing estimators.The CRB also serves as a performance baseline for direction estimation.
  • Parameter estimation: Increasing illuminated power directly improves target identification by lowering the CRB, so RIS-enabled illumination can facilitate sensing in NLoS conditions.The RIS reflection pattern should therefore be designed to maximize illuminated power.

D. Numerical Example

The numerical example shows that RIS beamforming adapts to target location and can preserve or improve sensing performance relative to systems without an RIS. Its clearest benefit occurs when the direct path is weak or blocked, where RISs can enable sensing that would otherwise be impossible.

  • Beamforming patterns: The beam patterns adapt to target location: the RIS reflects toward the target, while the transmit pattern changes according to the relative target, transmitter, and RIS positions.At the blocked point J, the precoder concentrates energy toward the RIS; near point F, it forms prominent beams toward both the RIS and target.
  • CRB and illumination comparison: For N = 100, RIS-aided illumination is comparable to the no-RIS system near the transmitter but improves over it near the RIS and remains available after direct-path blockage.Without an RIS, blocked points have zero illumination power and no sensing capability.
  • CRB and illumination comparison: The RIS-aided system is always at least as good as the no-RIS system in the example, and it enables sensing where the direct path is blocked.For blocked points, the RIS-aided and RIS-only systems coincide, while the no-RIS CRB becomes infinite.
  • CRB and illumination comparison: With N = 400 elements, the RIS provides larger array gain, increasing the RIS-path strength and making RIS-only sensing outperform no-RIS sensing at point F.The corresponding RIS-only CRB is lower than the no-RIS CRB at that point.
  • Implications: The sensing-only results indicate that RISs are mainly beneficial in harsh and NLoS environments, while the subsequent ISAC analysis examines gains beyond NLoS recovery.The paper identifies a fundamental RIS gain when sensing and communications are combined, not restricted to NLoS conditions.

IV. RIS-EMPOWERED ISAC

RIS-empowered ISAC gains arise from coupling between communication and sensing channels, which determines how effectively shared resources support both functions. Stronger coupling improves joint performance, while orthogonal channels provide no ISAC gain over separate systems.

  • S&C Trade-off: ISAC beamforming seeks a Pareto-efficient balance between communication rate and sensing accuracy under shared wireless resources.The design minimizes the angle-estimation CRB subject to a downlink communication QoS constraint.
  • S&C Coupling: Stronger coupling between communication and sensing channels enables higher ISAC performance gains.The weakly-coupled, moderately-coupled, and strongly-coupled cases correspond to increasing channel correlation and progressively better joint performance.
  • S&C Coupling Effect versus ISAC Beamforming: Fully aligned sensing and communication subspaces let both functionalities achieve their best performance without a trade-off.For feasible communication rates up to Rc, resources are fully reused and the ISAC gain is maximized.
  • S&C Coupling Effect versus ISAC Beamforming: Orthogonal sensing and communication subspaces provide no ISAC gain because their resources cannot be reused.The resulting resource efficiency matches that of separate sensing and communication systems.
  • S&C Coupling Effect versus ISAC Beamforming: Intermediate channel coupling produces partial signal-power reuse and an ISAC gain between the aligned and orthogonal extremes.Increasing correlation between the channels yields significant performance gains.

B. How Do RISs Improve ISAC Performance?

RISs improve ISAC by adding propagation paths and by tuning their phases to increase correlation between communication and sensing channels. These mechanisms expand and rotate the corresponding subspaces, improving the joint performance trade-off.

  • How RISs Improve ISAC Performance?: RIS-aided ISAC combines subspace expansion from added channel paths with subspace rotation from phase-controlled coupling.Together, these mechanisms explain the joint gains beyond improving the individual sensing and communication links.
  • RIS-Aided Channel Coupling Effect: RISs provide additional propagation paths that enhance channel gains for both sensing and downlink communications.The extra paths expand the respective communication and sensing subspaces.
  • RIS-Aided Channel Coupling Effect: RIS phase control can increase communication-sensing correlation, allowing more signal power to be reused by both functions.This artificial coupling improves ISAC efficiency and the performance trade-off.
  • RIS-Aided Channel Coupling Effect: Proper RIS placement and reflection-pattern tuning can transform weakly coupled direct channels into moderately coupled channels.The resulting beams are directed toward separated communication-user and sensing-target clusters.

C. RIS-Aided ISAC Beamforming based on Subspace Expansion and Rotation

The RIS-aided beamforming design first shapes the RIS to expand and rotate the sensing and communication subspaces, then designs a dual-functional transmit beamformer. Numerical results show that both mechanisms improve the ISAC trade-off, especially for weakly coupled channels.

  • RIS-Aided ISAC Beamforming: The two-step design first maximizes channel correlation and gains through RIS phase optimization, then designs the transmit beamformer for simultaneous sensing and communication.The beamformer minimizes sensing CRB subject to a communication-rate constraint.
  • Optimization: The RIS phase-optimization problem is non-convex because its objective and unit-modulus constraints are non-convex.Gradient descent with projection and semi-definite relaxation are identified as solution approaches for the phase and transmit-beamforming problems.
  • Numerical Example: In strongly coupled channels, RIS deployment mainly improves ISAC by enhancing channel gains through subspace expansion.With correlation coefficient 1, the trade-off curve is rectangular regardless of RIS deployment.
  • Numerical Example: In weakly coupled channels, RIS benefits are more pronounced because phase control also improves communication-sensing correlation through subspace rotation.Fig. 8 separates the gain from enhanced channel gains and the gain from improved channel correlation.
  • Numerical Example: The numerical example demonstrates the effectiveness of RIS-aided ISAC beamforming for the considered weakly and strongly coupled scenarios.The reported gains include both individual S&C improvements and additional joint-operation benefits from controlling channel coupling.

D. RIS-Aided ISAC Beamforming based on Beampattern Errors

RIS-aided ISAC beamforming can jointly control sensing beampatterns, communication performance, waveform correlation, and RIS reflections. The design uses a weighted sensing objective with communication SINR constraints and illustrates the resulting beam-pattern trade-offs.

  • Beampattern Design: The desired beampattern directs sufficient power toward target directions and the RIS while allowing communication and sensing beams to be jointly designed.It may be formed as a superposition of multiple rectangular beams.
  • Joint Beamforming Design: The joint design optimizes beampattern mismatch, reflected-signal cross-correlation, and communication SINR under unit-modulus RIS constraints.The resulting problem is non-convex and is addressed with suboptimal alternating-minimization solvers.
  • Numerical Example: The RIS reflection pattern and the individual sensing and communication beampatterns jointly realize the desired directions and communication-assisted sensing illumination.The numerical evaluation covers the realized beampattern, individual patterns, and RIS phase profile.
  • Numerical Example: The communication beamformer peaks toward the RIS, while the sensing beampattern places a dip there to limit sensing-waveform interference at the user.Communication symbols can still illuminate targets and increase their received power.

V. FUTURE RESEARCH DIRECTIONS

Future work must move RIS-empowered ISAC beyond simplified, static models toward physically compliant, mobile, and hardware-aware designs. Key directions include characterizing fundamental trade-offs, supporting realistic signal processing, and developing deployable RIS hardware and orchestration.

  • Future Research Directions: Future research should further test whether RIS-enabled ISAC gains extend beyond marginal adaptations of established RIS communications results.The article identifies open challenges intended to strengthen its initial observations about RIS-ISAC capabilities.
  • Beyond Simple RIS Models: Physically compliant RIS models are needed because ideal independent phase control and narrowband cascaded channels often fail in multipath, wideband, and rich-scattering environments.Future signal processing should account for finite or binary configurations and frequency-selective element responses.
  • Beyond Simple ISAC Models: Realistic RIS-ISAC processing should incorporate multiple users, interference, fading, practical modulations, moving targets, clutter, sensing waveforms, and transmitter architectures.These factors affect the beamforming capabilities of dual-function transmitters and therefore the design of signal processing methods.
  • Fundamental Limits of RIS-Empowered ISAC: Fundamental-limit studies should characterize how RISs alter the communications–sensing trade-off independently of any particular transmission scheme.Such analyses can rigorously identify the benefits RIS technology brings to ISAC.
  • Mobile ISAC Systems in RIS-Parameterized Settings: Mobile RIS-aided ISAC requires joint trajectory, beamforming, reflection-pattern, and transmission design while accounting for changing channel models.RISs may themselves be mounted on vehicles with unknown position and orientation, making mobility a distinct modeling challenge.
  • RIS Hardware Requirements and System Design: ISAC-specific RIS hardware should support efficient reconfigurability, simultaneous reflection and sensing, and programmable coupling across wideband or multiband signals.Deployment and management research should also study multiple RISs and their joint orchestration.

VI. CONCLUSION

The article surveys RISs for future ISAC systems and identifies when their combination is most useful. It concludes that coupled sensing and communications channels enable joint-design gains, while RISs provide control over that coupling and motivate further signal-processing research.

  • Conclusion: Joint sensing and communications designs are most beneficial when the respective channels are coupled.This conclusion is drawn from a basic MIMO ISAC system model.
  • Conclusion: RISs can facilitate efficient control of the beneficial coupling between sensing and communications channels.The article presents this as a key advantage of integrating RISs with ISAC.
  • Conclusion: The fusion of RISs and ISAC creates signal-processing challenges and motivates several future research directions.The article discusses these challenges as part of its overview of RIS-enabled ISAC.
Loading 2211.01003v1…