PaperlayerView exact revision on arXiv

Source-linked AI summary

Reconfigurable Intelligent Surfaces for N-LOS Radar Surveillance

Augusto Aubry, Antonio De Maio, Massimo Rosamilia

arXiv:2104.00456v2eess.SP

TL;DR

Radar surveillance can miss targets in shadow regions where obstacles block the direct radar–target path. This paper proposes RIS-controlled N-LOS operation and derives associated SNR and clutter models; its analysis assesses the framework for short-range N-LOS scenarios.

  • Problem

    Targets in urban or other shadow regions may lack a direct radar–target path, making N-LOS surveillance a challenging problem.

    Source: §I · p. 1, §III · p. 3

  • Method

    The paper uses RIS-controlled propagation to create a two-way double-hop radar–target channel and derives SNR and surface- and volume-clutter SCR models for the N-LOS mode.

    Source: §I · p. 2, §VIII · p. 11

  • Results

    Larger RIS size and more coherently integrated pulses improve SNR, while a larger RIS area increases far-field distance and control complexity; numerical analysis assesses SNR and detection performance for N-LOS short-range scenarios.

    Source: §VII · p. 10, §VIII · p. 11

  • Takeaways & Limitations

    The numerical analysis corroborates the framework’s capability to handle N-LOS short-range scenarios.

    Source: §VIII · p. 11

Abstract

from arXiv · show

This paper deals with the use of Reconfigurable Intelligent Surfaces (RISs) for radar surveillance in Non-Line Of Sight (N-LOS) scenarios. First of all, the geometry of the scene and the new system concept is described with emphasis on the required operative modes and the role played by the RIS. Then, the specific radar equation (including the RIS effect) is developed to manage the coverage requirements in the challenging region where the LOS is not present. Both noise and clutter interference cases (pulse length-limited and beamwidth-limited surface clutter as well as volume clutter) are considered. Hence, a digression on the use of the radar timeline for the new operative mode is presented together with the data acquisition procedure and the resolution issues for the range, azimuth, and Doppler domains. Finally, the interplay among the system parameters and, in particular, those involving the RIS is discussed and analyzed via numerical simulations.

I. INTRODUCTION

The paper proposes using RISs to extend radar surveillance into areas without a direct radar–target path, and develops the operating concept and analysis for that mode. It also introduces RIS properties and notation used in the discussion.

I. INTRODUCTION

  • RIS-assisted surveillance targets shadow regions where buildings or other obstacles block the direct radar–target path, a challenge in urban and other practical scenarios.Existing strategies described include natural multipath exploitation and radar networks.Source: §I · p. 1
  • A RIS can extend a standard radar’s coverage by creating a controlled two-way, double-hop path to targets lacking direct line of sight.The radar illuminates the RIS, which is configured to direct radiation toward the target and then back toward the radar; alternating standard and N-LOS modes is proposed for full-region coverage.Source: §I · p. 2
  • The paper derives an N-LOS SNR budget and surface- and volume-clutter SCR expressions, then discusses data acquisition, resolution, and performance analysis.After data acquisition, conventional range-Doppler processing, potentially with clutter cancellation and CFAR, can be used.Source: §I · p. 2

A. Notation

  • The notation section specifies conventions for vectors, matrices, transposes, conjugation, Hadamard products, complex-number sets, and norms.It uses bold lowercase and uppercase letters for vectors and matrices, respectively.Source: A. Notation · p. 2

II. GENERALITIES ON RIS

  • RISs are digitally controllable metasurfaces whose programmable units adjust signal responses, enabling changes to incident-wave shape and reflection or refraction behavior.The paper describes RISs as composed of many low-cost passive programmable electronic circuits.Source: §II · p. 2

III. RIS-BASED RADAR OPERATION FOR N-LOS SURVEILLANCE

The proposed N-LOS radar mode uses a controllable RIS to establish a radar–target path in a region without direct line of sight. The radar alternates conventional surveillance with scheduled RIS-assisted scanning.

III. RIS-BASED RADAR OPERATION FOR N-LOS SURVEILLANCE

  • Sub-region 1 has a direct radar–target path, whereas sub-region 2 is shadowed and receives only partial coverage from natural diffraction or multipath under standard radar operation.The RIS is inactive during standard surveillance of sub-region 1.Source: §III · p. 3, Figure 2 · p. 3
  • During scheduled scans of sub-region 2, the radar points toward the RIS, whose settings control the reflected beam toward the surveillance area.The RIS can be configured by a precanned program or communication-oriented software, which adds a consideration for radar scheduling.Source: §III · p. 3
  • In each dwell, the RIS first directs radar radiation into an angular sector, then steers target returns toward the radar during listening across Np pulses.Acquisition and processing occur at the radar and account for the extra two-way path; sequential scanning is one example among possible sweep patterns.Source: §III · p. 3

IV. RADAR EQUATION FOR SUB-REGION 2

The N-LOS radar equation is reformulated to account for the RIS and the two-hop propagation geometry, providing a tool for system sizing. Its derivation assumes far-field placement and tracks illumination and reflection through the RIS patches.

IV. RADAR EQUATION FOR SUB-REGION 2

  • The standard radar equation is insufficient for sub-region 2; a reformulated equation is needed to include RIS effects and support system sizing.Source: §IV · p. 3
  • The model assumes the RIS is in the far-field region of both the radar and target, and identifies the radar, RIS, and target phase-center positions.The system geometry is represented using coordinate systems centered at the radar and RIS.Source: §IV · p. 3, Figure 4 · p. 4
  • The derivation follows radar illumination to the RIS, where patch area and efficiency determine gathered power and the reflected field impinging on the target.The plane-wave spatial power density at the RIS is defined from radar transmit power, gain, radiation pattern, and radar-to-RIS distance.Source: §IV · pp. 4–5

RIS)T ,

The section derives an RIS-assisted radar SNR budget and interprets the RIS reflection pattern, including how beam steering and equivalent-monostatic parameters relate to coverage and resolution.

RIS)T ,

  • The received-power derivation assumes reciprocity: the RIS coefficients remain unchanged on the backward path and radar transmit and receive patterns match.The SNR formulation uses receiver noise figure, bandwidth, and pulse length, and is stated to hold for modulated and unmodulated pulses.Source: RIS)T · p. 5
  • Coherent integration over Np pulses and longer dwell time increase SNR, with dwell time defined as Td = NpT.The SNR model also includes system losses such as transmit, atmospheric, receiver, processing, and RIS losses.Source: RIS)T · pp. 5–6
  • The RIS-induced pattern is maximized by coherently aligning reflected contributions, and its upper bound is (NM)^4 under the stated passive-patch assumptions.The alignment choice compensates the phase terms of S1 ⊙ S2.Source: RIS)T · p. 6
  • The pattern’s directional dependence provides a basis for tiling sub-region 2 with RIS beams while accounting for acceptable angular SNR loss and dwell time.The pattern factors across directional-cosine offsets, with mainlobe widths set by N and M; the section gives broadside 12 dB beamwidths of 0.891/N and 0.891/M when dx = dy = λ0/2.Source: RIS)T · p. 6
  • The equivalent-monostatic azimuth beamwidth is chosen to match cross-range resolution between the equivalent point and the RIS-assisted target location.The matched quantities are φeq(r1 + r2) and φRIS r2.Source: RIS)T · p. 7, Figure 5 · p. 7

V. RADAR EQUATION FOR CLUTTER INTERFERENCE

The section evaluates clutter interference in the N-LOS operative region, considering surface and volume clutter and two surface-clutter geometries.

V. RADAR EQUATION FOR CLUTTER INTERFERENCE

  • The clutter-to-signal analysis for sub-region 2 covers pulse-length-limited and beamwidth-limited surface clutter, as well as volume clutter.The section computes SCR for each considered clutter case.Source: §V · p. 7

A. Surface clutter: pulse length-limited

For pulse-length-limited surface clutter, the illuminated cell depends on RIS azimuth beamwidth and pulse length, and the resulting SCR varies with target distance and beamwidth.

A. Surface clutter: pulse length-limited

  • The illuminated surface area is framed by RIS azimuth beamwidth and pulse length measured along the surface, with grazing angle ψ used in the geometry.The tangent approximation assumes a small RIS azimuth beamwidth.Source: A. Surface clutter: pulse length-limited · p. 7
  • Pulse-length-limited surface-clutter SCR decreases inversely with RIS-to-target distance r2 and RIS azimuth beamwidth φ̄RIS.It does not depend on radar-to-RIS distance r1 or radar transmit/receive beamwidths.Source: A. Surface clutter: pulse length-limited · p. 7

B. Surface clutter: beamwidth-limited

For beamwidth-limited surface clutter under a circular-beam assumption, the SCR has inverse-square dependence on target distance and RIS azimuth beamwidth.

B. Surface clutter: beamwidth-limited

  • Beamwidth-limited surface-clutter SCR decreases as 1/r2^2 and 1/φRIS^2.The illuminated area and clutter-power calculation use a circular beam and the small-angle approximation for RIS azimuth beamwidth.Source: B. Surface clutter: beamwidth-limited · p. 7

C. Volume clutter

The volume-clutter analysis assumes uniformly distributed scatterers in the illuminated volume, then derives clutter power and the resulting SCR. The SCR depends on RIS–target geometry and RIS beamwidths, but not on radar–RIS distance or radar beamwidths.

C. Volume clutter

  • Uniformly distributed scatterers across the illuminated volume are assumed for the volume-clutter model.Source: C. Volume clutter · p. 7
  • The clutter-power expression uses RIS beamwidth geometry, with the stated approximation relying on a small-angle assumption.Source: C. Volume clutter · pp. 7–8
  • The volume-clutter SCR varies inversely with squared RIS–target distance and both RIS beamwidths, while radar–RIS distance and radar beamwidths do not affect it.Source: C. Volume clutter · p. 8

VI. OPERATION IN SUB-REGION 2 AND RADAR ECHO MODEL

Sub-region 2 operation alternates radar transmission and RIS steering with a listening period for data acquisition. The echo model then supports conventional pulse-compression and range-Doppler processing, with range and Doppler resolution tied to waveform bandwidth and dwell time.

VI. OPERATION IN SUB-REGION 2 AND RADAR ECHO MODEL

  • Each burst includes radar transmission, an idle interval, RIS beam steering, and listening for returns before the next pulse.The listening interval lasts 2Rua/c, and the cycle repeats through the dwell.Source: §VI · p. 8, Figure 6 · p. 9
  • A target echo returns toward the RIS as a delayed, frequency-shifted transmit burst, which programmable reflecting coefficients focus back toward the radar under a stop-and-hop assumption.Source: §VI · p. 8
  • The received signal is down-converted and matched-filtered for pulse compression; Doppler-tolerant waveforms permit the stated ambiguity-function simplification.Source: §VI · p. 8
  • Range resolution is c/(2B) and Doppler resolution is 1/(NpT); target range includes the fixed radar–RIS bias and RIS–target distance.Azimuth resolution depends on RIS size and physical/geometrical properties and radar wavelength.Source: §VI · pp. 8–9
  • The pulse-compressed received signal fills a fast-time/slow-time data matrix used for subsequent processing such as MTI, pulse-Doppler, or CFAR.Source: §VI · p. 9

VII. NUMERICAL RESULTS

Simulations show that SNR improves with more RIS elements and coherently integrated pulses, although larger RISs increase far-field distance and control complexity. The reflected pattern peaks along the pointing direction, while steering off-boresight widens its beam.

VII. NUMERICAL RESULTS

  • The simulations consider square RISs with N = M ∈ {101, 133, 201}; the stated 10 MHz bandwidth yields 15 m range resolution and limits RIS size under the narrowband assumption.The narrowband assumption may no longer hold when max(dx N, dy M) > 3.0 m.Source: §VII · pp. 9–10
  • The reflected pattern peaks when the target lies along the RIS pointing direction; off-boresight steering widens beamwidths, while larger RISs produce narrower beams.Source: §VII · p. 10
  • Up to 12 dB SNR improvement occurs for Np = 128 versus Np = 8; larger RISs also improve SNR but increase far-field distance and control complexity.The reported SNR decreases as RIS–target distance increases, and larger RIS dimensions may require near-field processing strategies.Source: §VII · p. 10
  • For the micro UAV, the reported result is an improvement of 10 dB in the resulting SNR.Source: §VII · p. 10
  • Detection probability is evaluated for nonfluctuating Swerling 0 and scan-to-scan decorrelated Swerling 1 targets at PFA = 10^-4 and 10^-6.Source: §VII · p. 10

2 SNRc,

Detection performance declines with target range and is more challenging for SW1 targets; the RIS link also incurs substantial loss versus a clairvoyant monostatic reference. RIS scanning reaches elevation angles up to 45° with loss below 5 dB.

Detection performance

  • With PF A = 10−4, N = M = 133, and Np = 64, Pd > 0.9 is achievable to 1250 m for SW0, while SW1 reduces maximum range by nearly 500 m.At Pd = 0.9, PF A = 10−6 further reduces range by about 100 m versus PF A = 10−4.Source: §2 · p. 10

SNR loss

  • The two-way double-hop RIS link incurs significant loss versus clairvoyant monostatic LOS; larger RISs reduce LSNR but require greater far-field distance and controller complexity.LSNR = SNRm/SNR, with SNRm for the clairvoyant system.Source: §2 · p. 10

Scanning loss

  • RIS scanning covers elevation angles up to 45° with loss below 5 dB; above 20°, increased dwell time is needed to guarantee the desired performance.For θR ≤ 20°, scanning loss is described as quite negligible.Source: §2 · p. 10

VIII. CONCLUSION

The numerical analysis supports the RIS-assisted framework for short-range N-LOS radar surveillance, while identifying RIS drawbacks as an area that still needs fuller understanding. The conclusion also points to joint natural and RIS-induced multipath and adaptive RIS control as future directions.

VIII. CONCLUSION

  • The framework derives SNR and SCR for the RIS-induced two-way, double-hop channel and addresses N-LOS data acquisition and range, angle, and Doppler resolution.SCR expressions cover both surface and volume clutter.Source: §VIII · p. 11
  • Numerical analysis of SNR, detection, and SNR loss assesses RIS size and system parameters, supporting the framework’s capability for N-LOS short-range scenarios.The SNR loss is compared with a LOS monostatic geometry.Source: §VIII · p. 11
  • The drawbacks of RIS use remain incompletely understood while the technology is still maturing.The authors call for further work to weigh RIS pros and cons.Source: §VIII · p. 11
  • Future directions include combining natural and RIS-induced multipath for around-the-corner radar and adapting RIS parameters using radar perception outputs.The proposed adaptation can vary beam steering and polarization.Source: §VIII · pp. 11–12

Related papers

Loading 2104.00456v2…