PaperlayerView exact revision on arXiv

Source-linked AI summary

Radar Target Detection aided by Reconfigurable Intelligent Surfaces

Stefano Buzzi, Emanuele Grossi, Marco Lops, Luca Venturino

arXiv:2104.00768v3eess.SPcs.ITmath.IT

TL;DR

Radar detection with an RIS is considered as a way to exploit an additional target echo, addressing a topic the authors describe as previously unexplored. The paper analyzes closely- and widely-spaced geometries and several beam configurations, and reports substantial gains in its examples. It concludes that more complex scenarios require further study.

  • Problem

    The paper asks whether an RIS can improve radar target detection by supplying an additional target echo, a benefit the authors describe as unexplored.

    Source: §I · p. 2, §II · p. 3

  • Method

    The authors theoretically analyze RIS-assisted radar detection across closely- and widely-spaced geometries and different beam configurations.

    Source: §I · p. 2, §III · p. 3

  • Results

    The examples report large detection-performance improvement with an RIS, and Case a always outperforms Case b in the stated comparison.

    Source: §IV · p. 5

  • Takeaways & Limitations

    A large RIS close enough to fill the radar beam can provide substantial indirect-echo gain, while widely-spaced operation offers two observations at different aspect angles.

    Source: §V · p. 5

    The paper considers a basic, simple setting; effects in more complex scenarios such as MIMO radar and multiple distributed RISs remain for further research.

    Source: §V · p. 5

Abstract

from arXiv · show

In this work, we consider the target detection problem in a sensing architecture where the radar is aided by a reconfigurable intelligent surface (RIS), that can be modeled as an array of sub-wavelength small reflective elements capable of imposing a tunable phase shift to the impinging waves and, ultimately, of providing the radar with an additional echo of the target. A theoretical analysis is carried out for closely- and widely-spaced (with respect to the target) radar and RIS and for different beampattern configurations, and some examples are provided to show that large gains can be achieved by the considered detection architecture.

I. INTRODUCTION

RISs can electronically control reflected wavefronts, adding channel-shaping options beyond transmitter and receiver design. This paper addresses their potential for radar detection, which the authors describe as unexplored, and analyzes conditions and examples for RIS-assisted detection.

I. INTRODUCTION

  • RISs are nearly passive surfaces whose electromagnetic response can be electronically controlled to tune reflected wavefronts and shape the propagation channel.The passage lists phase, amplitude, frequency, and polarization as tunable properties.Source: §I · p. 2
  • Earlier RIS studies cited here focus on communication or localization, including energy efficiency, IoT access, indoor configuration, massive MIMO, and localization.The cited examples include –[11].Source: §I · p. 2
  • The paper addresses the previously unexplored question of whether an RIS can enhance radar target detection by providing an additional target echo.The authors present their work as a first contribution on RIS benefits for radar detection.Source: §I · p. 2
  • The analysis considers closely- and widely-spaced radar and RIS configurations, relates system design to bandwidth, carrier frequency, sizes, and positions, and reports substantial SNR improvements in two examples.The authors also analyze two radar beam configurations and phase adjustment to align echoes.Source: §I · p. 2

II. SYSTEM MODEL

The system model treats an RIS as an additional target-echo path for radar detection. It defines narrowband, far-field operating assumptions and three beam and delay configurations for comparing detection performance.

II. SYSTEM MODEL

  • The model assumes a narrowband waveform with unresolved echo delays, directive radar antennas, a large RIS, and plane-wave propagation on the radar-target, target-RIS, and radar-RIS paths.The radar and RIS need not be in each other's far field.Source: §II · p. 3
  • Case a uses one target-directed transmit beam and receive beams toward the target and RIS; Case b uses two transmit beams and resolvable direct and indirect echoes.Case c uses Case b's transmit beams but assumes the two echo delays are not resolvable.Source: §II · p. 3
  • The received discretized signal is considered after matched filtering and sampling when a target is present in the resolution cell under test.The model includes unknown target RCSs and channel phases, Gaussian noise, and a transmit-power fraction for Cases b and c.Source: §II · p. 3
  • The detection problem is to optimize a radar-RIS architecture to improve radar performance by exploiting the additional target echo available from the RIS.The model defines radar power, wavelength, RIS element count, distances, and beam gains.Source: §II · pp. 2–3

III. SYSTEM DESIGN

The design distinguishes radar-RIS geometries by the angle at which the target sees them, with the boundary set by wavelength and target size.

III. SYSTEM DESIGN

  • When ξ ≪ λ/D_t, the target sees the radar and RIS as closely spaced or co-located because they lie within the same angular scattering beam.The target scattering beam is proportional to λ/D_t.Source: §III · p. 3
  • When ξ ≥ λ/D_t, the radar and RIS are widely spaced with respect to the target.ξ is the angle between the target-to-radar and target-to-RIS line segments.Source: §III · p. 3
  • Which spacing regime applies depends on target size, antenna wavelength, and the mutual distances among radar, target, and RIS.Source: §III · p. 3

A. Closely-spaced radar and RIS

For the closely-spaced case, the design aligns signal phases and evaluates detection through GLRTs for unknown target parameters, with SNR bounds and a power-splitting choice for Case c.

A. Closely-spaced radar and RIS

  • The RIS phases can be adjusted to align the signal terms, using the phase relation between the radar-target and target-RIS channels.The target-RIS phase difference across RIS elements is determined by their positions relative to the target.Source: A. Closely-spaced radar and RIS · pp. 3–4
  • The detection tests use GLRTs with respect to the unknown complex target parameter, and the false-alarm probability is P_fa = e^-γ.The detection probability depends on the distribution of σ, with closed-form expressions noted for the non-fluctuating and Swerling models.Source: A. Closely-spaced radar and RIS · p. 4
  • For Case c, the stated SNR is achieved by properly splitting transmit power between the two available beams.The passage identifies an optimal power fraction but does not provide its complete expression.Source: A. Closely-spaced radar and RIS · p. 4

B. Widely-spaced radar and RIS

For the widely-spaced configuration, the RIS phases are chosen using channel-phase information, while the target responses are modeled as independent random variables. Unknown average target RCSs constrain power allocation in Cases b and c.

B. Widely-spaced radar and RIS

  • The RIS phases can be chosen using channel-phase information; the target-to-RIS phase differences are determined by the target and element positions under plane-wave illumination.The passages also describe choosing RIS phases to compensate for these phase differences.Source: B. Widely-spaced radar and RIS · p. 4
  • The target responses observed through the radar and RIS are modeled as independent random variables, and detection probability depends on the target RCS distribution.For Cases a and b, false-alarm probability is Pfa = e^-γ(1 + γ); for Case c it is Pfa = e^-γ.Source: B. Widely-spaced radar and RIS · p. 4
  • For Case b, the second observation’s SNR is SNRb,2 = (1 − ϵ)SNR0Kstσ̄s/σ̄.The passage provides this expression for the second observation in Case b.Source: B. Widely-spaced radar and RIS · p. 4
  • When average RCSs are unknown, ϵ cannot be directly optimized in Cases b and c; with bounded averages, Case c selects ϵ to maximize worst-case SNR.The bounds are specified for the average RCSs observed by the radar and RIS.Source: B. Widely-spaced radar and RIS · p. 4

IV. EXAMPLES

The examples report large SNR gains in the closely-spaced scenario and improved detection probability with RIS assistance in the widely-spaced scenario. They evaluate specified radar arrays, bandwidths, RIS sizes, and beam configurations.

IV. EXAMPLES

  • The examples use a 3 GHz radar with two 1 × 1 m arrays, λ/2 element spacing, cosine element patterns, and 1 or 10 MHz bandwidths.The RIS is a square surface with λ/2 element spacing and an element RCS model based on incidence and reflection angles.Source: §IV · p. 4
  • In the closely-spaced scenario, tested RIS sizes range from 2 to 5 m, with radar distance varied so its 3-dB beamwidth covers the RIS area.For two transmit and one receive beam, Case b holds at 10 MHz, while Case c holds for RIS sizes up to 3 m at 1 MHz.Source: §IV · pp. 4–5
  • Large performance improvements are reported across all operating conditions in the closely-spaced scenario.Table I reports SNR gain [dB] for this scenario, but the supplied passages do not provide its cell values.Source: §IV · p. 5, Table I · p. 5
  • In the widely-spaced scenario, RIS assistance improves detection probability over no RIS, and Case a outperforms Case b when the average RCSs are equal.The comparison uses an exponential target-RCS fluctuation model and Pfa = 10^-6.Source: §IV · p. 5

V. DISCUSSION

The discussion links RIS performance to how well the surface fills the radar beam and describes its role as a reconfigurable feed. It also identifies bistatic diversity and the need to test more complex scenarios; the appendix supplies derivations and supporting method details.

V. DISCUSSION

  • When the RIS fills the radar beam area (Asr ≈ Ars), the discussion says the resulting gain can be large for a small, low-cost radar and large RIS.The radar beam aimed at the RIS acts as a feed for a reconfigurable surface that can electronically tune its beamforming.Source: §V · p. 5
  • In the widely spaced scenario, the system acts as a low-cost bistatic radar, with two-view diversity in Cases a and b that may improve estimation.Source: §V · p. 5
  • The discussion limits its conclusions to a basic setting and calls for further study of MIMO radars and multiple distributed RISs used in transmission and reception.Source: §V · p. 5

APPENDIX

  • The appendix provides detailed derivations of the paper’s main results and equations.Source: APPENDIX · p. 6

A. RIS-target channel phases in Footnote 4

B. GLRT’s in Eqs. (6)

  • The GLRT derivation models the observation as a target response with unknown deterministic complex amplitude plus Gaussian noise with covariance PwI.Source: B. GLRT’s in Eqs. (6) · p. 6

C. SNRs in Eqs. (7)

  • For Case a, the test statistic uses αTx; the stated maximization for Cases b and c gives st = 1/(1+Kst), respectively.Source: C. SNRs in Eqs. (7) · p. 6

D. Detection probability in the closely-spaced scenario

The section gives detection-probability expressions under several target fluctuation models, then describes the RIS–target channel phases and refers back to earlier derivations for the GLRTs and SNRs.

D. Detection probability in the closely-spaced scenario

  • Detection probability is evaluated for nonfluctuating, exponential, and gamma-distributed target RCS models using the SNR and Marcum Q-function.The models correspond to Marcum’s nonfluctuating case and Swerling cases 1/3 and 2/4.Source: D. Detection probability in the… · p. 6

E. RIS-target channel phases in Footnote 5

  • In the widely-spaced geometry, RIS–target channel phases comprise a known element-dependent steering-vector component and a phase offset associated with the first RIS element.The two echoes have independent target RCS responses.Source: E. RIS-target channel phases in… · p. 7, Figure A2 · p. 7

F. GLRT’s in Eqs. (11) and SNRs in Eqs. (12)

G. SNRs in Eqs. (13) and (14)

This section derives the Case c SNR and describes a worst-case SNR optimization over the transmit-power fraction.

H. Detection probability in the widely-spaced scenario

For the widely-spaced scenario, the section describes detection-probability expressions and approximations for RIS gains, including the conditions under which the gain approximation applies.

H. Detection probability in the widely-spaced scenario

I. RIS gains in Eq. (15)

  • The RIS-gain approximation uses effective RIS area and radar beam coverage; its element-count term is L when the 3-dB beam covers the whole RIS and approximately Ars/Asr otherwise.The same approximation applies to Kst because Sst,ℓ = Ssr,ℓ.Source: §I · pp. 7–8

J. Radar Equation in (16)

Under an approximately matched-area assumption, the section gives the indirect-echo SNR for Cases b and c and relates the Case a expression to the equality of two gains.

Related papers

Loading 2104.00768v3…