⚡ electrical engineering

Radar Cross Section Characterization of Quantized Reconfigurable Intelligent Surfaces

This paper presents a theoretical framework and experimental validation for a low-complexity, quantized Reconfigurable Intelligent Surface (RIS) that enhances radar detection in non-specular and shadowed regions by deriving closed-form expressions for its radar cross section and demonstrating its ability to redirect beams and recover micro-Doppler signatures.

Original authors: Kainat Yasmeen, Shobha Sundar Ram, Debidas Kundu

Published 2026-03-31

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Kainat Yasmeen, Shobha Sundar Ram, Debidas Kundu

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to talk to a friend who is hiding behind a large, thick wall. You can't see them, and your voice (the radar signal) bounces off the wall and goes the wrong way. You are effectively blind to your friend's presence.

Now, imagine you have a smart mirror (called a Reconfigurable Intelligent Surface, or RIS) that you can stick to the wall. This isn't a normal mirror; it's a "programmable" mirror made of thousands of tiny, switchable tiles. You can tell this mirror, "Hey, don't just bounce the light straight back at me. Instead, catch my voice and shout it around the corner to my friend!"

This paper is about building and testing exactly that kind of smart mirror for radar systems, specifically using a very simple, low-cost version of it.

Here is the breakdown of what the researchers did, using some everyday analogies:

1. The Problem: The "Blind Spot"

Standard radar works like a flashlight. If you shine it at a wall, the light bounces back. If your target (a car, a person, a drone) is behind a building or in a "shadow" where the light doesn't reach, the radar sees nothing.

The Challenge: High-frequency signals (like 5G or modern radar) can't pass through walls. They rely on bouncing off surfaces. If the geometry is wrong, the signal never reaches the target.

2. The Solution: The "Smart Mirror" (RIS)

The researchers created a surface covered in tiny electronic tiles.

The Analogy: Think of a choir. If everyone sings the same note at the same time, the sound is loud and goes in one direction. If they are out of sync, the sound is messy.
The RIS: This surface acts like a conductor for radio waves. It catches the incoming wave from the radar, tweaks the timing (phase) of each tiny tile, and then re-broadcasts the wave in a new direction, effectively "bending" the signal around obstacles to hit a target that was previously invisible.

3. The "Cheap" Version: The "Pixelated" Mirror

Building a perfect mirror where every tile can be set to any angle is expensive and power-hungry.

The Innovation: The researchers used a 1-bit RIS. This is like a mirror made of tiles that can only be in two states: ON or OFF (or "0" and "1").
The Metaphor: Imagine trying to paint a smooth gradient of color using only black and white pixels. It's not perfectly smooth; it looks a bit "blocky" or "pixelated."
The Result: This "blocky" mirror works, but it creates some side effects. Just like a pixelated image has jagged edges, this mirror creates "grating lobes"—unwanted extra beams of energy shooting off in the wrong directions. The paper calculates exactly how much "signal strength" is lost because of this pixelation.

4. The "Two-Way Street" Surprise

Here is the most interesting discovery in the paper.

The Old Assumption: Scientists used to think that if a mirror sends a signal from Point A to Point B, it would send it back from B to A with the exact same strength. They assumed the path was perfectly symmetrical.
The Reality: The researchers found that it's not symmetrical.
- The Analogy: Think of throwing a ball at a bumpy wall. If you throw it at a specific angle to hit a target, the ball might bounce perfectly. But if you throw the ball back from the target at the same angle, the bumpy wall might scatter it differently because the "bumps" (the tiny tiles) are designed for the first direction, not the return trip.
- The Finding: The signal strength going to the target is different from the signal strength coming back. You have to calculate both separately to know if your radar will actually work.

5. The Experiment: Proving it Works

The team didn't just do math; they built a real prototype.

The Setup: They built a 16x10 grid of these "smart tiles" (about the size of a large pizza box) and put it in a soundproof room (an anechoic chamber) to test it without interference.
The Test: They used a radar to look for a person holding a special reflector.
- Without the mirror: The radar couldn't see the person because they were standing in the "blind spot."
- With the mirror: The radar "saw" the person clearly.
The Micro-Doppler Magic: They even detected the tiny movements of the person's hand (micro-Doppler). It's like hearing the swish of a hand waving, even though the person is hidden behind a wall. The mirror redirected the radar waves to "listen" to the hand's movement.

6. The Takeaway

This paper proves that you don't need a super-expensive, perfect mirror to make radar smarter.

Cost vs. Performance: Even a "cheap," pixelated (1-bit) mirror works well enough to see around corners.
The Trade-off: You lose a little bit of signal strength and get some "noise" (grating lobes), but you gain the ability to see things that were previously impossible to detect.
Future: This technology could be the key to self-driving cars seeing around blind corners, or security systems detecting intruders hiding behind walls, all using low-cost, energy-efficient surfaces.

In short: They built a "smart, pixelated mirror" that bends radar waves around corners. They proved that even though the mirror isn't perfect, it's good enough to let a radar "see" the invisible, changing the rules of how we calculate signal strength for these new systems.

1. Problem Statement

Reconfigurable Intelligent Surfaces (RIS) offer programmable control over electromagnetic wave propagation, potentially enabling radar sensing in non-line-of-sight (NLOS) and shadowed regions where direct paths are blocked. However, existing literature on RIS-assisted radar suffers from several critical gaps:

Idealized Assumptions: Most prior works assume continuous phase control, ignoring the practical constraints of hardware quantization (e.g., 1-bit or 2-bit phase shifts).
Neglect of Bistatic Asymmetry: Traditional radar models often assume reciprocity (symmetry between forward and backward paths). This paper identifies that for finite-sized, quantized RIS, the Radar Cross Section (RCS) differs significantly between the forward path (Radar $\to$ RIS $\to$ Target) and the backward path (Target $\to$ RIS $\to$ Radar) due to phase gradient mismatches and diffraction.
Lack of Experimental Validation: There is a scarcity of experimental data validating theoretical models against real-world hardware, specifically regarding the impact of quantization errors, grating lobes, and fabrication non-idealities on radar detection performance.

2. Methodology

The authors employ a multi-faceted approach combining analytical theory, full-wave simulation, and experimental validation:

Theoretical Framework (Aperture Field Theory):
- Derived closed-form expressions for the scattered electric field and bistatic RCS of phase-quantized RIS apertures.
- Modeled the effects of discrete phase states (1-bit, 2-bit, 3-bit) and their impact on grating lobe formation.
- Explicitly calculated separate RCS values for the forward ( $\sigma_f$ ) and backward ( $\sigma_b$ ) paths to demonstrate angular asymmetry.
- Developed a link budget analysis incorporating the four-way propagation path (Radar-RIS-Target-RIS-Radar).
Full-Wave Simulation:
- Used CST Microwave Studio to simulate a 2D RIS environment.
- Accounted for realistic factors ignored in simple theory, such as mutual coupling between unit cells, edge diffraction, and dielectric/ohmic losses.
- Simulated target detection performance across various RIS aperture sizes ( $8\times10$ , $16\times10$ , $32\times10$ ) and steering angles.
Experimental Validation:
- Fabrication: Constructed a $[16 \times 10]$ 1-bit RIS operating at 5.5 GHz. Each unit cell uses a PIN diode to switch between two phase states ($0$ and $\pi$ ).
- Setup: Conducted measurements in a 4m $\times$ 4m anechoic chamber.
- Tests:
  1. Scattering Patterns: Measured beam steering, beam-squint errors, and Peak-to-Specular Lobe Ratios (PSLR).
  2. RCS Characterization: Measured monostatic RCS for various incident ( $\theta_i$ ) and desired ( $\theta_d$ ) angles.
  3. Micro-Doppler Detection: Tested the system's ability to detect human motion (using a corner reflector) in non-specular regions where the radar antenna's main lobe does not reach.

3. Key Contributions

Closed-Form Bistatic RCS Expressions: The paper provides the first analytical derivation of scattered fields for quantized RIS that explicitly accounts for phase discontinuities and grating lobes.
Discovery of Bistatic Asymmetry: The authors prove that $\sigma_f \neq \sigma_b$ for most angle combinations in quantized RIS. This challenges the assumption of reciprocity in RIS link budgets and necessitates independent characterization of both paths for accurate SNR estimation.
Quantization Impact Analysis: Systematic analysis showing how 1-bit quantization introduces grating lobes and reduces PSLR, while higher bit-depths (2-bit, 3-bit) approach continuous performance but increase hardware complexity.
Hardware Prototype & Validation: Successful fabrication and testing of a 1-bit RIS, demonstrating strong agreement between theory, simulation, and measurement regarding beam steering and RCS patterns.
Non-Specular Detection: Experimental proof that RIS can redirect radar beams to detect micro-Doppler signatures of moving targets located outside the radar's direct field of view.

4. Key Results

RCS Asymmetry: For a $[16 \times 10]$ RIS with $\theta_i = 0^\circ$ and $\theta_d = 45^\circ$ , the forward RCS ( $\sigma_f$ ) was 3.5 dBsm, while the backward RCS ( $\sigma_b$ ) was only 1.8 dBsm (for 1-bit). This confirms that the two paths cannot be treated as identical.
Quantization Effects:
- 1-bit RIS: Exhibits significant grating lobes (e.g., at $-45^\circ$ when steering to $45^\circ$ ) and lower PSLR.
- 3-bit RIS: Performance closely matches continuous phase control, with minimal grating lobes.
Aperture Size: Increasing the RIS size (from $8\times10$ to $32\times10$ ) significantly improves SNR and detection probability. A $32\times10$ RIS detected targets in 41 out of 50 positions, compared to 21 for an $8\times10$ RIS.
Measurement vs. Theory:
- Measured beam-squint errors were slightly higher than theoretical predictions (e.g., $2.5^\circ$ error at $15^\circ$ steering vs. $0.3^\circ$ theoretical) due to calibration and fabrication tolerances.
- Measured RCS values were generally lower than theoretical predictions (due to losses and edge effects) but aligned well with full-wave simulations.
Micro-Doppler Detection: In the absence of RIS, the radar could not detect the target outside its main lobe. With the RIS steered to the target's location, clear micro-Doppler signatures (positive and negative Doppler shifts from the swinging reflector) were successfully captured.

5. Significance

This work bridges the gap between theoretical RIS models and practical radar implementation. By rigorously characterizing the bistatic RCS of quantized RIS, the authors provide a realistic framework for designing RIS-enhanced radar systems. The findings are crucial for:

Link Budget Design: Engineers must now account for asymmetric path losses ( $\sigma_f$ and $\sigma_b$ ) rather than assuming reciprocity.
Hardware Trade-offs: The study validates that while 1-bit RIS is cost-effective, it incurs performance penalties (grating lobes, lower SNR) that must be managed via aperture sizing or advanced signal processing.
Next-Gen Sensing: It demonstrates the viability of using low-cost, programmable metasurfaces to extend radar coverage into blind spots, enabling robust detection in cluttered urban environments and for security applications (e.g., human motion detection).