⚡ electrical engineering

Compact Reconfigurable Intelligent Surface with Phase-Gradient Coded Beam Steering and Controlled Substrate Loss

This paper presents a compact, low-cost 1-bit reconfigurable intelligent surface (RIS) fabricated on an FR4 substrate with an optimized air gap, featuring a simplified PIN-diode biasing network and phase-gradient coding that achieves measured beam steering up to ±30° and 9 dB gain enhancement in the 5G n78 band (3.38–3.67 GHz).

Original authors: Mahendra Kheti, Debapratim Ghosh, Soumya P. Dash

Published 2026-04-07

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Mahendra Kheti, Debapratim Ghosh, Soumya P. Dash

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 in a crowded, noisy room full of pillars and walls. Your voice hits the walls, bounces off in random directions, and gets lost before it reaches your friend. This is exactly what happens with wireless signals (like 5G) in cities or inside buildings: they get blocked, scattered, and weakened.

This paper presents a clever solution to that problem: a "Smart Mirror" for radio waves.

Here is the story of how the researchers built this mirror, explained simply:

1. The Problem: The "Dead Zone"

In modern wireless networks, signals often hit obstacles (like concrete walls or metal furniture) and die out. To fix this, scientists use something called a Reconfigurable Intelligent Surface (RIS). Think of an RIS as a giant, programmable mirror. Instead of just reflecting light, it reflects radio waves. But unlike a normal mirror that reflects everything in one direction, this smart mirror can be programmed to bend the signal exactly where you want it to go, like a laser pointer that you can aim with your mind.

2. The Challenge: Cost vs. Quality

Usually, making these "smart mirrors" requires expensive, high-tech materials that don't waste energy. The researchers wanted to use FR4, which is the cheap, standard material used in almost all computer motherboards and cheap electronics.

The Problem with FR4: It's cheap, but it's "spongy" for radio waves—it absorbs them and turns them into heat (loss), making the signal weak.
The Solution: They invented a way to make this cheap material act like expensive material.

3. The Secret Sauce: The "Air Gap" Sandwich

Imagine you have a heavy, wet sponge (the FR4 substrate) that absorbs water (radio energy). If you sandwich that wet sponge between two dry, airy layers, the water doesn't soak through as easily.

The Innovation: The team built their mirror using a three-layer "sandwich." They placed a tiny air gap between the layers.
Why it works: Radio waves hate traveling through solid plastic but love traveling through air. By forcing the waves to spend more time in the air gap and less time in the "spongy" plastic, they drastically reduced the energy loss. This allowed them to use cheap materials without sacrificing performance.

4. The "Switches": Tiny Light Switches for Waves

To steer the beam, the mirror needs to change how it reflects the signal. The researchers put a tiny electronic switch (called a PIN diode) on every little square of the mirror.

The Analogy: Imagine a wall covered in thousands of tiny tiles. Each tile has a switch.
- Switch ON: The tile reflects the wave one way.
- Switch OFF: The tile reflects the wave the opposite way (180 degrees flipped).
By turning specific switches on and off in a specific pattern, the whole mirror can "tilt" the reflected beam, just like tilting a solar panel to catch the sun.

5. The "Brain": Simple and Cheap

Usually, controlling thousands of these switches requires a super-complex computer brain (FPGA) and messy wiring.

The Innovation: The team used a simple, low-cost Arduino (a small, cheap microcontroller) and a few shift registers.
The Result: They created a system that is easy to build, cheap to make, and easy to control, yet it can steer the signal up to 30 degrees left or right.

6. The Results: A Clearer Connection

They built a prototype (a 10x10 grid of these smart tiles) and tested it in a special room designed to absorb all echoes (an anechoic chamber).

Beam Steering: They successfully aimed the signal to different angles, proving the mirror works.
Signal Boost: They showed that using this mirror made the signal 9 decibels stronger (which is a huge jump in clarity) compared to just letting the signal bounce off a normal wall.
Real-World Test: They even sent actual data (QPSK symbols, which are like digital letters) through the mirror. Without the mirror, the letters were scrambled and blurry. With the mirror, the letters came through clearly and sharply.

Summary

This paper is about taking a cheap, common material (FR4), giving it a smart "air gap" sandwich design, and adding a simple Arduino brain to create a programmable mirror for 5G signals.

The Big Picture: This technology could be the key to making 5G and future 6G networks work perfectly in dense cities and inside buildings, without needing to build expensive, high-tech infrastructure everywhere. It turns a cheap piece of plastic into a high-performance signal director.

1. Problem Statement

The deployment of 5G and beyond-5G (B5G) networks faces significant challenges in dense urban and indoor environments, including signal blockage, severe multipath fading, and poor coverage. While Reconfigurable Intelligent Surfaces (RISs) offer a promising solution by programmatically controlling reflected beams, existing implementations face three primary hurdles:

High Cost and Loss: Many designs rely on specialized, low-loss substrates (e.g., Rogers) to minimize dielectric losses, increasing manufacturing costs. Standard low-cost substrates like FR4 are often avoided above 3 GHz due to high loss tangents.
Complexity and Size: High-performance RISs often require complex multi-bit unit cells, intricate biasing networks, and large aperture sizes, making them difficult to deploy practically.
Control Overhead: Existing prototypes frequently depend on complex FPGA-based control systems and extensive wiring, hindering scalability.

2. Methodology

The authors propose a 1-bit RIS architecture that addresses these issues through a novel unit cell design, a simplified control scheme, and a phase-gradient coding strategy.

A. Unit Cell Design & Substrate Optimization

Structure: The unit cell utilizes a three-layer structure (Radiating Layer, Ground Layer, and Biasing Layer) separated by a manually stacked air gap.
Loss Mitigation: Instead of using expensive low-loss materials, the design employs standard FR4 substrate ( $\varepsilon_r \approx 4.4$ $ε_{r} \approx 4.4$ , $\tan \delta \approx 0.02$ $tan δ \approx 0.02$ ). To counteract FR4's high loss, an optimal air gap ( $h_{air} = 0.5$ $h_{ai r} = 0.5$ mm) is introduced between the ground and biasing layers.
- Mechanism: The air gap reduces the effective dielectric constant ( $\varepsilon_{eff}$ ) and redistributes the electromagnetic field energy away from the lossy FR4 substrate. This is quantified using a Loss Participation Ratio (LPR), which significantly lowers the effective loss tangent.
Via Placement: Unlike traditional designs that place vias at field minima, this work uses Floquet port excitation to analyze surface current distribution. Vias are placed at locations of maximum surface current density to maximize phase sensitivity while maintaining high reflection magnitude.
Geometry: The unit cell is compact ( $17 \text{ mm} \times 17 \text{ mm}$ or $0.198\lambda_g \times 0.198\lambda_g$ ) and features a circular metallic patch with slots for polarization independence.

B. Biasing and Control Network

Switching Mechanism: A PIN diode (Skyworks SMP1345-079LF) is used to switch between two states (ON/OFF), providing $0^\circ$ and $180^\circ$ phase shifts.
Simplified Circuit: A compact biasing network is designed using discrete reactive components (inductors, capacitors) and a radial stub to ensure effective RF/DC isolation and broadband impedance matching.
Control Architecture: The system is driven by an Arduino Mega 2560 microcontroller interfaced with cascaded shift registers (TPIC6B595N). This eliminates the need for complex FPGA setups, reducing hardware complexity and power consumption (approx. 0.05 W per cell).

C. Beam Steering Strategy

Phase-Gradient Coding: A phase-gradient coding scheme is implemented to steer the reflected beam. By assigning specific $0^\circ$ or $180^\circ$ states to the $10 \times 10$ array elements, a linear phase gradient is created across the aperture.
Theoretical Modeling: The paper derives the far-field scattered fields based on the spatial phase distribution, accounting for incident angles and the discrete nature of the 1-bit states.

3. Key Contributions

Low-Cost, Low-Loss Design: Demonstration that FR4 substrates can be effectively used for RISs above 3 GHz by introducing a controlled air gap and optimizing via placement, achieving a Figure of Merit (FoM) that balances phase control and loss reduction.
Compact 1-bit Unit Cell: Development of a $17 \text{ mm} \times 17 \text{ mm}$ unit cell with a simplified biasing network, resulting in a total array size of $2.9\lambda_g \times 2.9\lambda_g$ ( $255 \text{ mm} \times 255 \text{ mm}$ ).
Simplified Control System: Implementation of a microcontroller-based control system (Arduino + Shift Registers) that significantly reduces complexity compared to FPGA-based alternatives.
Experimental Validation: Comprehensive experimental verification in both anechoic and noisy environments, including QPSK signal transmission.

4. Results

Operating Band: The RIS operates effectively in the 3.38–3.67 GHz range (centered at 3.5 GHz), covering the 5G n78 band with an 8.3% fractional bandwidth.
Beam Steering Performance:
- Achieved measured beam steering up to $\pm 30^\circ$ in both anechoic and noisy environments.
- The phase difference between ON/OFF states remains within $0^\circ/180^\circ \pm 20^\circ$ .
- Gain Enhancement: The RIS provides a measured reflected wave gain enhancement of approximately 9 dB over the incident wave within the $\pm 30^\circ$ angular range.
Comparison with State-of-the-Art:
- Compared to other works (e.g., Ref [27], [13], [28]), this design offers a significantly smaller aperture size ( $2.9\lambda_g$ vs. $7.9\lambda_g$ – $10\lambda_g$ ) while maintaining comparable beam steering capabilities ( $\pm 30^\circ$ ).
- It uses fewer control lines (100 vs. 128–400) and a simpler MCU-based control method.
Communication Performance:
- In a noisy environment, the RIS-assisted channel successfully transmitted QPSK-modulated symbols.
- The received constellation diagram showed tighter symbol clusters and reduced error rates compared to the non-RIS scenario, confirming improved channel reliability and Euclidean distance between symbols.

5. Significance

This work bridges the gap between theoretical RIS concepts and practical, low-cost deployment. By proving that standard FR4 substrates can be used effectively through structural optimization (air gaps and via placement), the paper removes a major cost barrier for 5G/6G infrastructure. Furthermore, the simplified control architecture makes large-scale RIS deployment more feasible, as it avoids the high cost and power consumption of FPGA-based controllers. The successful validation of beam steering and data transmission (QPSK) in real-world conditions underscores the viability of this compact, 1-bit RIS for enhancing coverage in next-generation wireless networks.