⚡ electrical engineering

A Novel Stripe-based RIS Optimization for UAV Communications and Sensing in Low-Altitude Wireless Networks

This paper proposes a low-complexity, stripe-based Reconfigurable Intelligent Surface (RIS) optimization framework for Low-Altitude Wireless Networks that simultaneously enhances UAV communication reliability and enables passive sensing by leveraging structural phase gradients to reduce computational search space, with its effectiveness validated through both simulations and real-world outdoor experiments.

Original authors: Burak Ahmet Celebi, Sefa Kayraklik, Onur Salan, Ibrahim Hokelek, Ali Emre Pusane, Ali Gorcin

Published 2026-06-08

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Burak Ahmet Celebi, Sefa Kayraklik, Onur Salan, Ibrahim Hokelek, Ali Emre Pusane, Ali Gorcin

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 drone flying high above a city, but a tall building is blocking your direct line of sight. You can't shout loud enough to get through, and the signal bounces off the wrong way. This is the problem this paper tackles: how to keep a clear, strong connection with a drone (UAV) even when obstacles get in the way, while also using that connection to "see" where the drone is moving.

The solution proposed by the authors is a high-tech "smart mirror" called a Reconfigurable Intelligent Surface (RIS). Think of this RIS as a giant wall made of thousands of tiny, adjustable tiles. Each tile can catch a radio signal from a ground station and reflect it toward the drone. By tweaking the angle of each tile, the wall can focus the signal like a magnifying glass focuses sunlight.

Here is how the paper breaks down the innovation, using simple analogies:

1. The Problem: The "Pixel-by-Pixel" Nightmare

Usually, to make this smart mirror work perfectly, you have to adjust every single tiny tile individually to find the perfect angle.

The Old Way: Imagine trying to focus a massive mirror by turning one tiny screw on one tile, checking if the light got brighter, then moving to the next screw, and so on. If the mirror has 576 tiles (like in their experiment), you have to check hundreds of thousands of combinations. If the drone is moving, you have to do this all over again instantly. It's slow, computationally heavy, and if the signal is weak (like a whisper in a storm), you can't tell if a tiny adjustment helped or not.

2. The Solution: The "Stripe" Strategy

The authors propose a clever shortcut. Instead of treating every tile as a unique individual, they realized that in a clear, open space (far-field), the best way to reflect the signal looks like a stripe pattern.

The Analogy: Imagine the tiles on the mirror are like rows of people in a stadium doing "the wave." Instead of telling every single person exactly when to stand up, you just tell the first row, "Stand up," and the next row, "Stand up a split second later," and so on. The wave moves smoothly across the stadium.
How it works: The system only needs to figure out the "slope" or the "gradient" of the wave (how fast the angle changes from left to right and top to bottom). Instead of searching for the perfect setting for 576 tiles, it only searches for two numbers: the slope in the X direction and the slope in the Y direction.
The Benefit: This reduces the search space from a massive maze to a tiny, manageable grid. It's like finding a needle in a haystack by first looking for the haystack's shape, rather than poking every single piece of hay.

3. Handling the Moving Target: Tracking the Drone

Drones don't stay still; they fly around. If the mirror is set for a stationary drone, the signal will miss as soon as the drone moves.

The Challenge: You need to update the "stripe pattern" constantly without stopping to do a full search every time.
The Paper's Tools: The authors developed three "smart trackers" to keep the mirror focused:
1. Simple Tracking: Like a dog chasing a ball, it just looks a little bit around where the signal was last strong and adjusts slightly.
2. Kalman-Aided Tracking: This is like a GPS predicting where a car will be. It uses math to guess the drone's next move based on its current speed and direction, so the mirror can pre-adjust before the drone even gets there.
3. Extremum Seeking (ESC): This is like a blind person feeling their way up a hill. They take tiny, wiggly steps in different directions to feel which way the ground gets steeper (stronger signal) and keep walking that way.

4. Bonus Feature: "Passive Sensing"

While the mirror is adjusting itself to catch the signal, it accidentally (or rather, intentionally) learns where the drone is.

The Magic: Because the "stripe pattern" depends entirely on the angle of the drone, once the system finds the best pattern, it can mathematically reverse-engineer the drone's location. It's like knowing exactly where a flashlight is pointing just by looking at the shadow it casts on the wall. The system gets a "free" tracking capability without needing extra cameras or sensors.

5. Real-World Proof

The authors didn't just run computer simulations; they built a real prototype.

The Experiment: They set up a real smart mirror (made of 24x24 tiles) on a university campus. They used real radio equipment to send signals from a ground station to a receiver, bouncing them off the mirror.
The Result: The "stripe" method worked better and faster than the old "pixel-by-pixel" methods. It maintained a strong signal even when the environment was noisy, and it successfully tracked the signal path in the real world, proving the math works outside the computer.

Summary

In short, this paper introduces a low-complexity, high-speed method to control a smart radio mirror. Instead of micromanaging every single tile, it controls the "flow" of the reflection using a stripe pattern. This allows it to:

Connect with drones much faster and more reliably, even when signals are weak.
Track moving drones in real-time using smart prediction algorithms.
Locate the drone's position automatically as a side effect of the connection process.

The authors validated this with both computer simulations and real-world outdoor experiments, showing it is a practical solution for the future of low-altitude drone networks.

Technical Summary: A Novel Stripe-based RIS Optimization for UAV Communications and Sensing in Low-Altitude Wireless Networks

Problem Statement
Low-altitude wireless networks (LAWN) aim to support mission-critical aerial operations by integrating communication, sensing, and control across aerial and terrestrial nodes. However, the highly dynamic nature of unmanned aerial vehicle (UAV) operations, combined with signal blockages and rapidly fluctuating wireless channels, challenges the reliability and latency requirements of these systems. While Reconfigurable Intelligent Surfaces (RIS) offer a promising solution to enhance coverage and reliability by controlling the wireless propagation environment, existing optimization approaches often suffer from high computational complexity. Furthermore, many existing methods struggle to maintain performance in low Signal-to-Noise Ratio (SNR) regimes or when phase estimation errors occur, particularly as the number of RIS elements increases. There is a need for a low-complexity optimization framework that can simultaneously ensure reliable communication and provide passive sensing (tracking) capabilities for mobile UAVs without requiring exhaustive search spaces.

Methodology
The paper proposes a stripe-based RIS phase shift optimization framework designed for both stationary and mobile UAV scenarios. The core methodology leverages the inherent structural phase-gradient of adjacent RIS elements in the far-field region, rather than optimizing each element individually.

System Model: The system models a Base Station (BS), an RIS (Uniform Planar Array), and a UAV. In the far-field, the optimal phase profile across the RIS is linear, defined by two parameters: phase gradients $\Delta_x$ and $\Delta_y$ along the x and y axes.
Stationary Optimization (Stripe-Based Search): Instead of searching through $N$ $N$ individual element phases, the algorithm searches for the optimal pair $(\Delta_x, \Delta_y)$ $(Δ_{x}, Δ_{y})$ . This reduces the search space significantly. A two-stage hierarchical search is employed:
- Coarse Search: Uniformly discretizes the gradient space to find a local window.
- Fine Search: Refines the search within the identified window.
- Quantization: The method supports 1-bit RIS architectures (phase shifts of 0 or $\pi$ ), where the linear phase profile manifests as "stripe" patterns due to phase wrapping.
Error Analysis: The paper provides a mathematical analysis of the system's sensitivity to phase estimation errors. It demonstrates that structured gradient errors (affecting the stripe pattern) result in significantly lower SNR loss compared to random, uncorrelated element-wise errors, even in 1-bit quantized systems.
Mobile UAV Tracking: For moving UAVs, the framework extends to real-time tracking using three strategies to update the phase gradients:
- Simple Tracking: Repeated fine-grid searches around the previous optimum.
- Kalman-Aided Tracking: Uses a Kalman filter to predict the evolution of phase gradients based on UAV velocity, narrowing the search window.
- Extremum Seeking Control (ESC) Aided Tracking: Applies continuous, low-amplitude perturbations to the gradients to track the SNR maximum without discrete grid searches, minimizing latency.
Passive Sensing: The estimated phase gradients are mathematically transformed to infer the UAV's elevation and azimuth angles relative to the RIS, providing angular localization as a byproduct of the communication optimization.

Key Contributions

Low-Complexity Algorithm: The paper develops a stripe-based optimization algorithm that reduces computational complexity from linear ( $O(N)$ ) to sublinear ( $O(\sqrt{N})$ ) with respect to the number of RIS elements, making it scalable for large arrays.
Robustness Analysis: It provides analytical derivations and simulations showing that the proposed method maintains robust, high-SNR connectivity even under phase estimation errors and in low-SNR regimes where iterative element-wise methods fail.
Integrated Tracking and Sensing: The framework is extended to support mobile UAVs through Kalman-aided and ESC-aided tracking methodologies. Crucially, it derives a mathematical transformation that converts RIS phase-shift optimization into a continuous UAV tracking mechanism, enabling passive angular localization without dedicated sensing signals.
Experimental Validation: The study validates the approach using a real RIS prototype (24 $\times$ 24 elements) and Software-Defined Radios (SDRs) in an outdoor campus environment, demonstrating practical viability beyond theoretical simulations.

Results

SNR Performance: Simulations indicate that the stripe-based method consistently achieves high SNR across varying UAV distances, closely matching the theoretical upper bound. In contrast, conventional iterative methods (e.g., 1 $\times$ 1 element updates) degrade rapidly at longer distances or low SNR because single-element phase changes become indistinguishable from noise.
Computational Efficiency: Runtime analysis shows that while iterative methods scale linearly with the number of RIS elements, the stripe-based method scales with the square root of the element count, offering substantial time savings for large arrays.
Tracking Accuracy: In mobile scenarios, the ESC-aided tracking approach achieved the lowest average SNR loss (approx. 0.31 dB) compared to the practically optimal configuration, outperforming both simple and Kalman-aided trackers by reducing update latency and mismatch.
Experimental Verification: Outdoor measurements confirmed that the stripe-based method yields higher received power than iterative baselines (1 $\times$ 1 and 4 $\times$ 4) at various transmitter distances. The measured radiation patterns verified that the RIS successfully directed reflected signals toward the receiver.

Significance
The paper claims that this work bridges the gap between theoretical RIS optimization and practical deployment in LAWNs. By exploiting the structural phase-gradient of the RIS, the proposed method offers a computationally efficient solution that is robust to hardware imperfections and low SNR conditions. Its ability to simultaneously optimize communication links and provide passive UAV tracking (angular localization) makes it a viable enabler for next-generation, service-aware 3D wireless networks. The successful outdoor demonstration with a real prototype underscores the practical feasibility of deploying such low-complexity, stripe-based strategies in real-world environments.