⚡ electrical engineering

RSMA Enabled Hierarchical UAV Networks with Non Linear Energy Harvesting: Outage Probability Analysis and UAV Placement Optimization

This paper proposes a hierarchical UAV network integrating non-linear energy harvesting and rate-splitting multiple access (RSMA) under hardware impairments and imperfect channel state information, deriving outage probability expressions and optimizing UAV placement to demonstrate superior reliability and performance compared to existing benchmarks.

Original authors: Faicel Khennoufa, Khelil Abdellatif, Metin Ozturk, Halim Yanikomeroglu, Safwan Alfattani

Published 2026-06-08

📖 5 min read🧠 Deep dive

CC0 1.0

Original authors: Faicel Khennoufa, Khelil Abdellatif, Metin Ozturk, Halim Yanikomeroglu, Safwan Alfattani

Original paper dedicated to the public domain under CC0 1.0 (http://creativecommons.org/publicdomain/zero/1.0/). ✨ 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 a team of drones (UAVs) working together to deliver Wi-Fi signals to people on the ground, perhaps in a disaster zone where cell towers are down. This paper proposes a new way to organize these drones so they can work longer, talk to each other better, and not run out of battery.

Here is the breakdown of their idea using simple analogies:

1. The Team Structure: The "Boss" and the "Helpers"

Think of the network as a hierarchy:

The Boss Drone (CHU): This is a large, powerful drone hovering high up. It has a big battery (maybe solar-powered) and acts as the main source of information and energy.
The Helper Drones (CMUs): These are smaller drones hovering lower. They don't have big batteries. Instead, they act like middlemen. They catch the signal from the Boss, decode it, and then shout it down to the people on the ground (IoT devices).

The Problem: The Helper Drones run out of battery quickly.
The Paper's Solution: Instead of waiting for someone on the ground to charge them (which is hard if the ground infrastructure is broken), the Helpers "steal" a little bit of energy from the Boss Drone's signal while they are listening to it. This is called Wireless Energy Harvesting.

2. The Energy Trick: The "Non-Linear" Battery

Usually, scientists assume that if you double the radio signal, you get double the energy. But in the real world, batteries and circuits are picky. They get full and stop accepting more energy, like a cup that overflows.

The Paper's Insight: They used a realistic model (Non-Linear) that accounts for this "overflow." They found that if you don't account for this limit, your math is wrong. By using the realistic model, they can plan better so the drones don't waste energy trying to fill a cup that's already full.

3. The Conversation Method: "Splitting the Message" (RSMA)

When the Boss Drone talks to the Helpers, there is a lot of noise and interference.

Old Way (NOMA): Imagine the Boss shouting a message to two people at once. One person is close, one is far. The Boss shouts loudly, and the far person hears it, but the close person has to wait for the far person to finish "listening" before they can understand their part. It's a bit rigid.
The Paper's Way (RSMA - Rate Splitting Multiple Access): The Boss splits the message into two parts: a Common Part (everyone needs to hear this) and a Private Part (just for specific people).
- The Helpers listen to the Common part first.
- Then, they use that knowledge to help filter out the noise and hear the Private parts clearly.
- The Result: It's like having a translator who helps you understand a noisy room better. This method is much more efficient at handling interference than the old way.

4. The Real-World Mess: "Bad Glasses" and "Static"

The paper acknowledges that real life isn't perfect.

Hardware Impairments (HWI): The drones' radios aren't perfect. They have tiny manufacturing defects, like a slightly bent antenna or a noisy speaker. This adds "static" to the signal.
Imperfect Channel State Information (ICSI): The drones try to guess how good the connection is, but their guess isn't 100% accurate. It's like trying to aim a flashlight in the dark with slightly foggy glasses.
The Paper's Claim: Most previous studies pretended these problems didn't exist. This paper builds its math including these imperfections, making the results much more reliable for real-world use.

5. The Optimization Games: "Choosing the Best Team"

The authors created two "games" (algorithms) to make the system work better:

Game 1: Picking the Best Helpers (CMU Selection)
- Imagine you have 10 Helper Drones, but you only need 6 to cover the area.
- Random Choice: Picking 6 at random might pick 6 weak drones.
- The Paper's Algorithm: It looks at every drone, checks their signal strength, and picks the specific 6 that will cause the fewest "outages" (failed connections). It also pairs the ground users smartly so the strong and weak users help each other.
- Result: This method significantly reduces the chance of a dropped connection compared to random selection.
Game 2: Finding the Perfect Hover Spot
- Where should a Helper Drone hover? Too far from the Boss? No energy. Too far from the people? Bad signal.
- The Paper's Algorithm: It calculates the exact 3D position (X, Y, and Z coordinates) and the exact amount of energy to "steal" (the Power Splitting factor) to get the best balance.
- Result: They found a smart way to calculate this spot that is faster and more accurate than previous methods.

The Bottom Line

The paper proves that by combining hierarchical drones, smart message splitting (RSMA), and realistic energy harvesting, you can build a much more reliable network.

RSMA beats the old "NOMA" method by about 3–4 dB (a significant jump in signal quality).
Drone-to-Drone energy harvesting beats trying to harvest energy from the ground by about 8–12 dB because the air is clearer and there are fewer obstacles.
The Algorithms ensure you aren't wasting time with bad drones or bad positions.

In short, they built a mathematically rigorous "recipe" for a drone network that is tougher, smarter, and more energy-efficient, even when the equipment isn't perfect.

Technical Summary: RSMA‑Enabled Hierarchical UAV Networks with Non‑Linear Energy Harvesting

Problem Statement
Uncrewed aerial vehicles (UAVs) are pivotal for enhancing connectivity in sixth-generation (6G) networks, particularly in disaster relief and civil applications. However, multi-UAV systems face critical limitations regarding onboard energy, which restricts mission duration and communication continuity. While wireless energy harvesting (EH) offers a sustainable solution, existing literature often relies on linear EH models and ideal system assumptions. Furthermore, the integration of advanced multiple access techniques, specifically Rate-Splitting Multiple Access (RSMA), into hierarchical UAV networks remains under-explored, particularly under practical constraints such as hardware impairments (HWI) and imperfect channel state information (ICSI). Most existing studies focus on single-UAV architectures or ideal conditions, failing to capture the complexity of multi-UAV deployments where inter-user interference and heterogeneous channel conditions are prevalent.

Methodology
The authors propose a novel hierarchical ad hoc UAV network architecture comprising a Cluster Head UAV (CHU) acting as the source and multiple Cluster Member UAVs (CMUs) acting as relays to ground Internet of Things (IoT) devices.

System Model: The CHU transmits signals to CMUs, which harvest energy from these signals using a Power Splitting (PS) protocol. The harvested energy powers the CMUs to relay information to ground IoT devices. The system operates under Nakagami- $m$ fading channels.
Practical Constraints: The analysis incorporates non-linear EH models to reflect the saturation characteristics of practical RF circuits, alongside the effects of HWI (e.g., oscillator phase noise, amplifier distortion) and ICSI.
Access Scheme: RSMA is employed, splitting user messages into common and private components to manage interference more flexibly than Non-Orthogonal Multiple Access (NOMA).
Analytical Derivation: The paper derives closed-form expressions for the outage probability of individual ground IoT devices, single CMUs, and the overall system outage probability. Approximate expressions for high transmit power regimes are also provided.
Optimization: Two optimization problems are formulated:
1. Joint CMU and IoT Selection (OPM-JCIS): Aims to minimize the overall system outage probability by selecting optimal CMUs and pairing ground IoT devices based on channel conditions.
2. Joint Position and PS Factor Optimization: Aims to minimize the CMU outage probability by jointly optimizing the 3D hovering coordinates of the CMU and the power splitting factor ( $\rho$ ). Two algorithms are proposed to solve this non-convex problem: a sub-optimal decomposition method (CPPF) and a metaheuristic approach (JCPPM).

Key Contributions

Novel Framework: This is the first study to investigate a hierarchical ad hoc UAV network integrating non-linear EH, RSMA, HWI, and ICSI simultaneously.
Analytical Insights: The authors provide rigorous outage probability derivations that account for the saturation effects of non-linear EH and the degradation caused by hardware imperfections and channel estimation errors.
Optimization Algorithms: The paper introduces the OPM-JCIS algorithm for user pairing and CMU selection, and the JCPPM metaheuristic algorithm for joint position and power splitting optimization, offering solutions with varying computational complexities.
Performance Benchmarking: The proposed system is compared against benchmarks including NOMA-based hierarchical UAVs and ground-based EH systems.

Results

Analytical Validation: Simulation results confirm the accuracy of the derived analytical expressions, showing a close match with numerical simulations.
Impact of Constraints: The presence of HWI and ICSI significantly degrades system performance, with ICSI having a more pronounced negative impact. The non-linear EH model introduces an outage probability floor at high transmit powers due to saturation, unlike linear models which predict continuous improvement.
RSMA vs. NOMA: The proposed RSMA-based system outperforms NOMA-based systems, achieving approximately 3–4 dB gain in outage probability. This advantage is attributed to RSMA's superior interference management capabilities.
UAV vs. Ground EH: The proposed UAV-to-UAV energy harvesting architecture significantly outperforms traditional ground-based EH systems, showing gains of approximately 8 dB (NOMA) and 12 dB (RSMA). This is attributed to the favorable Line-of-Sight (LoS) air-to-air propagation conditions.
Optimization Efficacy: The OPM-JCIS algorithm demonstrates superior performance over random selection (RSEL) and scenarios where all CMUs are active, effectively balancing coverage and reliability. The JCPPM algorithm achieves the lowest CMU outage probability among the tested optimization methods by jointly optimizing position and power splitting, outperforming the CPPF and HGCD algorithms.

Significance
The paper claims that its proposed framework provides a more realistic and efficient communication architecture for future UAV-assisted networks. By integrating non-linear EH modeling with RSMA under practical hardware and channel constraints, the study addresses critical gaps in current literature regarding the sustainability and reliability of multi-UAV systems. The findings underscore that optimizing CMU selection, IoT pairing, and EH parameters is essential for enhancing system efficiency. The authors conclude that while reducing outage probability does not equate to maximizing energy harvesting, a balanced design is crucial for reliable operation. This work lays the groundwork for future research into mobility-induced effects and machine learning-based real-time optimization.