Energy Efficient and Throughput Maximization of UAV-Assisted RIS in Aerial-Terrestrial Communication Networks

Technical Summary: Energy Efficient and Throughput Maximization of UAV-Assisted RIS in Aerial-Terrestrial Communication Networks

Problem Statement
The integration of Unmanned Aerial Vehicles (UAVs) and Reconfigurable Intelligent Surfaces (RIS) is identified as a transformative paradigm for 6G networks, offering solutions to terrestrial network limitations such as restricted coverage, insufficient capacity, and high deployment costs. While UAVs provide mobility and flexible deployment, they face challenges in urban environments where Line-of-Sight (LoS) is obstructed. Conversely, RIS technology offers software-controlled wireless environments to manipulate signal propagation without additional Radio Frequency (RF) chains. However, the existing literature presents conflicting findings regarding the actual performance improvements in Energy Efficiency (EE) and throughput of RIS-assisted UAV systems. Furthermore, there is a lack of consensus on optimal design parameters (e.g., RIS element count, UAV altitude) and a significant absence of standardized statistical reporting, which hinders quantitative meta-analysis and fair comparison between different optimization approaches.

Methodology
This study conducts a systematic review of 47 peer-reviewed studies published between 2020 and 2026. The research adheres to PRISMA 2020 guidelines, utilizing a search strategy across IEEE Xplore, Scopus, Web of Science, and arXiv.org.

• Inclusion Criteria: Studies must be peer-reviewed, published between 2022–2026, provide quantitative performance comparisons, report EE and/or throughput metrics, and include at least 10 simulation iterations.
• Exclusion Criteria: Theoretical papers without simulation validation, terrestrial RIS-only systems, papers lacking means or confidence intervals, and review papers were excluded.
• Analysis Framework: The review synthesizes data on system models (network architectures, channel models, energy consumption), optimization objectives (throughput vs. EE), algorithmic approaches (DRL, SCA, AO, etc.), and reported performance gains. A critical component of the methodology involves a meta-analysis of methodological rigor, specifically checking for the reporting of standard deviations, confidence intervals, and independent replication counts.

Key Contributions

1. Comprehensive Synthesis: This is the first systematic review to synthesize results specifically on energy efficiency and throughput optimization in UAV-assisted RIS networks, analyzing 47 distinct studies.
2. Evidence-Based Design Guidelines: The paper establishes specific design recommendations, including an optimal RIS element count of 128–256 and an optimal UAV altitude of 100–120 meters for urban environments to maximize LoS probability.
3. Algorithmic Taxonomy: The review categorizes optimization algorithms, highlighting a shift toward Deep Reinforcement Learning (DRL) methods, which constitute 40.4% of recent studies.
4. Methodological Critique: A major contribution is the identification of a critical gap in statistical reporting. The review reveals that 68% of studies do not report standard deviations and 91.5% do not specify independent replication counts, rendering quantitative meta-analysis impossible.
5. Standardization Proposal: The authors propose a set of reporting standards (means, standard deviations, replication counts, confidence intervals) to ensure reproducibility and enable future quantitative synthesis.

Results

• Performance Gains: RIS-assisted UAV systems demonstrate reported throughput gains ranging from 15% to 55% and energy efficiency improvements from 12% to 58% over baseline systems.
• Algorithm Performance: DRL-based algorithms (specifically DDPG, DQN, and PPO) demonstrate superior performance in dynamic environments, achieving up to 34.48% throughput gain and 40% EE improvement compared to conventional methods.
• Parameter Optimization:
  • RIS Elements: The optimal range for the best cost-benefit trade-off is 128–256 elements. Gains show diminishing returns beyond 257 elements.
  • UAV Altitude: In urban settings, the optimal altitude is 100–120 meters, balancing signal strength against path loss and maintaining high LoS probability (0.88–0.94).
  • Energy Consumption: Propulsion energy accounts for 60–80% of total energy consumption in UAVs, making trajectory optimization critical.
• Statistical Deficiencies: The review found that 95.7% of studies do not report confidence intervals, and 91.5% lack independent replication counts. This absence of statistical rigor prevents the calculation of effect sizes (e.g., Hedges' g) and heterogeneity statistics.

Significance and Claims
The paper claims its significance lies in providing a comprehensive, evidence-based reference for the design and deployment of energy-efficient UAV-RIS communication systems for 6G networks. It moves beyond narrative surveys by offering a quantitative distribution of reported gains and a critical assessment of the field's methodological maturity.

The authors explicitly state that while the volume of research has grown, the quality of statistical reporting has not kept pace, limiting the field's ability to draw definitive conclusions about the relative effectiveness of different approaches. Consequently, the paper argues that the immediate priority for the research community is not just technical innovation, but the adoption of standardized reporting protocols. The findings serve as a call to action for researchers to include statistical rigor (standard deviations, replication counts) and for journal editors to enforce these standards, thereby enabling future quantitative meta-analyses and accelerating scientific progress in UAV-RIS networks.