Modeling and Analysis of Air-to-Ground Cellular KPIs in a 5G Testbed using Android Smartphones

Simran Singh, Anıl Gürses, Özgür Özdemir, Ram Asokan, Mihail L. Sichitiu, İsmail Güvenç, Rudra Dutta, Magreth Mushi · 2026

Read Modeling and Analysis of Air-to-Ground Cellular KPIs in a 5G Testbed using Android Smartphones by Simran Singh, Anıl Gürses, Özgür Özdemir, Ram Asokan, Mihail L. Sichitiu, İsmail Güvenç, Rudra Dutta, Magreth Mushi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-04452.txt.

Introduction

Singh and colleagues model air-to-ground link quality for drones. Flights at AERPAW Lake Wheeler track how a base station's indicators vary with drone altitude, distance, elevation and azimuth. They log 4G and 5G physical-layer indicators and app throughput using two Android phones. Signal strength is checked against free-space path-loss predictions from the base station's antenna patterns, with polynomial curve fits derived. Random forests, gradient boosting and neural networks model indicators from drone position. The models aid design, simulation and deployment of cellular-linked drones.

Catalog record

CREATORSimran Singh, Anıl Gürses, Özgür Özdemir, Ram Asokan, Mihail L. Sichitiu, İsmail Güvenç, Rudra Dutta, Magreth Mushi
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE32039
LICENSEhttp://creativecommons.org/licenses/by-sa/4.0/
RIGHTSCC BY SA
SOURCEarXiv

Source: https://arxiv.org/abs/2604.04452

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Modeling and Analysis of Air-to-Ground Cellular KPIs in a 5G Testbed using Android Smartphones Simran Singh, Anıl Gürses, Özgür Özdemir, Ram Asokan, Mihail L. Sichitiu, İsmail Güvenç, Rudra Dutta, Magreth Mushi Abstract The integration of cellular communication with Unmanned Aerial Vehicles (UAVs) extends the range of command and control (C2) and payload communications of autonomous UAV applications. Accurate modeling of this air-to-ground wireless environment aids UAV mission planning. Models built on and insights obtained from real-life experiments intricately capture the variations in air-to-ground link quality with UAV position, offering more fidelity for simulations and system design than those that rely on generic theoretical models designed for ground scenarios or ray-tracing simulations. In this work, we conduct multiple aerial flights at the Aerial Experimentation and Research Platform for Advanced Wireless (AERPAW) Lake Wheeler testbed site to study the variation in key performance indicators (KPIs) of a private 4G/5G cellular base station (BS) with the UAV’s altitude, distance from the BS, elevation, and azimuth relative to the BS. Variations in both 4G and 5G physical layer KPIs and application layer throughput are logged and analyzed, using two Android smartphones: a Keysight Nemo device, with enhanced KPI access, obtained through a rooted operating system (OS), and a standard smartphone running a custom application that utilizes open-source Android APIs. The BS, provided by Ericsson, consisted of two sectors, such that handover events occur during UAV flights. The observed signal strength measurements are compared to theoretical predictions from free space path loss models that incorporate the cell tower’s antenna radiation patterns. Furthermore, mathematical model parameters for polynomial curve approximations are derived to fit the observed data. Light machine learning approaches, namely random forests, gradient boosting regressors and simple neural networks, are also used to model KPI behaviour as a function of UAV position relative to the BS. Model performance is evaluated using metrics such as goodness-of-fit, mean average error (MAE), and root mean square error (RMSE). The insights and models generated from real-life experiments in this study can serve as accurate and valuable tools in the design, simulation, and deployment of cellular communication-based UAV systems. I Introduction The use of Unmanned Aerial Vehicles (UAVs) has seen a significant increase in recent years, with applications ranging from aerial photography and surveillance [ 1 ] to delivery services and emergency disaster response [ 2 ] . Such applications rely on communication links between the UAV and ground control stations for real-time transmission of mission-specific and command and control communications (C2). 4G and 5G cellular technologies are a viable means of serving such communication requirements, due to the widely deployed and standardized infrastructure, coupled with mature and industry-tested secure protocol stacks [ 3 ] . Cellular networks enable UAVs to operate beyond the line of sight and in areas where traditional communication methods may be unavailable or unreliable. The success of such autonomous cellular-communication based UAV systems requires accurate modeling of the air-to-ground wireless environment. Models that accurately capture the variations in air-to-ground link quality with UAV position can aid in mission planning, route optimization, and predictable operation and control. While theoretical models and ray-tracing simulations can provide a valuable reference, insights and models derived from real-world measurements can reflect the unique scenarios that arise in air-to-ground communication links, as UAV flight trajectories extend beyond the antenna’s main lobes. Besides signal strength models, cellular network key performance indicators (KPIs) such as reference signal received power (RSRP), reference signal received quality (RSRQ), channel quality index (CQI), and application layer throughput provide quantifiable and interpretable metrics for system design. 4G/LTE and 5G cellular KPIs have been analyzed, for terrestrial environments, [ 4 ] , [ 5 ] , both with public cellular networks [ 1 ] , and using dedicated experimental testbeds [ 6 ] . The benefits of integrating cellular technology with UAVs [ 7 ] , has motivated such performance analysis for aerial scenarios as well [ 8 ] . For example, authors in [ 9 ] conducted experiments to evaluate the performance of LTE networks to support UAV applications, while [ 10 ] analyzed the performance of 5G networks to support UAV applications. 4G LTE RSRP and RSRQ KPIs were modeled for suburban aerial scenarios, based on empirical experiments, as a function of the two-dimensional (2D) distance of the UAV from the base station (BS) and its elevation angle in [ 11 ] . Polynomial, logarithmic, and support vector regression models were employed. However, 5G signal strength was not modeled nor was the free space path loss (FSPL) model evaluated. An SDR-based channel sounder was built and utilized in [ 12 ] to measure channel impulse response and path loss, which was compared with reference statistical models. Similarly, the channel impulse response was modeled in [ 13 ] using a SDR-based channel sounding with a fixed-wing aircraft, for a rural region, using Rician fading model. However cellular KPIs such as channel rank and channel quality index (CQI) were not measured in these studies. The end to end performance of cellular networks optimized for terrestrial environments was evaluated in [ 14 ] , in terms of meeting the throughput and latency requirements of video transmission from a UAV, and the reliability gaurantees of the control and command (C2) link. However, signal strength metrics were not modeled and physical layer KPIs were not analyzed. In this work, we conduct multiple aerial flights at the Aerial Experimentation and Research Platform for Advanced Wireless (AERPAW) [ 15 ] Lake Wheeler testbed site to study the variation in KPIs of a private 5G cellular BS with the UAV’s altitude, distance from the BS, elevation, and azimuth relative to the BS. Our contributions are as follows: • Experimental analysis of UAV 4G and 5G KPIs at AERPAW’s testbed with a private BS. • Comparison of FSPL, polynomial, and ML models in predicting 5 G RSRP as a function of the three dimensional (3D) distance of the UAV from the BS, and elevation and azimuth angle of the UAV with respect to the BS. • Open-source data and Android software for reproducibility and future research. The rest of this paper is organized as follows. The AERPAW measurement testbed and the Android measurement software is explained in Section II . The signal strength prediction models are defined in Section III , and their performance is evaluated in Section IV . 4G and 5G cellular KPIs for various UAV trajectories are analyzed in Section V and Section VI concludes this work. Fig. 1 : AERPAW Lake Wheeler Road Field Labs, where the UAV flights were conducted. The site has rural terrain characteristics. ((a)) ((b)) Fig. 2 : Radiation pattern of the 5G antenna used in the AERPAW Ericsson BS, in the azimuth (a) and elevation (b) planes [ 16 ] . Fig. 3 : PawPrints architecture and components [ 17 ] . II AERPAW 4G/5G Measurement Testbed In this section, we provide details of the AERPAW testbed, configuration of the 4G/5G BS and antenna radiation pattern, and the Android software used to collect radio and throughput measurements. The flights were conducted on AERPAW’s Lake Wheeler Road Field Labs, which is located in Raleigh, NC and resembles a rural terrain, shown in Fig. 1 . The AERPAW

Read the complete text