Source-linked AI summary
UAV Air-to-Ground Channel Characterization for mmWave Systems
Wahab Khawaja, Ozgur Ozdemir, Ismail Guvenc
TL;DR
The paper characterizes mmWave air-to-ground channels for UAV communications across frequencies, heights, and environments. Using ray tracing and a USRP-based sounder, it finds that RSS often follows a two-ray model, while scatterers and UAV height strongly affect RSS and RMS-DS.
Problem
MmWave UAV communications require air-to-ground channel characterization because large path loss makes LOS or favorable NLOS connectivity important for link quality.
Method
The study uses ray tracing simulations at 28 GHz and 60 GHz across urban, suburban, rural, and over-sea environments, complemented by a USRP-based channel sounder.
Results
RSS generally follows the two-ray propagation model, while scatterers and UAV height substantially influence RSS fluctuations and RMS-DS across environments.
Takeaways & Limitations
Environmental scatterer density and UAV altitude should be considered when characterizing mmWave UAV channels and designing receivers.
Abstract
from arXiv · showhide
Communication at mmWave bands carries critical importance for 5G wireless networks. In this paper, we study the characterization of mmWave air-to-ground (AG) channels for unmanned aerial vehicle (UAV) communications. In particular, we use ray tracing simulations using Remcom Wireless InSite software to study the behavior of AG mmWave bands at two different frequencies: 28~GHz and 60~GHz. Received signal strength (RSS) and root mean square delay spread (RMS-DS) of multipath components (MPCs) are analyzed for different UAV heights considering four different environments: urban, suburban, rural, and over sea. It is observed that the RSS mostly follows the two ray propagation model along the UAV flight path for higher altitudes. This two ray propagation model is affected by the presence of high rise scatterers in urban scenario. Moreover, we present details of a universal serial radio peripheral (USRP) based channel sounder that can be used for AG channel measurements for mmWave (60 GHz) UAV communications.
I. INTRODUCTION
The paper motivates mmWave UAV air-to-ground links for high-throughput 5G applications and studies their propagation at 28 GHz and 60 GHz. It combines ray-tracing analysis across environments with a compact USRP-based channel sounder for future 60 GHz measurements.
- UAVs can support mobile hotspots, content caching, and first-responder communications because of their agile three-dimensional mobility.
- mmWave bands can sustain UAV data-rate demands, but their large path loss makes line-of-sight or favorable non-line-of-sight connectivity important.
- The study addresses limited prior evidence on mmWave UAV air-to-ground propagation by analyzing 28 GHz and 60 GHz links across urban, suburban, rural, and over-sea environments.
- Simulations use a fixed ground transmitter and a UAV receiver flying a linear path at multiple heights, with RSS and RMS-DS evaluated for both frequencies.
- RSS generally follows a two-ray model, but high-rise urban scatterers produce rapid fluctuations and make direct application of that model unreliable.
- The paper also presents a USRP X310-based 60 GHz channel sounder whose compact setup can be airlifted by UAVs.
II. RAY TRACING SIMULATIONS FOR MMWAVE UAVS
The paper uses deterministic ray tracing to characterize mmWave UAV air-to-ground channels while modeling UAV motion over representative trajectories. The simulations produce RSS and RMS-DS results for different environments and UAV heights.
- Ray tracing provides a deterministic way to characterize mmWave channels in different scenarios when UAV-based mmWave measurements are challenging.
- The simulations use Remcom Wireless InSite to imitate real-time UAV motion over a specified trajectory.
- The analysis first defines the ray-tracing scenarios and then evaluates received signal strength and root mean square delay spread.
A. Ray Tracing Scenarios
The ray-tracing scenarios vary environmental scatterers, UAV altitude, frequency, and trajectory conditions to represent urban, suburban, rural, and over-sea air-to-ground links.
- Ray-Tracing Environments: Four simulated environments are considered: urban, suburban, rural, and over sea.
- Environmental Factors: The number, material, and height of environmental scatterers—especially buildings—are key factors in air-to-ground ray tracing.
- Simulation Parameters: The terrain measures 10 km by 10 km, with a fixed 2 m transmitter and UAV heights of 2 m, 50 m, 100 m, and 150 m.
- Simulation Parameters: The UAV travels at 15 m/s along an approximately 2 km trajectory while half-wave dipole antennas and 28 GHz and 60 GHz sounding frequencies are used.
B. Ray Tracing Results for the RSS
RSS generally follows the two-ray propagation model, but scatterers, UAV height, frequency, and environment alter its fluctuations and decay behavior.
- RSS commonly follows the two-ray propagation model, with scatterers introducing environment-dependent fluctuations.Fluctuations increase near scatterers and are strongest in urban areas with many scatterers.
- 60 GHz produces more rapid two-ray maxima and minima, steeper RSS decay, and lower RSS than 28 GHz.The faster interference variation is attributed to 60 GHz's smaller wavelength, while higher-frequency losses reduce RSS.
- The simulations' relatively low UAV heights, selected because of FAA regulations, limit the observed effect of UAV height on scatterer-induced RSS.At greater heights, scatterer effects on RSS could be reduced.
- Over sea, RSS closely follows the two-ray model with minimal fluctuations because scatterers have negligible effect.Rural and suburban environments show fluctuations near scatterers, while urban areas show larger fluctuations from denser scatterers.
C. Ray Tracing Results for the RMS-DS
RMS-DS depends strongly on UAV height and environment, with urban behavior differing from rural and suburban settings. At 60 GHz, the height-related pattern resembles 28 GHz, but RMS-DS is lower.
- Urban RMS-DS is largest for most UAV heights and increases with UAV altitude.Higher UAVs move above tall buildings and receive signals scattered from more surrounding buildings.
- Rural and suburban RMS-DS decreases as UAV altitude increases, opposite to the urban trend.Buildings are shorter and more sparsely deployed, so their scattered signals are less likely to reach higher UAVs.
- The relative impact of UAV height on RMS-DS at 60 GHz is similar to that at 28 GHz.
- Fig. 5 presents RMS-DS CDFs for different UAV heights across over-sea, rural, suburban, and urban scenarios at 60 GHz.
- RMS-DS is lower at 60 GHz than at 28 GHz, attributed to larger 60 GHz path loss.
III. CHANNEL SOUNDING FOR MMWAVE UAVS
The paper motivates direct mmWave AG measurements because ray tracing provides insight but may not capture subtle propagation features as accurately as measurements. UAV payload limits require a lightweight, compact sounder.
- Propagation measurements can characterize subtle AG mmWave propagation features more accurately than ray tracing simulations.The paper therefore introduces a preliminary framework for AG channel sounding at mmWave frequencies.
A. USRP-Based Experimental Setup
The proposed 60 GHz channel sounder combines USRP X310 SDRs, frequency-conversion hardware, GNU Radio processing, and a pseudonoise-based sounding signal. Receiver processing estimates carrier offset and recovers the channel impulse response.
- The 60 GHz setup uses Pasternack TX/RX development hardware with USRP X310-generated differential I/Q signals and splitter-based conversion.
- The transmitter GNU Radio flow graph generates a degree-12 maximal-length PN sequence of length 4095 and periodically extends it.
- Fig. 6 depicts the experimental channel sounder setup, while Fig. 7 depicts the transmitter GNU Radio flow graph.
- The periodic sounding signal is transmitted by the USRP at a 25 MHz sampling rate.
- The receiver estimates carrier frequency offset from the squared BPSK signal spectrum and compensates the received signal.
- After CFO correction, a matched filter using the known sequence produces a signal representing the channel impulse response.
IV. CONCLUSIONS
The paper finds that mmWave UAV AG channels are shaped by propagation geometry, scatterers, UAV height, and frequency, and presents a lightweight USRP/GNU Radio sounder for 60 GHz measurements.
- The two-ray propagation model applies with limitations across urban, suburban, rural, and over-sea scenarios, while scatterers particularly affect urban RSS.
- RSS fluctuations with distance are higher at 60 GHz than at 28 GHz.
- RMS-DS depends strongly on UAV height and surrounding scatterer density and height.
- Suburban and rural RMS-DS decreases with increasing UAV height because scatterers become less significant at higher altitudes.
- The USRP-based sounder is lightweight enough for carriage by a DJI S-1000 octocopter for 60 GHz channel-sounding experiments.