⚡ electrical engineering

Height-Dependent Spectrum Activity Measurements and Modeling: A Case Study with FM Radio Bands

This paper presents a height-dependent spectrum activity measurement and modeling framework using a helikite-mounted SDR, demonstrating through an FM radio case study that an altitude-aware path loss model accurately captures the transition from non-line-of-sight to line-of-sight regimes and significantly outperforms standard free space models in complex urban environments.

Original authors: Sung Joon Maeng, Amir Hossein Fahim Raouf, Ozgur Ozdemir, İsmail Güvenç, Mihail L. Sichitiu

Published 2026-02-25

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Sung Joon Maeng, Amir Hossein Fahim Raouf, Ozgur Ozdemir, İsmail Güvenç, Mihail L. Sichitiu

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 listen to your favorite radio station while flying a kite. If you hold the kite low, near the ground, the buildings and trees around you block the signal, making the music crackle and fade. But as you fly the kite higher and higher, eventually you pop above the rooftops. Suddenly, the signal becomes crystal clear and stays strong, no matter how much higher you go.

This paper is essentially a detailed study of that exact phenomenon, but instead of a child's kite, the researchers used a giant, tethered balloon called a helikite carrying a high-tech radio receiver.

Here is the breakdown of their adventure, explained simply:

1. The Problem: The "Vertical" Blind Spot

For a long time, engineers have been great at predicting how radio signals travel across the ground (like a map). They know how far a signal goes from a tower to your car. However, they haven't paid much attention to the vertical dimension (up and down).

With the rise of drones and flying taxis, we need to know: What happens to radio signals when a device is 50 feet in the air? 100 feet? 500 feet? Old models assumed the air was empty and signals traveled in a straight line perfectly, but in a city, that's not true. Buildings block signals, creating "shadows" near the ground.

2. The Experiment: The Sky-High Radio Lab

To figure this out, the team attached a Software-Defined Radio (SDR)—a super-smart radio that can listen to many frequencies at once—to a helikite. They flew this balloon over a university campus in North Carolina for about 90 minutes.

The Range: They listened to everything from very low radio waves (FM radio) all the way up to high-speed 5G frequencies (6 GHz).
The Focus: To make sure their math was right, they focused on FM Radio. Why? Because FM stations are like open books. Everyone knows exactly where they are, how loud they broadcast, and what their antennas look like. It's like having a "cheat sheet" to check if their measurements are accurate.

3. The Discovery: The "Magic Altitude"

As they pulled the data back, they saw a very clear pattern, which they call the "LoS Transition" (Line-of-Sight Transition).

The "Shadow Zone" (Low Altitude): When the balloon was low (below about 50 meters or 160 feet in the city), the signal was weak and unpredictable. The buildings were acting like walls, blocking the radio waves. This is the Non-Line-of-Sight (NLoS) zone.
The "Sky Zone" (High Altitude): Once the balloon climbed above the rooftops (around 50 meters), the signal suddenly got strong and stopped changing. The balloon was now "seeing" the radio tower directly, with no buildings in the way. This is the Line-of-Sight (LoS) zone.

The Analogy: Imagine you are in a crowded concert hall (the city). If you stand on the floor (low altitude), people in front of you block your view of the stage (the radio tower). But if you stand on a ladder and rise above the crowd (high altitude), you have a perfect, unobstructed view. The paper found the exact height of that ladder.

4. The Solution: A New Rulebook

The researchers realized that the old "Free Space Path Loss" model (which assumes signals travel perfectly through empty air) was wrong for low-flying drones in cities. It predicted signals would be stronger than they actually were near the ground.

So, they created a new mathematical model that acts like a two-speed gear system:

Gear 1 (Low Altitude): The model accounts for the "messy" city environment where signals bounce off buildings and get blocked.
Gear 2 (High Altitude): Once you cross the "Magic Altitude" (the transition point), the model switches to the simple, clean physics of open-air travel.

5. Why Does This Matter?

This isn't just about FM radio. This is the foundation for the future of flying technology.

Drones and Delivery: If you want to send a drone to deliver a pizza, you need to know if it will lose its connection to the control tower when it flies between skyscrapers.
5G and 6G: Future networks will rely on drones and high-altitude platforms to beam internet to the sky. This paper gives engineers a better map to predict where those signals will work and where they will fail.

The Bottom Line

The paper proves that altitude matters. You can't just look at a 2D map to understand radio signals anymore; you need a 3D model. By using a balloon and some math, they figured out exactly where the "city shadow" ends and the "clear sky" begins, helping us build a safer, more connected future for the skies above our cities.

1. Problem Statement

The increasing deployment of aerial platforms (drones, high-altitude platforms) for next-generation communications (5G/6G) necessitates a deeper understanding of 3D signal behavior, specifically how signal propagation changes with altitude.

The Gap: Traditional spectrum models often assume 2D ground-based propagation or rely on standard Free Space Path Loss (FSPL) models. These fail to capture the critical transition from Non-Line-of-Sight (NLoS) to Line-of-Sight (LoS) regimes as an aerial receiver ascends through urban obstacles (buildings, trees).
The Challenge: There is a lack of empirical data and modeling that specifically addresses the altitude threshold where signal behavior shifts, particularly for complex urban topologies. Furthermore, validating models for proprietary bands (like 5G/6G) is difficult due to a lack of public transmitter data.

2. Methodology

The authors conducted a comprehensive spectrum monitoring campaign using a helikite-mounted Software-Defined Radio (SDR) and GPS receiver.

Experimental Setup:
- Platform: A helikite (a tethered aerostat) carrying an SDR and GPS.
- Locations: Urban environment (NC State University campus, "Packapalooza" festival) and a rural environment (Lake Wheeler).
- Frequency Range: Swept from 88 MHz to 6 GHz, with a specific focus on the FM Radio Band (87–108 MHz) for validation.
- Duration: 90-minute flights involving altitude changes from ~5m to ~140m.
- Data Collection: 364 spectrum sweeps were performed.
Why FM Radio?
- FM bands were chosen as a strategic case study because transmitter locations, effective radiated power (ERP), and antenna patterns are publicly available. This provides a reliable "ground truth" for analytical comparison, which is often unavailable for proprietary cellular bands.
Modeling Approach:
- The authors analyzed the relationship between altitude and received signal strength.
- They observed a distinct transition point where signal strength stops increasing and plateaus, indicating a shift from NLoS (blocked by obstacles) to LoS (clear view).
- They proposed a new Altitude-Dependent Break-Point Path Loss Model to mathematically capture this transition.

3. Key Contributions

Identification of Altitude Thresholds: The study empirically identifies specific altitude thresholds (e.g., 50m in urban areas) where signal propagation transitions from NLoS to LoS.
Novel Path Loss Model: The authors propose a continuous, altitude-dependent path loss model (Equation 1) that incorporates a transition function. This model smoothly shifts from a high path-loss exponent (NLoS) to the standard Free Space Path Loss (FSPL) exponent of 2 once the LoS threshold ( $h_0$ ) is reached.
Validation Framework: The paper demonstrates a scalable methodology using FM bands to validate aerial spectrum models. This framework can be extended to emerging technologies (5G/6G) once transmitter data becomes available.
Empirical Characterization: The study provides extensive datasets characterizing spectrum activity in both time and altitude domains, distinguishing between urban and rural propagation dynamics.

4. Key Results

Urban Environment (Packapalooza):
- Transition Point: A clear transition from NLoS to LoS was observed at approximately 50 meters. Below this altitude, signal power increased with height; above it, power plateaued.
- Model Accuracy: The proposed break-point model significantly outperformed the standard FSPL model in urban settings. The FSPL model failed to predict the low-altitude signal attenuation caused by building blockage.
- Error Reduction: The Cumulative Distribution Function (CDF) of modeling errors showed the proposed model had significantly lower error margins, especially at low altitudes where NLoS effects dominate.
Rural Environment (Lake Wheeler):
- Variable Thresholds: In rural areas, the LoS threshold varied by station direction. Stations with clear lines of sight (e.g., WNCB, WBBB) showed LoS conditions as low as 10 meters, fitting the FSPL model well. Stations with semi-urban obstructions (e.g., WQDR) showed a delayed transition up to 80 meters.
- Model Selection: The study confirmed that a hybrid approach is necessary: using the break-point model for obstructed paths and FSPL for clear rural paths.
Statistical Analysis:
- NLoS vs. LoS: NLoS conditions exhibited lower mean signal power but higher standard deviation due to multipath scattering and reflection. LoS conditions showed more uniform power distribution.
- Calibration: By comparing analytical FSPL results with measurements, the authors identified a systematic offset (approx. 29–38 dB) attributed to SDR receiver calibration errors and antenna gain variations.

5. Significance and Impact

Enabling Aerial Spectrum Management: The proposed model provides a more accurate framework for Radio Dynamic Zones (RDZ) and spectrum sharing. It allows regulators and network operators to predict signal interference and coverage for aerial users more reliably than current 2D models.
3D Network Planning: The findings are critical for the deployment of 5G/6G networks in the vertical dimension (e.g., drone delivery, aerial base stations), ensuring that altitude is treated as a primary variable in link budget calculations.
Scalability: By validating the methodology on FM bands, the paper establishes a proof-of-concept for modeling higher frequency bands (mmWave, sub-6GHz) where transmitter data is currently restricted.

In conclusion, this paper bridges the gap between theoretical propagation models and empirical reality for aerial communications, proving that altitude is a critical determinant of signal behavior and that standard models are insufficient for complex 3D environments.