Bridging the Terrestrial to Aerial Cellular Coverage Gap

39 concepts28 papers1 proposal

Existing commercial cellular networks (LTE and 5G) are highly optimized for a two-dimensional world. To maximize street-level signal strength and minimize interference between neighboring cells, network operators physically and electrically tilt base station antennas downward. While this design works exceptionally well for ground users, it creates a severe structural mismatch for Unmanned Aerial Vehicles (UAVs), drones, and low-altitude aircraft operating in three-dimensional space.

As aerial users ascend, they leave the primary coverage zones of terrestrial networks, exposing a significant coverage gap in the sky. To integrate drones safely into next-generation networks, researchers are analyzing the physics of this mismatch and developing infrastructure and algorithmic solutions to bridge the gap.

The Geometry of the Aerial Coverage Gap

To understand why aerial connectivity degrades, we must look at the radiation pattern of a standard base station antenna. An antenna does not spray signals equally in all directions; it concentrates energy into a "main lobe" directed slightly downward toward the ground.

Standard Base Station Configuration:

         /|   (Signal Nulls / Weak Sidelobes)   -->  [ Drone (High Altitude) ]
        / |  /
       /  | / 
      [===] 
       \  | 
        \ | \ 
         \|  \--> (Main Lobe: Downward-Tilted)  -->  [ Ground User ]

When a drone flies at a significant altitude, it operates entirely above this main lobe. As a result, the drone is forced to communicate via the antenna's sidelobes—secondary, much weaker bands of radiation that escape upward—or it falls into signal nulls, which are blind spots where the radiated power drops to near zero.

As Aziz Altaf Khuwaja et al. point out in A Survey of Channel Modeling for UAV Communications, these coverage gaps mean that a drone can experience intermittent connectivity and sudden signal drops even when it is physically close to a base station tower. Empirical flight tests, such as those conducted by M. Gharib et al. in 5G Wings: Investigating 5G-Connected Drones Performance in Non-Urban Areas, confirm this altitude penalty: as a drone climbs above the main signal beam, the received signal power and overall throughput drop significantly.

The Line-of-Sight Paradox

Intuitively, one might assume that flying higher would improve wireless performance. After all, ascending lifts the drone above terrestrial obstacles like buildings, trees, and terrain, creating a clean, unobstructed Line-of-Sight (LoS) path to the base station.

However, this introduces a fundamental paradox in aerial cellular networking: favorable propagation conditions actually worsen network interference.

In a typical ground network, buildings act as natural shields, blocking signals from distant towers and allowing frequencies to be reused nearby without issue. But as Fernando Rosas demonstrates in Coexistence of Terrestrial and Aerial Users in Cellular Networks, a high-altitude drone lacks this structural shielding. Because it has an unobstructed view of the horizon, it simultaneously establishes LoS paths with dozens of surrounding base stations.

This creates a massive influx of inter-cell interference. The drone struggles to hear its serving base station because it is being bombarded by signals from every other tower in the region. When you combine this high interference with the fact that the drone is already relying on weak, sub-optimal antenna sidelobes (Hongguang Sun et al. in CoMP Transmission in Downlink NOMA-Based Cellular-Connected UAV Networks), the Signal-to-Interference-plus-Noise Ratio (SINR) plummets, leading to unreliable command-and-control links.

Hardware and Infrastructure Solutions

To resolve this mismatch, researchers are exploring how to reshape cellular infrastructure to serve both ground and sky simultaneously.

Up-Tilted Antennas and Dedicated Tiers

One direct approach is to equip existing cell towers with a secondary set of dedicated, upward-tilted antennas. Md Moin Uddin Chowdhury et al. demonstrate in Ensuring Reliable Connectivity to Cellular-Connected UAVs with Up-tilted Antennas and Interference Coordination that by using optimization algorithms to tune the angles of these upward beams alongside inter-cell interference coordination, networks can provide a reliable "blanket" of coverage for drones without degrading performance for users on the ground.

Alternatively, operators can deploy a multi-tier network architecture. Boris Galkin et al. show in Experimental Evaluation of a UAV User QoS from a Two-Tier 3.6GHz Spectrum Network that while traditional macrocells struggle to support high-altitude users, a dense secondary tier of urban small cells can offer highly stable throughput and low handover rates for drones.

Dedicated Aerial Infrastructure

In areas where ground networks are sparse, relying on sidelobes or reflections becomes impossible (Seongjoon Kang et al. in Millimeter-Wave UAV Coverage in Urban Environments). To address this, Yujie Qin et al. propose a dedicated infrastructure strategy in Coverage Analysis and Trajectory Optimization for Aerial Users with Dedicated Cellular Infrastructure, suggesting the deployment of specialized, low-altitude aerial base stations mounted directly onto existing roadside furniture like lampposts and traffic lights to provide targeted, upward-directed coverage.

Dedicated Multi-Beam Infrastructure:

                      [ Drone ] 
                         ^
                         | (Upward-Tilted Aerial Beam)
                       /---\
                       |   |
                       |===|  <-- Dual-Service Base Station
                       |   |
                         | (Downward-Tilted Terrestrial Beam)
                         v
                  [ Ground User ]

Algorithmic Adaptations in the Sky

Modifying physical infrastructure takes time. In the interim, drones must be smart enough to navigate the patchy, irregular signal landscapes created by current down-tilted networks. Because antenna sidelobes create highly complex, fluctuating 3D patterns of signal strength (Yuanjian Li et al. in Path Planning for Cellular-Connected UAV: A DRL Solution with Quantum-Inspired Experience Replay), drones cannot simply fly in a straight line and assume their connection will hold.

Connectivity-Aware Path Planning

To prevent drones from flying into signal dead zones, researchers are using Deep Reinforcement Learning (DRL) to design connectivity-aware trajectories. For instance, Xin Wang et al. introduce a framework in Joint optimization of UAV communication connectivity and obstacle avoidance in urban environments using a double-map approach that constructs internal radio maps during flight. This allows the drone to predict and actively navigate around "coverage black holes" in complex urban environments, balancing the shortest physical path with the safest communication path.

Intelligent Cell Association

On the ground, a phone connects to whichever tower has the strongest signal. In the air, this strategy fails because the closest tower might have its main beam pointed away from the drone, while a distant tower's sidelobe might offer a cleaner link.

To solve this, Boris Galkin et al. developed REQIBA in REQIBA: Regression and Deep Q-Learning for Intelligent UAV Cellular User to Base Station Association. Instead of relying on raw signal strength, this approach uses machine learning to factor in surrounding building topologies and interference patterns, allowing the drone to intelligently select and maintain connections with the most stable base station, drastically reducing dropped connections and unnecessary handovers.

Key Foundational Work

Coexistence of Terrestrial and Aerial Users in Cellular Networks (2017) — contributed 1 concept elements to this direction.
CHANNEL MODELLING AND PERFORMANCE ANALYSIS OF V2I COMMUNICATION SYSTEMS IN BLIND BEND SCATTERING ENVIRONMENTS (2014) — contributed 1 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_aerial_terrestrial_infrastructure_mismat_02aec6ee,
  title = {aerial-terrestrial infrastructure mismatch},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/aerial-terrestrial-infrastructure-mismatch/rdr_02aec6ee8cc945cd9f6d5143aabef0c8},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}

This exact version (version 1, frozen text):

@misc{lacuna_aerial_terrestrial_infrastructure_mismat_02aec6ee_v1,
  title = {aerial-terrestrial infrastructure mismatch},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/aerial-terrestrial-infrastructure-mismatch/rdr_02aec6ee8cc945cd9f6d5143aabef0c8/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake