Source-linked AI summary
From Ground to Sky: Architectures, Applications, and Challenges Shaping Low-Altitude Wireless Networks
Weijie Yuan, Yuanhao Cui, Jiacheng Wang, Fan Liu, Lin Zhou, Geng Sun, Tao Xiang, Jie Xu, Shi Jin, Sinem Coleri, Sumei Sun, Shiwen Mao, Abbas Jamalipour, Dong In Kim, Mohamed-Slim Alouini, Xuemin Shen
TL;DR
Low-altitude missions require wireless infrastructure that can operate amid dynamic environments, safety constraints, and changing mission demands. This article presents LAWN, a reconfigurable 3D architecture integrating data, control, sensing, and intelligence and computing, and surveys enabling technologies, applications, and cross-layer challenges. It concludes that LAWN provides a framework for intelligent, resilient, multifunctional low-altitude wireless systems, while current standards still lack native support for its integrated planes.
Problem
Existing aerial-communication research often treats topics such as drone-to-ground communication or trajectory design in isolation, while low-altitude missions require integrated connectivity, sensing, control, and computing.
Method
The article develops a holistic tutorial framework for LAWN, describing its four functional planes and surveying enabling technologies, applications, case studies, and cross-layer challenges.
Results
LAWN is presented as a dynamically reconfigurable 3D infrastructure whose tightly integrated planes support adaptation to mission objectives, environmental conditions, and airspace constraints.
Takeaways & Limitations
LAWN extends drones beyond communication-relay or remote-sensing roles toward an intelligent, resilient, multifunctional cyber-physical infrastructure for low-altitude operations.
Abstract
from arXiv · showhide
In this article, we introduce a novel low-altitude wireless network (LAWN), which is a reconfigurable, three-dimensional (3D) layered architecture. In particular, the LAWN integrates connectivity, sensing, control, and computing across aerial and terrestrial nodes that enable seamless operation in complex, dynamic, and mission-critical environments. Different from the conventional aerial communication systems, LAWN's distinctive feature is its tight integration of functional planes in which multiple functionalities continually reshape themselves to operate safely and efficiently in the low-altitude sky. With the LAWN, we discuss several enabling technologies, such as integrated sensing and communication (ISAC), semantic communication, and fully-actuated control systems. Finally, we identify potential applications and key cross-layer challenges. This article offers a comprehensive roadmap for future research and development in the low-altitude airspace.
I. INTRODUCTION
The article introduces LAWN as a reconfigurable 3D network integrating communication, sensing, control, and computing across aerial and terrestrial nodes. It frames LAWN as a response to dynamic, cluttered, safety-critical low-altitude environments and surveys its architecture, technologies, applications, and challenges.
- I. INTRODUCTION: Low-altitude operations combine dense aerial deployments, rapid maneuvers, changing LoS/NLoS links, and stringent control requirements below 10 ms latency with 99.999% reliability.These coupled requirements make the operational environment substantially more dynamic than conventional fixed or high-altitude systems.
- I. INTRODUCTION: Unlike isolated aerial communication systems, LAWN jointly adapts connectivity, real-time sensing, closed-loop control, and distributed intelligence across a changing network.Communication, sensing, control, and intelligence continually reshape one another to sustain safe and efficient operation.
- I. INTRODUCTION: The article surveys LAWN architectures, enabling technologies, applications, case studies, and cross-layer challenges including spectrum management, cyber-physical security, and regulation.It positions the work as a holistic tutorial addressing a gap left by prior studies focused on isolated communication or trajectory-design topics.
- A. What is the LAWN?: LAWN integrates aerial platforms and terrestrial nodes into a multifunctional, reconfigurable 3D network operating typically below 3,000 meters.Its architecture supports data transmission, control, sensing, and intelligence and computing.
- A. What is the LAWN?: The four functional planes divide responsibilities among mission-data transmission, flight control, environmental sensing, and distributed onboard–edge–cloud intelligence and computing.The data plane uses A2G, G2A, and A2A links, while the control and sensing planes support safe maneuvering and environmental awareness.
- A. What is the LAWN?: LAWN data-plane operation must adapt packet scheduling and traffic treatment to mobility-driven LoS/NLoS changes and differing service requirements.Delay-tolerant traffic can be buffered, latency-sensitive video favors stable LoS paths, and group-oriented perception maps use the aerial mesh.
B. How LAWN Differs from Aerial Communications?
LAWN extends aerial networking beyond connectivity by jointly organizing aerial and terrestrial nodes for communication, sensing, control, and computing. Its reconfigurable planes support cooperative perception, safety-critical services, and distributed decision-making.
- Traditional aerial networks primarily extend connectivity, while sensing and control are typically handled through separate networks or standalone systems.This separation limits their suitability for emerging low-altitude applications.
- LAWN jointly organizes aerial and terrestrial nodes as components of a cyber-physical system rather than using drones solely as communication tools.The architecture is reconfigurable, intelligent, and mission-aware.
- LAWN combines data, control, and sensing planes with an intelligence-and-computing plane.Its nodes dynamically reconfigure, and its computing hierarchy spans onboard, edge, and cloud resources.
- Cooperative sensing lets aerial and ground nodes share environmental data in real time for formation flying, obstacle avoidance, and urban air traffic management.
C. Standardization for LAWN-NTN Integration
3GPP has begun incorporating aerial platforms into NTN standards, but current specifications remain centered on data-plane functions. Native support for LAWN’s integrated control, sensing, and intelligence-and-computing planes is still incomplete.
- 3GPP Release 17 included satellites and airborne platforms within NTN scope, with aerial drones and HAPs identified as special use cases.
- Release 17 architecture work added cellular-network functions for aerial drone identification, authentication, and authorization.
- Fig. 3 identifies ISAC, Delay–Doppler waveforms, semantic communication, edge intelligence, large language models, and fully-actuated control as LAWN enablers.
- A relocatable low-altitude cell site can dynamically provide coverage to users in a cell.
- Current 3GPP specifications lack native support for LAWN’s integrated control, sensing, and intelligence-and-computing planes.Standard primitives for safety-critical control reliability, cooperative perception trust, and task or model continuity remain missing or loosely defined.
III. ENABLING TECHNOLOGIES AND POTENTIAL APPLICATIONS
LAWN realization requires enabling technologies across waveform design, integrated sensing, intelligent control, and network-level intelligence. These tools and protocols support the architecture’s broader operational goals.
- LAWN realization spans physical-layer waveform design, integrated sensing, intelligent control, and network-level intelligence.
- Advanced tools and protocols are essential to unlock the full potential of the LAWN architecture.
A. Enabling Technologies
LAWN enabling technologies address sensing and communication integration, dynamic-channel transmission, and network intelligence under demanding operating conditions. ISAC reuses wireless resources for environmental awareness, while Delay–Doppler waveforms improve resilience in highly dynamic channels.
- LAWN technologies span physical-layer waveform design, integrated sensing, intelligent control, and network-level intelligence under fast topology changes and severe resource constraints.
- ISAC uses the same signaling, hardware platform, and radio resources for wireless communication and environmental sensing.A transmitting drone can process reflected echoes to detect obstacles or track other aerial agents.
- Distributed drones can form a virtual large aperture through ISAC, improving angular resolution and imaging capability.Sensing-assisted CSI acquisition can also reduce reliance on frequent pilot transmission.
- Delay–Doppler waveforms, including OTFS and its variants, were proposed for reliable transmission in highly dynamic wireless channels.
- OTFS modulates symbols in the two-dimensional Delay–Doppler domain and converts time-varying channels into near time-invariant ones.The approach relies on channel parameters remaining nearly unchanged during a short symbol interval.
3) Semantic Communications:
Semantic communications transmit task-relevant meaning rather than complete raw messages, supporting bandwidth- and energy-efficient operation in LAWNs while introducing intelligence and trust requirements.
- Semantic communication: Semantic communication prioritizes task-relevant meaning over accurate bit-level recovery by discarding redundancy that does not affect downstream inference or action.Systems interpret context, intent, and importance rather than reconstructing raw messages with minimal bit error.
- Semantic communication: A task-oriented objective constrains task degradation, data rate, and safety-violation probability simultaneously.The formulation uses Ltask(·), R ≤ Rmax, and Pr(unsafe) ≤ ϵ.
- LAWN operation: Semantic-aware nodes can extract high-level descriptions from sensor streams and transmit concise representations instead of full 4K video.An example report is “three pedestrians crossing in two seconds,” reducing bandwidth and energy demands for battery-constrained drones.
- LAWN operation: Semantic awareness supports task-prioritized control and scheduling, but interpretation requires onboard or edge intelligence.Incorrect or over-compressed semantic messages may cause poor decisions in safety-critical missions.
- Intelligence and computing: AI and edge intelligence support distributed perception, decision-making, channel prediction, trajectory optimization, and cooperative behavior across LAWN nodes.Lightweight neural networks can run on drones or edge servers, while task partitioning assigns functions according to latency and compute demands.
- Intelligence and computing: Edge offloading is beneficial only when transmission, edge computation, result delivery, and protocol overhead remain below local-processing time.Three-dimensional mobility additionally causes intermittent connectivity and requires task and model continuity during movement and handover.
6) Multi-Connectivity and Redundant Networking:
LAWN applications span logistics, sensing, emergency response, urban air mobility, infrastructure inspection, and entertainment because the architecture combines altitude-segmented corridors with coupled functional planes.
- Multi-Connectivity and Redundant Networking: Multi-connectivity maintains parallel links to improve reliability under blockage, interference, and node failures.Examples include a sub-6 GHz control anchor with mmWave/FR2 data boosting and terrestrial connectivity with satellite backup.
- Applications: The architecture can simultaneously support mission-critical public-safety services and commercial or consumer ecosystems in low-altitude airspace.This breadth follows from combining altitude-segmented corridors with tightly coupled functional planes.
- Applications: Urban logistics uses vertiports, dynamic geofencing, and integrated traffic management to support compliant parcel, food, and medical-supply transport.These mechanisms are especially relevant to dense urban last-mile delivery and emergency medical deployment.
- Applications: Rural LAWNs support crop monitoring, spraying, irrigation planning, and environmental surveillance through multimodal sensing and distributed processing.Edge facilities handle aggregation and processing, while cloud infrastructure supports long-horizon analytics.
- Applications: Disaster-response LAWNs rapidly restore communications and situational awareness using airborne relays, real-time imagery, robust C2, and coordinated sensing.Drones can re-establish links and help localize survivors and identify hazards when terrestrial infrastructure is damaged.
- Applications: Urban Air Mobility requires low-altitude corridors with URLLC, cooperative sensing for collision avoidance, and integration with UTM/U-space systems.These capabilities are presented as requirements for scaling eVTOL demonstrations into services.
- Applications: Drone-based infrastructure inspection covers assets such as bridges, pipelines, power lines, and offshore wind farms, with reported savings and improved worker safety.Field reports specifically describe major California utilities saving millions of dollars after switching to drone inspections for power lines.
- Applications: LAWNs also support coordinated aerial light shows and potential immersive AR/VR experiences, extending applications into entertainment and media.Drone displays are described as programmable, environmentally friendly, and reusable alternatives to traditional fireworks.
C. Lessons Learned
The lessons learned emphasize that LAWN’s integrated architecture supports real-time processing, reconfiguration, crisis deployment, precision surveillance, and cross-domain operations.
- C. Lessons Learned: Integrated data and intelligence-computing planes enable real-time collection and processing of high-resolution data for resource optimization and decision-making.Precision agriculture is given as an example.
- C. Lessons Learned: The reconfigurable three-dimensional architecture supports adaptive urban-logistics operations under changing traffic or weather conditions.The lesson is framed as dynamic network reconfiguration.
- C. Lessons Learned: Ad-hoc networking enables robust emergency communication and operational continuity in disrupted environments.The lesson is tied to public-safety applications and rapid deployment in crises.
- C. Lessons Learned: Coordinated sensing and control planes support environmental and infrastructure monitoring through precise fault detection and data collection.The paper connects this capability with sustainability.
- C. Lessons Learned: Integration with systems such as IoT supports interdisciplinary operations in domains including smart cities and wildlife tracking.The architecture is presented as versatile across varied application domains.
- C. Lessons Learned: By combining connectivity, perception, control, mobility, adaptability, and edge intelligence, LAWN supports next-generation aerial systems.The conclusion presents unified architecture as the basis for these capabilities.
IV. CASE STUDY: LAWN-ASSISTED SWARM COORDINATION
The case study models post-disaster swarm coordination through four integrated planes, allowing drones to sense, communicate, compute, and adapt when external connectivity is impaired.
- A. System Setup: A post-disaster fleet surveys damage and locates survivors using a ground command node, aerial drones, and integrated LAWN planes.The ground node provides multi-band radios and edge servers while maintaining terrestrial, sub-GHz C2, and LEO backup links.
- A. System Setup: The intelligence-and-computing plane distributes heavy processing between onboard processors and ground edge servers.This division supports the case study’s multimodal sensing and coordination workflow.
- A. System Setup: The sensing plane fuses camera, LiDAR, IMU, and ISAC data to map terrain, detect survivors or obstacles, and support collision avoidance.Sensing augments control when clutter and LoS/NLoS transitions create rapidly changing conditions.
- A. System Setup: Even with partial or complete loss of external connectivity, the four planes adapt to sustain autonomous swarm operation.Drone-to-drone mesh links share local obstacle maps, while onboard lightweight AI supports formation reconfiguration and mission continuation.
1) Simulations:
The LAWN swarm study combines coordinated control, multi-link communication, sensing, and computing across ground and aerial nodes. Simulations and a real-world experiment evaluate formation control in dynamic low-altitude settings.
- Simulation setup: Five drones start randomly within a 1500 × 1500 × 40 m3 volume and target a 40 m-radius pentagon formation.
- Control plane: A 100 Hz leader-follower MPC uses synchronized sub-GHz C2 beacons to support sub-10 ms end-to-end control latency.
- LAWN coordination: The ground node distributes flight and formation updates through G2A links, collects mission payloads through A2G links, and supports A2A information sharing.
- Sensing plane: ISAC echoes derived from payload transmissions provide continuous environmental awareness for obstacle detection.
- Data plane: The DD waveform is evaluated against an identically configured OFDM baseline in a 3D multipath channel with Jakes Doppler distribution.
- Results: Across 1,000 Monte Carlo realizations, DD significantly outperforms OFDM under high mobility, while the swarm reaches its target formation within 15 s on average.
A. 3D Spectrum Coexistence
LAWN spectrum coexistence is complicated by three-dimensional mobility, changing antenna orientations, and shared use of bands occupied by terrestrial wireless infrastructure. The roadmap also highlights synchronization, latency, security, and physical-layer mechanisms for managing these conditions.
- A. 3D Spectrum Coexistence: Drones share sub-6 GHz, C-band, and mmWave bands with 5G and 6G infrastructure while moving in three dimensions.
- A. 3D Spectrum Coexistence: Vertical and horizontal maneuvers continually shift antenna beams, producing interference patterns that classic spectrum management frameworks cannot handle.
- A. 3D Spectrum Coexistence: Predictive beam scheduling should combine flight-mode data with channel state information to manage interference proactively.
- Synchronization and localization: GNSS-challenging environments expose LAWN operations to clock drift, multipath, and intermittent satellite visibility, creating biased positions and inconsistent timing.
- Latency and reliability: Crowded, rapidly maneuvering airspace demands short frames, minimal preambles, Doppler-resilient waveforms, and stringent end-to-end control latency.
- Cyber-physical security: GNSS spoofing, C2 injection, and jamming can threaten physical safety, motivating efficient cryptography, onboard anomaly detection, and physical-layer security methods.
E. Scalability Versus Battery Limits
Scaling LAWN operations is constrained by battery capacity, rapidly changing connectivity, regulatory requirements, and the robustness of embedded AI. The paper frames these issues as cross-layer challenges requiring coordinated technical and policy development.
- E. Scalability Versus Battery Limits: Each added wireless link or computational task increases energy consumption, making limited drone battery capacity a central scalability constraint.
- E. Scalability Versus Battery Limits: Energy-aware scheduling and routing, alongside higher-density batteries and faster recharging, are proposed responses to this constraint.
- Connectivity resilience: Mobility and blockage can abruptly break LoS links and partition aerial meshes, challenging routing methods that assume persistent paths and stable neighbors.
- Traffic management and regulation: Low-altitude deployment requires UTM frameworks that manage cooperative and uncooperative aircraft, maintain separation, and adapt to weather or communication failures.
- AI integration: Deep-learning integration creates explainability, robustness, and adversarial-input challenges for LAWN decision-making.
- Conclusion and outlook: The proposed LAWN roadmap positions the architecture as a reconfigurable infrastructure spanning communications, control, sensing, computing, robotics, and aviation regulation.