Low-Altitude Economy Networking

• Low-altitude economy networking is a paradigm that integrates non-terrestrial, terrestrial, and air–ground systems to enable coordinated operations below 1,000 meters.
• It leverages multifunctional aircraft for sensing, communication, and applications in urban mobility, logistics, and environmental monitoring.
• Critical challenges include achieving seamless 3D wireless coverage, robust interference management, and optimized multi-modal sensing with collaborative autonomy.

Low-altitude economy networking describes the large-scale integration and optimization of non-terrestrial networks (NTN), terrestrial networks, and advanced air–ground systems to enable commercial and social activities in airspaces below 1,000 meters altitude. This paradigm shift leverages aircraft as multi-functional platforms for sensing, communication, and transportation, driving new economic ecosystems for applications ranging from logistics and urban mobility to environmental monitoring and smart city services. The distinct technological, architectural, and operational challenges necessitate advances in 3D wireless coverage, interference management, multi-modal sensing, and collaborative autonomy.

1. Technological Foundations and Network Architecture

The core enablers of the low-altitude economy (LAE) are (i) comprehensive three-dimensional (3D) network coverage, and (ii) robust aircraft detection architectures (Jiang et al., 2023). 3D network coverage mandates seamless, persistent connectivity for aircraft operating up to 1,000 meters altitude with the following key components:

• Cellular Access: Aircraft detect synchronization signal blocks (SSBs) from multiple base stations (BSs), exploiting line-of-sight (LoS) links for cooperative multi-cell access. Channel modeling for air-to-ground (A2G) and air-to-air (A2A) links accounts for Doppler shift, fast mobility, and airframe shadowing, distinct from ground-to-ground (G2G) models.
• Spectrum Sharing & 3D Beamforming: Aircraft employ spectrum sensing (energy, matched filter, or periodicity detection) to opportunistically access less interfered resource blocks, with full-dimensional antenna arrays enabling azimuth-elevation beamforming.
• Aircraft Detection Techniques: Safety and resource utilization are ensured via cooperative active sensing (onboard radars, cameras, GNSS, IMUs) and non-cooperative passive sensing (e.g., micro-Doppler radar signatures for unauthorized/illegal aircraft identification).

Architecture is characterized by tightly integrated layers (Wang et al., 30 Apr 2025):

• Airborne physical layer (sensors, flight control, communications)
• Digital airspace collaboration (communication, computation, positioning, surveillance, digital twin-based management)
• Service/collaboration assurance (regulation, scheduling, interoperability)

Satellite-assisted solutions further extend coverage, introducing distributed satellite–aerial MIMO, two-timescale optimization for cross-layer resource control, and satellite-based navigation/federated learning support (He et al., 7 May 2025).

2. Sensing, Communication, and Integrated Functionality

Aircraft in LAE fulfill dual roles:

• Sensing:
  • Non-terrestrial target sensing employs aircraft swarming for cooperative collision avoidance, tracking other aircraft and small targets (birds, UAVs).
  • Terrestrial sensing leverages distributed aircraft for wide-area or high-resolution environmental acquisitions (e.g., pollutant mapping, meteorological observation), with joint trajectory–beamforming optimization managing LoS intermittency.
• Communication:
  • Ubiquitous coverage: Aircraft act as aerial base stations (ABS) for areas lacking ground connectivity or in post-disaster conditions.
  • Relaying and Traffic Offloading: Dynamic relaying bridges BS-to-UE links and facilitates traffic offloading in terrestrial hotspots, with dynamic flight path adaptation enhancing signal quality.
• Integrated Sensing and Communication (ISAC): The ISAC paradigm allows joint optimization of flight trajectories and beamforming to satisfy simultaneous sensing (target illumination) and communication (throughput/SINR) requirements (Cheng et al., 13 May 2024, Cheng et al., 19 Jun 2024, Li et al., 25 Jun 2025).

Cooperative bistatic ISAC frameworks exploit distributed 5G NR MIMO-OFDM networks, utilizing CANDECOMP/PARAFAC (CP) tensor decomposition for efficient multi-target range, velocity, and angle estimation, and minimum spanning tree (MST)-based fusion for consistent 3D position/velocity reconstruction (Zhang et al., 22 Jun 2025).

Table: Representative LAE Functionalities

Functionality	Mechanism	Key Technical Requirement
Sensing (NT/TT)	Swarm, multi-modal sensing	Trajectory/beamforming optimization
Ubiquitous Coverage	ABS deployment	3D coverage, LoS link exploitation
Relaying/Offloading	Aerial relay, path control	Low-latency handover, adaptive flight

3. Collaborative Operation, Computation, and Energy Optimization

Collaboration in LAE is multi-faceted:

• Aircraft Collaboration: Efficient, low-latency sharing of sensed data is addressed through over-the-air computation (AirComp), allowing simultaneous, analog aggregation of sensor values (Jiang et al., 2023).
• Edge Computing and Offloading: UAVs are integrated with mobile edge computing (MEC) to enable low-latency, high-throughput processing. Multi-UAV resource allocation involves the formulation of complex, NP-hard multi-objective problems spanning offloading, computation, trajectory, and energy constraints (Wen et al., 19 Jun 2025, Xue et al., 27 Jun 2025). HybridRAG-based LLM agents facilitate structured retrieval of expert knowledge for automated optimization problem generation, and double-regularization diffusion-enhanced soft actor-critic (R²DSAC) algorithms use both diffusion and action entropy for robust, low-carbon policy learning.
• Energy Efficiency: Energy is a first-class constraint due to limited onboard power. Solutions involve trajectory and transmit power co-optimization, strategic placement of charging infrastructure, and exploration of wireless power transfer (WPT) (Jiang et al., 2023, Wang et al., 30 Apr 2025).

Generative AI (e.g., diffusion models), reinforcement learning, and Lyapunov optimization underpin learning-based real-time control for stability and multiple objectives (Liu et al., 27 Jan 2025, Zhao et al., 25 Feb 2025, Zhao et al., 21 May 2025).

4. Security, Privacy, and Robustness

LAE networking faces acute security threats arising from open airspace and LoS dominance:

• Physical Layer Security: Techniques for anti-eavesdropping (mobility/trajectory optimization, convex secrecy rate maximization, deep RL for channel uncertainty) and lightweight mutual authentication (physical unclonable functions, channel fingerprinting) are central (Cai et al., 12 Apr 2025).
• Availability: Anti-jamming and anti-spoofing leverage both optimization and deep RL/MARL for robust channel selection, transmit power adaptation, and anomaly detection (HDBN, GDBN, SVMs).
• Integrity: Data integrity is enforced through anomaly detection (statistical distances) and injection defense (SIC, digital beamforming, subspace projection).
• Privacy: Cluster authentication schemes such as LP2-CASKU employ message aggregation, cross-cluster authentication, and dynamic session key update for privacy-preserving, scalable UAV cluster association with strong forward and backward secrecy (Gong et al., 7 Sep 2025).
• Robustness: GenAI-enabled wireless systems employ diffusion-based beamforming and transformer/MoE actor networks to achieve up to 44% worst-case secrecy rate improvements under uncertainty (Zhao et al., 25 Feb 2025).

5. Spectrum Management and Resource Allocation

Spectrum management is challenged by rapid Doppler shifts, A2G/A2A mobility, and spectrum crowding:

• Federated Learning-Based Sensing: Decentralized federated learning (FedSNR) improves spectrum sensing via SNR-weighted aggregation, in which UAVs with higher SNR contribute more strongly to global models (Tekbıyık et al., 2 Jun 2025). Sensing accuracy approaches 97.6% as the agent pool size grows.
• Dynamic Cartography and Resource Allocation: Generative AI frameworks using masked autoencoders and multi-agent diffusion policies support high-precision temporal spectrum cartography and decentralized, adaptive UAV sensor deployment, yielding MSE reductions up to 88.68% compared to baseline methods (Zhao et al., 21 May 2025).
• Graph Attention Diffusion: For task offloading/resource allocation, graph attention networks with diffusion enable efficient, joint optimization of discrete (offloading) and continuous (resource allocation) variables, yielding superior average cost ratios and solution accuracy in heterogenous, dynamic MEC networks (Xue et al., 27 Jun 2025).

6. Predictive and Hierarchical Networking Paradigms

Foresight-driven communication paradigms exploit mission trajectory determinism and high-resolution radio environment models to transition from reactive adaptation to predictive optimization (Chen et al., 1 Sep 2025):

• Hierarchical resource allocation splits decision-making into (i) strategic routing (global, long-term, route reservation based on mission plans and radio map), (ii) tactical timing (medium-term, scheduling to avoid interference hot-spots), and (iii) operational power control (short-term, dynamic power allocation using local predictions).
• Predictive frameworks achieve order-of-magnitude reductions in cross-tier interference and enable scalable, robust aerial network operation compared to myopic, reactive baselines.

7. Economic, Regulatory, and Future Directions

Tokenizing on-board computing power as real-world assets (RWAs) mapped to blockchain smart contracts enables trusted, shareable, and market-driven computility (computing utility) networks, or "LACNets" (Luo et al., 25 Aug 2025). RWAs are implemented as NFTs (ERC-721/1155), allowing UAVs and eVTOLs to auction and monetize idle computing resources in decentralized marketplaces administered via Stackelberg game models and smart contracts. Case studies in urban logistics demonstrate improvements in processing latency, trust assurance, and resource efficiency. Real-world deployment raises legal, regulatory, and security challenges—including cross-chain integration, industrial compliance, and resilience to attacks on integrated avionics and smart contract layers.

Looking forward, major directions include:

• AI–empowered orchestration and collaborative resource allocation (including on-chain DAOs)
• Generative AI–driven optimization, generative governance, and digital twin simulation for self-evolving cognitive agents
• Integration of quantum-driven coordination
• Deep collaboration between LAE, LEO satellite constellations, and terrestrial networks for unified 3D coverage and resource virtualization (Wang et al., 30 Apr 2025).

The multi-layered, fully integrated LAE ecosystem—leveraging NTN–TN convergence, advanced ISAC, intelligent automation, and blockchain-enabled resource sharing—signifies a large-scale shift towards smart, adaptive, and sustainable urban economic operations in low-altitude airspace.