Predictive Communications for Low-Altitude Networks

TL;DR

Predictive communication for low-altitude networks addresses the challenge of extreme channel dynamics and cross-tier interference by leveraging foresight into both mission trajectories and radio environments. The approach fuses deterministic $4D$ trajectories (3D space + time) with $6D$ radio environment models to generate a predictive, time-evolving view of link qualities, enabling proactive optimization. A three-layer framework—strategic routing, tactical timing, and operational power—maps long-horizon foresight to network-wide decisions while accommodating accuracy-range trade-offs. Through a case study on interference mitigation, the authors demonstrate substantial cross-tier interference reductions and outline security and ISAC extensions for resilience. The work provides a scalable blueprint for robust, security-aware low-altitude networks that can support mission-driven services.

Abstract

The emergence of dense, mission-driven aerial networks supporting the low-altitude economy presents unique communication challenges, including extreme channel dynamics and severe cross-tier interference. Traditional reactive communication paradigms are ill-suited to these environments, as they fail to leverage the network's inherent predictability. This paper introduces predictive communication, a novel paradigm transforming network management from reactive adaptation to proactive optimization. The approach is enabled by fusing predictable mission trajectories with stable, large-scale radio environment models (e.g., radio maps). Specifically, we present a hierarchical framework that decomposes the predictive cross-layer resource allocation problem into three layers: strategic (routing), tactical (timing), and operational (power). This structure aligns decision-making timescales with the accuracy levels and ranges of available predictive information. We demonstrate that this foresight-driven framework achieves an order-of-magnitude reduction in cross-tier interference, laying the groundwork for robust and scalable low-altitude communication systems.

Figures (5)

• Figure 1: The operational concept of the low-altitude network and its implications for predictive communication. (a) Diverse low-altitude operations, such as infrastructure inspection and cargo delivery, are performed by aircraft following predetermined trajectories that are tailored to their mission requirements. (b) For reasons of air traffic safety and operational management, these diverse, mission-driven flight plans are filed with a central control authority, which possesses a priori, network-wide knowledge of all aircraft movements. This centralized oversight of a predictable, time-varying network topology creates the foundational opportunity for predictive communication.
• Figure 2: Illustration of an aerial-assisted radio map generation and update process. Multiple collaborate to sample the radio environment along their trajectories, creating a sparse point cloud of channel data. This aggregated data is then used to generate or update a dense, multi-layered environmental model, enabling a continuously refined understanding of the radio environments.
• Figure 3: Generation process of the predictive large-scale , illustrating the information processing flow. The architecture leverages two foundational inputs: (a) a set of 4D mission trajectories (3D space + 1D time) for all aircraft, and (b) a 6D radio environment model that maps the 6D spatial coordinates of a transmitter-receiver pair to a 1D prediction of large-scale channel quality. In the (c) synthesis step, these inputs are fused. For any two aircraft, their independent 4D trajectories are first temporally aligned to form a 7D state descriptor (3D transmitter position, 3D receiver position, and 1D time). At each time step, the 6D spatial component is mapped by the environmental model to a 1D channel quality value. This process transforms the initial trajectory data into two representative predictive views: a link-level channel prediction, which shows the 2D (channel quality vs. time) forecast for a series of single links, and a network-level channel prediction, which displays the evolution of the entire low-altitude network topology.
• Figure 4: The layered predictive communication framework, illustrating the hierarchical task decomposition for a proactive communication task. (a) The overall objective is to deliver data from a source to a destination via a multi-hop aerial route while minimizing interference. The main decisions, route selection, timing, and power allocation, require information at different scales. This complex problem is decomposed across three layers: (b) The Large-Scale Strategic Layer uses global, long-term information to transform mission requirements into a reserved end-to-end path. (c) This path is passed as a directive to the Middle-Scale Tactical Layer, where a local cluster uses more precise, local-area predictions to coordinate the timing of each hop, potentially adjusting the route to bypass a blockage. (d) Finally, the Small-Scale Operational Layer receives a specific transmission schedule. It uses link-level statistics to design a predictive power allocation policy that optimizes physical-layer efficiency while managing interference.
• Figure 5: Interference-aware predictive communications. (a) Case study. (a.1) Considered scenario with predictive aircraft mission trajectories, where black dots denote aircraft positions and red arrows indicate flight directions, over a ground layout comprising a source (square), a destination (triangle), and interference-sensitive nodes (circles). (a.2) Planned route and associated transmit power (color bar) for a delay-tolerant task (delay tolerance 20 s). (a.3) Planned route and transmit power for a delay-sensitive task (delay tolerance 2 s). (b) Aggregate interference power of the proposed method versus two baselines under increasing network load (number of commodities).