Goal-Oriented Learning at the Edge: Graph Neural Networks Over-the-Air for Blockage Prediction

TL;DR

This paper introduces a goal-oriented framework using over-the-air graph neural networks for efficient, low-latency blockage prediction in 6G wireless networks, combining analog aggregation with learning to improve scalability and adaptability.

Contribution

It proposes a novel analog over-the-air GNN framework that enhances scalability and reduces communication overhead for distributed edge inference tasks.

Findings

Achieves low-latency message passing with analog aggregation.

Demonstrates strong generalization and domain adaptation in blockage prediction.

Outperforms digital baselines with less communication overhead.

Abstract

Sixth-generation (6G) wireless networks evolve from connecting devices to connecting intelligence. The focus turns to Goal-Oriented Communications, where the effectiveness of communication is assessed through task-level objectives over traditional throughput-centric metrics. As communication intertwines with learning at the edge, distributed inference over wireless networks faces a critical trade-off between task accuracy and efficient radio resource use. Traditional communication schemes (e.g., OFDMA) are not designed for this trade-off, often facing challenges related to scalability and latency. Therefore, we propose a novel goal-oriented framework that integrates over-the-air computation with spatio-temporal graph learning. Leveraging the wireless channel as an analog aggregation layer, the proposed framework enables low-latency message passing while efficiently aggregating semantically relevant features from distributed nodes. Theoretical analysis confirms that our analog architecture converges to the expressive power of digital message passing, while offering decisive scalability advantages. We assess the framework in proactive line-of-sight blockage prediction for millimeter-wave networks. Through high-fidelity ray-tracing simulations, the framework exhibits strong inductive generalization to unseen networks and adapts to domain shifts via lightweight transfer learning, matching or even outperforming digital baselines with significantly reduced communication overhead.