AirPASS: Over-the-Air Federated Learning via Pinching Antenna Systems

2026-07-07

AI Summary

This work addresses over-the-air federated learning (AirFL) in multi-waveguide pinched antenna systems (PASS) by proposing the AirPASS framework, which maximizes the number of participating devices under a constraint on aggregation distortion. For the first time, non-colocated reconfigurable pinched antennas are integrated into AirFL, and an alternating optimization algorithm is developed that jointly optimizes device selection, receive beamforming, and reconfigurable antenna placement. The algorithm effectively decouples the highly non-convex and coupled problem by combining homotopy-Riemannian boundary condensation with homotopy-assisted geometric optimization. Experimental results demonstrate that AirPASS significantly outperforms conventional colocated MIMO, SDR-DC, and matching pursuit-based scheduling schemes, achieving performance close to ideal FedAvg while striking a favorable balance between computational complexity and system performance.

Abstract

This paper investigates over-the-air federated learning (AirFL) in wireless systems where the access point is equipped with a multi-waveguide pinching antenna system (PASS). We adopt the widely studied learning-oriented AirFL formulation, which seeks to maximize the number of selected devices while keeping the aggregation distortion below a prescribed threshold. The resulting joint optimization of device selection, receive beamforming, and pinching-antenna placement is highly nonconvex due to the intricate coupling among these system variables. To address this challenge, we develop AirPASS, an alternating optimization framework with two main components: a homotopy-Riemannian margin-consolidation method for device selection and receive beamforming under fixed PASS configuration, and a homotopy-assisted geometry optimization method for updating the pinching-antenna positions under fixed selected devices and beamformer. Experiments show that AirPASS consistently outperforms conventional co-located MIMO baselines, remains close to ideal FedAvg, and achieves an attractive performance-complexity tradeoff relative to SDR-DC and matching-pursuit scheduling alternatives.

Problem

• over-the-air federated learning
• pinching antenna system
• device selection
• aggregation distortion
• joint optimization

Innovation

• over-the-air federated learning
• pinching antenna system
• alternating optimization
• homotopy-Riemannian optimization
• geometry optimization