Over-the-Air Federated Learning in Cell-Free MIMO with Long-term Power Constraint

TL;DR

This paper investigates Over-the-Air Federated Learning in cell-free MIMO systems, deriving error bounds and proposing an optimization algorithm to balance training performance with long-term power constraints, validated through experiments.

Contribution

It introduces the MOP-LOFPC algorithm using Lyapunov optimization for joint power control and beamforming under long-term power constraints in cell-free MIMO federated learning.

Findings

MOP-LOFPC outperforms existing methods in training loss and power constraint adherence.

The algorithm effectively balances model accuracy and power consumption.

Experimental results validate the theoretical advantages of the proposed approach.

Abstract

Wireless networks supporting artificial intelligence have gained significant attention, with Over-the-Air Federated Learning emerging as a key application due to its unique transmission and distributed computing characteristics. This paper derives error bounds for Over-the-Air Federated Learning in a Cell-free MIMO system and formulates an optimization problem to minimize optimality gap via joint optimization of power control and beamforming. We introduce the MOP-LOFPC algorithm, which employs Lyapunov optimization to decouple long-term constraints across rounds while requiring only causal channel state information. Experimental results demonstrate that MOP-LOFPC achieves a better and more flexible trade-off between the model's training loss and adherence to long-term power constraints compared to existing baselines.