Analog-digital Scheduling for Federated Learning: A Communication-Efficient Approach

TL;DR

This paper introduces ADFL, a hybrid analog-digital federated learning scheme that optimally schedules devices for OTA or digital updates, significantly improving communication efficiency and model accuracy over wireless networks.

Contribution

The paper proposes a novel hybrid scheduling scheme for federated learning that combines analog OTA and digital transmissions, optimizing device scheduling and quantization for improved performance.

Findings

ADFL outperforms OTA-only and digital-only schemes in simulations.

Scheduling most devices in OTA mode with some digital transmissions yields better results.

The scheme is effective in both i.i.d. and non-i.i.d. data settings.

Abstract

Over-the-air (OTA) computation has recently emerged as a communication-efficient Federated Learning (FL) paradigm to train machine learning models over wireless networks. However, its performance is limited by the device with the worst SNR, resulting in fast yet noisy updates. On the other hand, allocating orthogonal resource blocks (RB) to individual devices via digital channels mitigates the noise problem, at the cost of increased communication latency. In this paper, we address this discrepancy and present ADFL, a novel Analog-Digital FL scheme: in each round, the parameter server (PS) schedules each device to either upload its gradient via the analog OTA scheme or transmit its quantized gradient over an orthogonal RB using the ``digital" scheme. Focusing on a single FL round, we cast the optimal scheduling problem as the minimization of the mean squared error (MSE) on the estimated global gradient at the PS, subject to a delay constraint, yielding the optimal device scheduling configuration and quantization bits for the digital devices. Our simulation results show that ADFL, by scheduling most of the devices in the OTA scheme while also occasionally employing the digital scheme for a few devices, consistently outperforms OTA-only and digital-only schemes, in both i.i.d. and non-i.i.d. settings.