标题： 模拟数字调度用于联邦学习：一种通信高效的方案 显示英文标题

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

摘要： 过空气（OTA）计算最近作为一种通信高效的联邦学习（FL）范式出现，用于在无线网络上训练机器学习模型。 然而，其性能受到信噪比最差的设备的限制，导致快速但噪声较大的更新。 另一方面，通过数字信道为每个设备分配正交资源块（RB）可以缓解噪声问题，但代价是增加了通信延迟。 在本文中，我们解决了这一差异并提出了 ADFL，一种新颖的模拟-数字FL方案：在每一回合中，参数服务器（PS）安排每个设备通过模拟OTA方案上传其梯度，或使用"数字"方案通过正交RB传输其量化梯度。 专注于单个FL回合，我们将最优调度问题建模为在PS处估计的全局梯度的均方误差（MSE）最小化问题，受延迟约束限制，从而得到最优设备调度配置和数字设备的量化比特数。 我们的仿真结果表明，ADFL通过在OTA方案中调度大多数设备，同时偶尔对少数设备采用数字方案，无论是在独立同分布（i.i.d.）还是非独立同分布（non-i.i.d.）设置中，都能持续优于仅OTA和仅数字方案。 显示英文摘要

摘要： 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.