非相干空中联邦学习：协议、收敛性与设备调度
Non-Coherent Over-the-Air Federated Learning: Protocol, Convergence, and Device Scheduling
浏览论文内容
中文总结 AI 辅助
本文提出非相干空中联邦学习协议，免除瞬时CSI需求，实现与FedAvg同阶收敛速率，并设计联合设备选择与功率控制策略以加速收敛。
中文摘要 AI 辅助
为缓解联邦边缘学习（FEEL）中无线接入网（RAN）的可扩展性瓶颈，空中联邦学习（AirFL）利用多址信道（MAC）上的波形叠加进行模拟模型聚合。然而，相干AirFL通常依赖于严格的物理层条件，如精确的信道状态信息（CSI）、紧密的时间/频率同步以及频繁的收发器校准以实现信号对齐。然而，这些要求即便并非无法满足，也会产生大量的通信和计算开销。在本文中，我们提出了一种在宽带单天线MAC上的非相干AirFL（NCAirFL）协议，利用二进制抖动、无偏非相干检测和长期误差反馈来免除对瞬时CSI的需求。对于具有一般光滑非凸目标和恒定学习率的NCAirFL，我们建立了一个收敛界，其收敛速率与通信理想的FedAvg同阶，为$\mathcal{O}(1/\sqrt{T})$，其中$T$是总通信轮数。为了在数据和无线资源异构性下进一步提高通信效率，我们还推导了在设备调度条件下全局损失期望单轮目标下降的下界，并在此基础上获得了一个用于联合最优设备选择和功率控制的替代目标函数。在MNIST和CIFAR-10上的实验结果表明，在实际设置中，NCAirFL实现了接近FedAvg的学习性能，而所提出的设备调度策略显著加速了收敛。
英文摘要
To mitigate the scalability bottleneck in the radio access network (RAN) in federated edge learning (FEEL), over-the-air federated learning (AirFL) exploits waveform superposition over multiple-access channels (MACs) for analog model aggregation. However, coherent AirFL typically relies on stringent PHY-layer conditions such as accurate channel state information (CSI), tight time/frequency synchronization, and frequent transceiver calibration for signal alignment. However, these requirements, if not impossible to be met, incur substantial communication and computation overhead. In this paper, we propose a non-coherent AirFL (NCAirFL) protocol over a broadband single-antenna MAC, leveraging binary dithering, unbiased non-coherent detection, and long-term error feedback to waive the need for instantaneous CSI. For NCAirFL with general smooth non-convex objectives and a constant learning rate, we establish a convergence bound achieving the convergence rate in the same order of $\mathcal{O}(1/\sqrt{T})$ as communication-ideal FedAvg, where $T$ is the total number of communication rounds. To further improve communication efficiency under data and wireless resource heterogeneity, we also derive a lower bound on the expected single-round objective decrease in the global loss conditioned on device scheduling, building upon which a surrogate objective function is obtained for jointly optimal device selection and power control. Experimental results on MNIST and CIFAR-10 corroborate that NCAirFL achieves learning performance close to FedAvg in practical settings, with the proposed device scheduling policy substantially accelerating convergence.
发表机构
- The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））
- The Hong Kong University of Science and Technology（香港科技大学）
- Arizona State University（亚利桑那州立大学）
- Northeastern University London（伦敦东北大学）
- Shanghai HKU Education Center（上海港大教育中心）
机构由 AI 辅助整理，请以论文原文为准。
