Journal
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS
Volume 20, Issue 3, Pages 2120-2135Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TWC.2020.3039309
Keywords
Wireless communication; Wireless sensor networks; Quadrature amplitude modulation; Channel estimation; Broadband communication; Servers; Convergence; Over-the-air computation; federated learning; multiple access channels; quantization; digital modulation
Funding
- National Natural Science Foundation of China [62001310]
- National Key Research and Development Program of China [2018YFB1800800]
- Guangdong Province Key Area Research and Development Program [2018B030338001]
- Shenzhen Peacock Plan [KQTD2015033114415450]
- European Research Council (ERC) through the Starting Grant BEACON [677854]
- U.K. EPSRC [EP/T023600/1]
- Guangdong Basic and Applied Basic Research Foundation [2019B1515130003]
- Hong Kong Research Grants Council [17208319, 17209917]
- Innovation and Technology Fund [GHP/016/18GD]
- EPSRC [EP/T023600/1] Funding Source: UKRI
Ask authors/readers for more resources
Federated edge learning (FEEL) is a framework for training models at edge servers using data from edge devices, with recent efforts focusing on a digital aggregation scheme to overcome communication bottlenecks. Analysis indicates that wireless channel hostilities can slow down the convergence rate of the learning process, but these negative effects diminish with an increase in the number of participating devices.
Federated edge learning (FEEL) is a popular framework for model training at an edge server using data distributed at edge devices (e.g., smart-phones and sensors) without compromising their privacy. In the FEEL framework, edge devices periodically transmit high-dimensional stochastic gradients to the edge server, where these gradients are aggregated and used to update a global model. When the edge devices share the same communication medium, the multiple access channel (MAC) from the devices to the edge server induces a communication bottleneck. To overcome this bottleneck, an efficient broadband analog transmission scheme has been recently proposed, featuring the aggregation of analog modulated gradients (or local models) via the waveform-superposition property of the wireless medium. However, the assumed linear analog modulation makes it difficult to deploy this technique in modern wireless systems that exclusively use digital modulation. To address this issue, we propose in this work a novel digital version of broadband over-the-air aggregation, called one-bit broadband digital aggregation (OBDA). The new scheme features one-bit gradient quantization followed by digital quadrature amplitude modulation (QAM) at edge devices and over-the-air majority-voting based decoding at edge server. We provide a comprehensive analysis of the effects of wireless channel hostilities (channel noise, fading, and channel estimation errors) on the convergence rate of the proposed FEEL scheme. The analysis shows that the hostilities slow down the convergence of the learning process by introducing a scaling factor and a bias term into the gradient norm. However, we show that all the negative effects vanish as the number of participating devices grows, but at a different rate for each type of channel hostility.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available