射频卷积神经网络
Radio-Frequency Convolutional Neural Networks
查看机构详情
- Duke University（杜克大学）
- Massachusetts Institute of Technology（麻省理工学院）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
本文提出射频卷积神经网络（RF-CNN），利用无线设备现有混频器在频域执行卷积，实现边缘端高效CNN推理，能耗低至0.72飞焦耳/乘加，性能接近全精度。
中文摘要 AI 辅助
直接在智能手机、可穿戴设备和无人机等边缘设备上运行人工智能（AI）模型，具有低延迟、普遍可扩展性和数据隐私等优势，但这些设备很少具备现代神经网络所需的计算能力。为此，人们开发了边缘加速器，然而每种加速器都会给本已在尺寸、重量、功耗和成本（SWaP-C）方面受限的设备增加计算硬件。另一种思路是利用这些设备已有的部件：每台无线设备中的频率混频器在时域上对信号进行乘法运算，天然地在频域中执行卷积。在此，我们提出了射频卷积神经网络（RF-CNN），它将现有的通信硬件重新用于CNN推理。多通道卷积被映射到频率音调上，由无源混频器单次通过即可完成。我们通过实验证明，RF-CNN能够运行深度达2640万参数、九层结构的深度CNN，涵盖无线信号和图像的分类乃至可控图像生成等任务，其性能接近全精度水平。由于权重通过无线方式到达，且模拟硬件与通信共享，边缘设备仅在数据准备和读出上消耗能量——每次乘加运算低至0.72飞焦耳，比在额外数字处理器上运行低两个数量级。这些结果表明，已部署的无线基础设施可以为它所连接的数十亿设备带来高效、最先进的AI推理能力。
英文摘要
Running artificial intelligence (AI) models directly on edge devices such as smartphones, wearables, and drones offers low latency, pervasive scalability, and data privacy, but these devices rarely carry the computing capability that modern neural networks demand. Edge accelerators have been developed in response, yet each adds computing hardware to devices already constrained in size, weight, power, and cost (SWaP-C). An alternative lies in what these devices already carry: the frequency mixer in every wireless radio multiplies signals in time, natively performing convolution in the frequency domain. Here we introduce radio-frequency convolutional neural networks (RF-CNNs), which repurpose existing communication hardware for CNN inference. Multi-channel convolutions are mapped onto frequency tones for a passive mixer to execute in a single pass. We experimentally demonstrate that RF-CNN runs deep CNNs up to 26.4 million parameters and nine layers from classification of wireless signals and images to controllable image generation, close to full-precision performance. Because the weights arrive over the air and the analog hardware is shared with communication, the edge device spends energy only on data preparation and readout-down to 0.72 femtojoules per multiply-accumulate, two orders of magnitude less than it would cost on an added digital processor. These results suggest that deployed wireless infrastructure can bring efficient, state-of-the-art AI inference to the billions of devices it already connects.
