AirNN: Over-the-Air Computation for Neural Networks via Reconfigurable Intelligent Surfaces

• Published in: IEEE/ACM transactions on networking Vol. 31; no. 6; pp. 1 - 0
• Main Authors: Sanchez, Sara Garcia, Reus-Muns, Guillem, Bocanegra, Carlos, Li, Yanyu, Muncuk, Ufuk, Naderi, Yousof, Wang, Yanzhi, Ioannidis, Stratis, Chowdhury, Kaushik Roy
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.12.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: analog convolution; Artificial neural networks; Computation offloading; Computer architecture; convolutional neural network; FIR filters; Neural networks; Over-the-air computation; programmable wireless environment; reconfigurable intelligent surface; Reconfigurable intelligent surfaces; Wave reflection

Over-the-air analog computation allows offloading computation to the wireless environment through carefully constructed transmitted signals. In this paper, we design and implement the first-of-its-kind convolution that uses over-the-air computation and demonstrate it for inference tasks in a convolu...