Analog RF Computing: A New Paradigm for Energy-Efficient Edge AI Over MU-MIMO Systems

TL;DR

This paper introduces an analog RF computing framework for energy-efficient edge AI over MU-MIMO systems, enabling low-power neural network inference at wireless receivers.

Contribution

It proposes a physical layer design for analog RF computing, including models, optimization algorithms, and accuracy control for energy-efficient edge inference.

Findings

Analog RF computing reduces client energy consumption by nearly 100x compared to digital methods.

Mixed-precision inference further lowers energy use than uniform-precision.

Simulation results demonstrate significant energy savings under 3GPP wireless standards.

Abstract

Modern edge devices increasingly rely on neural networks for intelligent applications. However, conventional digital computing-based edge inference requires substantial memory and energy consumption. In analog radio frequency (RF) computing, a base station (BS) encodes the weights of the neural networks and broadcasts the RF waveforms to the clients. Each client reuses its passive mixer to multiply the received weight-encoded waveform with a locally generated input-encoded waveform. This enables wireless receivers to perform the matrix-vector multiplications (MVMs) that account for most of the computation burden in edge inference with ultra-low energy consumption. Unlike conventional downlink transmissions which are optimized for communications, analog RF computing requires a computing-centric physical layer that controls both the analog MVM accuracy and the energy consumption for inference. Motivated by this, in this paper, we propose a physical layer design framework for analog RF computing in MU-MIMO wireless systems. We derive tractable models for computing accuracy and energy consumption for inference, formulate a joint BS beamforming and client-side scaling problem subject to computing accuracy, transmit power, and hardware constraints, and develop a low-complexity algorithm to solve the non-convex problem. The proposed design provides client- and layer-specific accuracy control for both uniform- and mixed-precision inference. Simulations under 3GPP specifications show that analog RF computing can significantly reduce client-side energy consumption by nearly two orders of magnitude compared to digital computing, while mixed-precision inference requires even lower energy consumption than uniform-precision inference. Overall, these results establish analog RF computing over wireless networks as a promising paradigm for energy-efficient edge inference.