# Multi-layer MIMO Relay as Deep Physical Neural Networks: Power Amplifiers as Activation Functions

Source: [Multi-layer MIMO Relay as Deep Physical Neural Networks: Power Amplifiers as Activation Functions](<https://www.sota2.com/research/paper/6a604e6fef5cfba72b49788a>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-07-23T05:00:05.031Z

Authors: Meng Hua, Itsik Bergel, Deniz G\\"und\\"uz

Published: 2026-07-20T09:45:07.000Z

## Abstract

Wireless physical neural networks (WPNNs) embed neural computation directly into analog hardware, offering lower energy consumption and latency than conventional digital implementations. In this paper, we propose a deep WPNN in which nonlinear activations are realized by a multi-hop multiple-input multiple-output (MIMO) relay network, in which each relay implements a trainable complex linear gain and bias, followed by the power amplifier's intrinsic nonlinearity acting as an activation function. The cascade of multiple relays therefore realizes an over-the-air fully connected network whose parameters can be trained end-to-end. We develop two transceiver designs for different channel state information (CSI) availability scenarios: a least squares (LS)-based scheme requiring only receiver-side CSI, and a singular-value-decomposition (SVD)-based scheme requiring both transmitter-side and receiver-side CSI. Simulation results show that the proposed architecture enables accurate over-the-air inference for image classification. In particular, the results highlight the advantage of exploiting hardware nonlinearity for enhanced inference capability.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2607.18354>)

[PDF](<https://arxiv.org/pdf/2607.18354.pdf>)
