PaperSwipe

Nonlinear EM-based Signal Processing

Published 2 days agoVersion 1arXiv:2512.04595

Authors

Mattia Fabiani, Giulia Torcolacci, Davide Dardari

Categories

eess.SP

Abstract

The use of high-frequency bands, combined with antenna arrays containing an extremely large number of elements (XL-MIMO), is pushing current technology to its limits in terms of hardware complexity, latency, and power consumption. A promising approach to achieving scalable and sustainable solutions is to shift part of the signal processing directly into the electromagnetic (EM) domain. In this paper, we investigate novel architectures that harness the interaction of reconfigurable passive linear and nonlinear (NL) scattering elements positioned in the reactive near field of signal sources. The objective is to enable multifunctional linear and NL EM signal processing to occur directly "over-the-air." Numerical results highlight the potential to significantly reduce both system complexity and the number of RF chains, while still achieving key performance metrics in applications such as direction-of-arrival and position estimation, without the need for additional analog or digital processing.

Nonlinear EM-based Signal Processing

2 days ago
v1
3 authors

Categories

eess.SP

Abstract

The use of high-frequency bands, combined with antenna arrays containing an extremely large number of elements (XL-MIMO), is pushing current technology to its limits in terms of hardware complexity, latency, and power consumption. A promising approach to achieving scalable and sustainable solutions is to shift part of the signal processing directly into the electromagnetic (EM) domain. In this paper, we investigate novel architectures that harness the interaction of reconfigurable passive linear and nonlinear (NL) scattering elements positioned in the reactive near field of signal sources. The objective is to enable multifunctional linear and NL EM signal processing to occur directly "over-the-air." Numerical results highlight the potential to significantly reduce both system complexity and the number of RF chains, while still achieving key performance metrics in applications such as direction-of-arrival and position estimation, without the need for additional analog or digital processing.

Authors

Mattia Fabiani, Giulia Torcolacci, Davide Dardari

arXiv ID: 2512.04595
Published Dec 4, 2025

Click to preview the PDF directly in your browser