Harnessing Wavefront Curvature and Spatial Correlation in Noncoherent MIMO Communications

TL;DR

This paper introduces a channel model that accounts for wavefront curvature in noncoherent MIMO systems, showing that near-field effects can be exploited to improve performance and enable user multiplexing with low complexity.

Contribution

It presents a new channel correlation model valid in both near and far fields, demonstrating the benefits of wavefront curvature in noncoherent MIMO communications.

Findings

Wavefront curvature enhances noncoherent system performance.

Large arrays enable user multiplexing without coherent processing.

Classical far-field assumptions underestimate system capabilities.

Abstract

Noncoherent communication systems have regained interest due to the growing demand for high-mobility and low-latency applications. Most existing studies using large antenna arrays rely on the far-field approximation, which assumes locally plane wavefronts. This assumption becomes inaccurate at higher frequencies and shorter ranges, where wavefront curvature plays a significant role and antenna arrays may operate in the radiative near field. In this letter, we adopt a model for the channel spatial correlation matrix that remains valid in both near and far field scenarios. Using this model, we demonstrate that energy-based noncoherent systems can leverage the benefits of wavefront spherical curvature, even beyond the Fraunhofer distance, revealing that the classical far-field approximation may significantly underestimate system performance. Moreover, we show that large antenna arrays enable the multiplexing of various users even with a noncoherent processing, as well as permitting near-optimal detection with low computational complexity.