RIS-Aided Near-Field Channel Estimation under Mutual Coupling and Spatial Correlation

Ahmad Dkhan, Simon Tarboush, Hadi Sarieddeen, Tareq Y. Al-Naffouri · 2025

Read RIS-Aided Near-Field Channel Estimation under Mutual Coupling and Spatial Correlation by Ahmad Dkhan, Simon Tarboush, Hadi Sarieddeen, Tareq Y. Al-Naffouri (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-01828.txt.

Introduction

Ahmad Dkhan, Simon Tarboush, Hadi Sarieddeen, and Tareq Al-Naffouri improve channel estimation for future wireless systems in a 2025 paper. Smart surfaces that reflect radio waves can boost huge antenna arrays, but they make channel estimation hard. Dense parts also couple to each other, signals correlate in space, and near field effects grow.

The authors extend a subspace based estimator to handle coupling in near field settings within a broad system framework. Ignoring coupling costs dearly: accounting for it gained about a set margin at a modest signal level in their analysis. They also study how coupling changes the usable spatial channels.

Catalog record

CREATORAhmad Dkhan, Simon Tarboush, Hadi Sarieddeen, Tareq Y. Al-Naffouri
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE33 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-01828

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

RIS-Aided Near-Field Channel Estimation under Mutual Coupling and Spatial Correlation Thanks: This work is supported by the AUB’s University Research Board and Vertically Integrated Projects Program, and KAUST’s Office of Sponsored Research under Award No. ORFS-CRG12-2024-6478. Ahmad Dkhan, Simon Tarboush, Hadi Sarieddeen, Tareq Y. Al-Naffouri Abstract The integration of reconfigurable intelligent surfaces (RIS) with extremely large multiple-input multiple-output (MIMO) arrays at the base station has emerged as a key enabler for enhancing wireless network performance. However, this setup introduces high-dimensional channel matrices, leading to increased computational complexity and pilot overhead in channel estimation. Mutual coupling (MC) effects among densely packed unit cells, spatial correlation, and near-field propagation conditions further complicate the estimation process. Conventional estimators, such as linear minimum mean square error (MMSE), require channel statistics that are challenging to acquire for high-dimensional arrays, while least squares (LS) estimators suffer from performance limitations. To address these challenges, the reduced-subspace least squares (RS-LS) estimator leverages array geometry to enhance estimation accuracy. This work advances the promising RS-LS estimation algorithm by explicitly incorporating MC effects into the more realistic and challenging near-field propagation environment within the increasingly relevant generalized RIS-aided MIMO framework. Additionally, we investigate the impact of MC on the spatial degrees of freedom (DoF). Our analysis reveals that accounting for MC effects provides a significant performance gain of approximately 5 dB at an SNR of 5 dB, compared to conventional methods that ignore MC. I Introduction Reconfigurable intelligent surfaces have gained significant attention for their ability to dynamically shape wireless propagation by adjusting the impedance of unit cells to control signal reflections and transmissions [ 1 , 2 , 3 ] . Such signal shaping enhances coverage, signal reliability, and overall communication performance. Accurate channel modeling is crucial, particularly for large-scale reconfigurable intelligent surfaces, where the expanded surface aperture extends the near-field region [ 4 ] . Unlike far-field conditions, near-field channels exhibit spherical wave propagation and spatial non-stationarity [ 4 , 5 ] . Additionally, the mutual coupling effect in dense multiple-input multiple-output arrays and reconfigurable intelligent surfaces introduces electromagnetic interactions that distort radiation patterns, alter channel characteristics, and intensify spatial correlation, leading to performance bottlenecks [ 6 , 7 ] . Studies such as [ 8 ] investigate mutual coupling effects on holographic reconfigurable intelligent surfaces, analyzing their influence on eigenvalues, spatial correlation, and array gain. Moreover, neglecting mutual coupling leads to significant estimation errors and system degradation, as demonstrated in recent works [ 9 , 10 , 11 ] . The authors in [ 9 ] highlight that mutual coupling critically affects multiple-input multiple-output channel estimation, even more than spatial correlation. Similarly, [ 10 ] demonstrates that in active reconfigurable intelligent surfaces with large amplitudes, mutual coupling can significantly degrade estimation performance in tightly spaced or large-scale setups. The study in [ 11 ] highlights the crucial role of accounting for mutual coupling in reconfigurable intelligent surface-aided channel estimation and beamforming. Despite extensive studies on mutual coupling, its impact in the near-field remains largely unexplored for large arrays and surfaces. Unlocking the potential of reconfigurable intelligent surface technology depends on accurate channel estimation, which is challenging due to the unique characteristics of reconfigurable intelligent surface-aided links [ 12 ] . Deploying numerous reflecting surfaces alongside multiple-input multiple-output arrays at the base station results in high-dimensional channel matrices, increasing estimation overhead and computational complexity [ 13 ] . The minimum mean square error estimator requires perfect knowledge of channel statistics, which is difficult to obtain in high-dimensional settings, while least square estimation, though independent of prior information, performs poorly, especially at low signal-to-noise ratio [ 13 ] . To address these challenges, [ 14 ] proposed the reduced-subspace least squares estimator, which leverages channel geometry to reduce dimensionality and improve estimation accuracy in far-field reconfigurable intelligent surface-aided systems; its effectiveness in near-field multiple-input multiple-output systems was later explored in [ 15 ] . The potential of the reduced-subspace least squares estimator to exploit mutual coupling information in a generalized MIMO-RIS setting with accurate near-field modeling is yet to be investigated. This paper addresses this gap by integrating mutual coupling effects and near-field characteristics into the channel estimation process. We analyze their combined impact on linear estimators and propose an enhanced reduced-subspace least squares estimator that leverages mutual coupling information to improve estimation accuracy. We further derive the general error covariance matrix for reconfigurable intelligent surface-aided multiple-input multiple-output systems and examine the degradation of spatial degrees of freedom (DoF) due to strong mutual coupling in dense antenna arrays, highlighting its implications for estimation performance. Our analysis shows that integrating mutual coupling into the reduced-subspace least squares estimator improves accuracy by approximately 5\,\mathrm{dB} at an signal-to-noise ratio of 5\,\mathrm{dB} compared to the reduced-subspace least squares estimator without mutual coupling consideration. Furthermore, we identify the most promising scenarios for the reduced-subspace least squares estimator, demonstrating that its highest performance gain occurs at small inter-element spacing. Regarding notation, scalars (a,A) , vectors ({\bf a}) , and matrices ({\bf A}) are represented by non-bold, bold lowercase, and bold uppercase letters, respectively. \mathbf{I}_{M} is an identity matrix of size M\times M . (\cdot)^{\sf T} , (\cdot)^{\sf*} , (\cdot)^{\sf H} , and {(\cdot)}^{-1} stand for the transpose, conjugate, Hermitian, and inverse operators, respectively. \mathbf{A}\odot\mathbf{B} , \mathbf{A}\otimes\mathbf{B} , and \mathbf{A}\diamond\mathbf{B} denote the Hadamard, Kronecker, and Khatri–Rao products, respectively. \mbox{\bb E}[\cdot] denotes the expectation operator. \mathrm{diag}\left(a_{1},a_{2},\dots,a_{N}\right) is an N\times N diagonal matrix with diagonal entries \{a_{1},a_{2},\dots,a_{N}\} . The floor and ceiling operations are denoted by \lfloor\cdot\rfloor and \lceil\cdot\rceil , respectively. The modulo operation, \mathrm{mod}(a,b) , returns the remainder of the division of a by b . II System Model We analyze the uplink of a time-division duplex reconfigurable intelligent surface-aided multiple-input multiple-output communication system, where a user equipment with N=N_{\rm H}\times N_{\rm V} antennas communicates with a base station with M=M_{\rm H}\times M_{\rm V} antennas through a nearly-passive reconfigurable intelligent surface consisting of K=K_{\rm H}\times K_{\rm V} unit cells. We assume the direct user equipment-base station link is blocked, focusing entirely on the role of the reconfigurable intelligent surface in signal propagation and channel estimation. We assume a Cartesian coordinate system where the reconfigurable intelligent surface, base station, and user equipment elements form a uniform planner array on the Y-Z plane, as shown in Fig. 1 . The position of the k th unit cell of the reconfigurable intelligent surface ( k\!\in\!\{1,\dots,K\} ), the m th antenna of the base station ( m\!\in\!\{1,\dots,M\} ), and the n th antenna of the user equipment ( n\!\in\!\{1,\dots,N\} ) are given by {\bf p}_{k}=[0,i_{k}\delta_{{\rm R}},j_{k}\delta_{{\rm R}}] , {\bf p}_{m}=[0,i_{m}\delta_{{\rm B}},j_{m}\delta_{{\rm B}}] , and {\bf p}_{n}=[0,i_{n}\delta_{{\rm U}},j_{n}\delta_{{\rm U}}] , where i_{k}=\text{mod}(k\!-\!1,K_{\rm H}) , j_{k}=\left\lfloor(k\!-\!1)/{K_{\rm H}}\right\rfloor , i_{m}=\text{mod}(m\!-\!1,M_{\rm H}) , j_{m}=\left\lfloor(m\!-\!1)/{M_{\rm H}}\right\rfloor , i_{n}=\text{mod}(n\!-\!1,N_{\rm H}) , and j_{n}=\left\lfloor(n\!-\!1)/{N_{\rm H}}\right\rfloor . Here, \delta_{{\rm R}} , \delta_{{\rm B}} , and \delta_{{\rm U}} represent the inter-element spacings for the reconfigurable intelligent surface, base station, and user equipment, respectively, and K_{\rm H} , M_{\rm H} , and N_{\rm H} are the number of elements along the horizontal axis for each. We model a narrowband channel with block fading, assuming static channels within coherence intervals. During uplink channel training, the phase shifts of the reconfigurable intelligent surface elements are adjusted. At the \ell th training step (

Read the complete text