Said Mikki

We propose a numerical method grounded in electromagnetic random field theory (EM-RFT) for characterizing and modeling random fields in electromagnetics, with future applications to information transmission and processing. By combining EM Green’s functions with the Karhunen–Loève (KL) expansion, the method enables efficient estimation of spatially continuous EM random fields through a finite set of fundamental KL modes. The key idea in EM-RFT is the use of EM Green’s functions to avoid separate KL expansions for each field or current; once the system response is determined, input information fields can be propagated to generate consistent random field models. Building upon this formulation, we further construct stochastic transmitter–scatterer–receiver configurations and analyze the channel capacity, enabling the modeling of random propagation environments. This unified framework provides physics-based analysis of dense-aperture information fields and evaluation of capacity over arbitrary continuous regions, establishing a computational foundation for future studies of realistic cMIMO channels through a synthesis of EM theory and information-theoretic principles.

We provide an electromagnetics (EM)-complete wireless system modeling framework for line-of-sight (LoS) communications, integrating Active Element Pattern (AEP)-based Infinitesimal Dipole Modeling (IDM) with dyadic Green’s function (GF) representation of EM wave propagation in practical generic MIMO links with arbitrary mutual coupling between the transceiver’s antenna elements. This novel AEP-IDM-GF methodology addresses three fundamental limitations of standardized models: (i) full-wave characterization of mutual coupling, (ii) incorporation of dyadic GFs beyond conventional spherical wave propagation physics, and (iii) polarization-complete channel representation. Among the main contributions of this work is the first complete derivation and decomposition of the MIMO channel matrix in terms of currents and free-space Green’s functions: we establish a rigorous integral representation that fully accounts for EM mutual coupling and introduce a novel decomposition of the channel matrix into isolated and mutual-coupling-induced components, which provides a foundational tool for quantitatively analyzing the effects of wave propagation and coupling effect. An efficient specialized implementation based on the IDM is also presented and compared with conventional full-wave EM solvers. The framework achieves sub-1◦ phase coherence and 0.1 dBm near-field (NF) power response fidelity relative to full-wave EM simulations, while maintaining computational efficiency with speedups of up to 42.9× compared to full-wave method of moments (MoM) simulations. This combination of accuracy and efficiency exposes inherent limitations in 3GPP models that significantly overestimate capacity in high-SNR regimes and underestimate performance in strong coupling configurations and low-SNR regimes. Numerical analysis of selected examples comprised of dense MIMO configurations demonstrates that mutual coupling interacts with full-wave spatial correlation to redistribute energy across eigenmodes: while spatial correlation leads to extreme concentration in dominant modes, mutual coupling of specific array configuration mitigates this concentration and improves energy spread to moderate-order modes, thereby increasing effective degrees of freedom (eDoF) and enhancing multiplexing capability, though with constraints on absolute multiplexing gain at high SNR due to fundamental spatial correlation limits.

As a rigorous and comprehensive foundation for electromagnetic information theory (EIT), we develop a general theory that elucidates the universal stochastic structure of radiated electromagnetic (EM) fields and induced currents in generic EM information transmission systems. The framework encompasses arbitrary random scatterers, input information fields, and EM mutual coupling. The system is modeled as a multiply connected, arbitrary Riemannian manifold within the language of differential geometry. Our approach exploits exact Green’s functions (GFs) on manifolds to construct a novel electromagnetic random field theory (EM-RFT). Interpreted as response functions localized on the surfaces of transceivers and scatterers, the GFs allow us to treat the internal physical details of the EM system as a black box, redirecting analytical attention toward external input–output relations in line with signal processing and communication theory. This integration of random fields (RFs), electromagnetics, and GFs yields a unified framework for deriving and characterizing the stochastic structure of arbitrary EM information transmission systems. We rigorously establish that EM random fields satisfying Maxwell’s equations can always be constructed using system GFs driven by external information fields. The theory further decouples stochastic input RFs from random fluctuations associated with the communication medium (e.g., scatterers), and introduces general correlation propagators valid for arbitrary EM links. Using the Karhunen–Loève expansion, all EM random fields are represented as sums of random variables, providing both a simulation framework for arbitrary EM RFs and a basis for evaluating mutual information between input and output spatial domains at arbitrary locations in the system.

We present a comprehensive full-wave random field (RF) Green’s function (GF) computational framework tailored to electromagnetic (EM) information theory (EIT) and stochastic electromagnetics, with direct relevance to communication systems. The framework is anchored in exact dyadic Green’s functions, including the transceiver antenna current Green’s function (ACGF), hence shifting the emphasis from internal structural details to input–output relations aligned with the spirit of signal processing principles. The Karhunen–Loève expansion (KLE) is employed to represent all fields in terms of orthogonal modal bases, enabling physically consistent simulations over arbitrary spatial domains. Our results demonstrate that, even for input Gaussian RFs, EM communication systems may give rise to non-Gaussian and improper complex RFs. We also show how mutual information can be computed directly from intrinsic EM fluctuations in noise-free settings. Overall, this novel computational method furnishes a flexible and efficient platform for analyzing and optimizing EM information in emerging communication technologies.

Observing the evolution of mobile communication networks across successive generations, frequency band allocation has progressively expanded both in quantity and frequency. The relative simplicity of generating high-frequency signals via heterodyning of optical carriers in analog radio-over-fiber systems demonstrates considerable potential for enabling millimeter-wave and even terahertz-range wireless communication systems. Nevertheless, the prospective utilization of sub-terahertz or higher frequency bands in future wireless networks may introduce significant phase fluctuations in the received signal due to degraded phase coherence, a consequence of the large frequency separation between optical carriers. This degradation underscores the necessity of implementing phase correction techniques at the receiver. This paper reexamines the heterodyne-based detection approach as a potential phase noise correction method at the receiver, incorporating machine learning techniques. This study primarily addresses the inherent limitation of the conventional heterodyne-based detection method, which requires a minimum frequency spacing between the reference tone and the wireless signal to prevent signal-to-signal beating interference. The implementation of a machine learning algorithm aims to suppress signal-to-signal beating interference without relying on additional bandpass filters or previously proposed machine learning-based solutions, which often involve more complex configurations. The objective is to preserve the simplicity of conventional heterodyne-based receivers, while eliminating the need for high-frequency oscillators for frequency downconversion. Our results indicate that integrating a machine learning algorithm into the heterodyne-based detection approach can effectively suppress signal-to-signal beating interference, leading to improved detection accuracy. This enhancement surpasses the performance of conventional heterodyne-based detection, reducing the symbol error rate from 10⁻² to 10⁻⁶.

We present a rigorous, highly efficient, and fully general framework for estimating cross-correlation in multiple antenna systems based on the concept of random electromagnetic (EM) sources. The proposed theory computes the stochastic port-to-port cross-correlation between two arbitrary receiving antennas by modeling the excitation as a random current source localized in either the far-or near-field region. This method leverages stochastic source models, which are mathematically and computationally more tractable than conventional random field approaches, and it is derived using the electromagnetic reciprocity theorem. The validity of the model is confirmed through comparisons with full-wave EM Monte Carlo simulations. When reproducing the Clarke-Jakes model in the far field, naive uniform sampling of random sources leads to discrepancies with the exact EM predictions. To resolve this issue and accurately recover the classical Clarke-Jakes power angle spectrum (PAS), we introduce a modified angular distribution scheme for randomizing source positions and orientations. Furthermore, we demonstrate that using standard zero-mean Gaussian-distributed random variables in Cartesian coordinates yields statistically equivalent results to the proposed angular sampling strategy. Numerical studies show excellent agreement and confirm the validity and robustness of the proposed framework.

We provide an electromagnetics (EM)-complete wireless system modeling framework for line-of-sight (LoS) communications, integrating Active Element Pattern (AEP)-based Infinitesimal Dipole Modeling (IDM) with dyadic Green's function (GF) representation of EM wave propagation in practical generic MIMO links with arbitrary mutual coupling between the transceiver's antenna elements. This novel AEP-IDM-GF methodology addresses three fundamental limitations of standardized models: (i) full-wave characterization of mutual coupling, (ii) incorporation of dyadic GFs beyond conventional spherical wave propagation physics, and (iii) polarization-complete channel representation. Among the main contributions of this work is the first complete derivation of the MIMO channel matrix in terms of currents and free-space Green's functions, independent of any specific numerical method, by employing an integral representation approach while fully taking into acount EM mutual coupling. An efficient specialized implementation based on the IDM is also presented and compared with conventional full-wave EM solvers. The framework achieves sub-1◦phase coherence and 0.1 dBm near-field (NF) power response fidelity relative to full-wave EM simulations, while maintaining computational efficiency with speedups of up to 42.9× compared to full-wave method of moments (MoM) simulations. This combination of accuracy and efficiency exposes inherent limitations in 3GPP models that significantly overestimate capacity in high-SNR regimes and underestimate performance in strong coupling configurations and low-SNR regimes. Numerical analysis of selected examples comprised of dense MIMO configurations demonstrates that mutual coupling enhances dominant eigenmodes while suppressing higher-order modes, thereby increasing effective degrees of freedom (eDoF) through EM energy concentration with the cost of constraining the theoretical multiplexing gain in high-SNR regimes.