Electromagnetic Information Theory-Based Statistical Channel Model for Improved Channel Estimation

TL;DR

This work introduces an electromagnetic information theory (EIT) based statistical channel model by formulating the electromagnetic correlation function (EMCF) as a physically grounded prior derived from Maxwell's equations and a von Mises–Fisher distribution. The EMCF enables closed-form spatio-temporal channel modeling and supports efficient, EM-informed covariance estimation (EIT-Cov) and EIT-MMSE channel estimation by embedding EM priors into classical MMSE inference. Through Gaussian process regression and kernel learning, the approach achieves superior covariance estimation and channel estimation performance compared to traditional baselines, especially in data-scarce settings. The framework offers a principled path to integrate EM physics into wireless system design, with potential extensions to wideband and Doppler-driven channel prediction tasks.

Abstract

Electromagnetic information theory (EIT) is an emerging interdisciplinary subject that integrates classical Maxwell electromagnetics and Shannon information theory. The goal of EIT is to uncover the information transmission mechanisms from an electromagnetic (EM) perspective in wireless systems. Existing works on EIT are mainly focused on the analysis of EM channel characteristics, degrees-of-freedom, and system capacity. However, these works do not clarify how to integrate EIT knowledge into the design and optimization of wireless systems. To fill in this gap, in this paper, we propose an EIT-based statistical channel model with simplified parameterization. Thanks to the simplified closed-form expression of the EMCF, it can be readily applied to various channel modeling and inference tasks. Specifically, by averaging the solutions of Maxwell's equations over a tunable von Mises distribution, we obtain a spatio-temporal correlation function (STCF) model of the EM channel, which we name as the EMCF. Furthermore, by tuning the parameters of the EMCF, we propose an EIT-based covariance estimator (EIT-Cov) to accurately capture the channel covariance. Since classical MMSE estimators can exploit prior information contained in the channel covariance matrix, we further propose the EIT-MMSE channel estimator by substituting EMCF for the covariance matrix. Simulation results show that both the proposed EIT-Cov covariance estimator and the EIT-MMSE channel estimator outperform their baseline algorithms, thus proving that EIT is beneficial to wireless communication systems.

Figures (9)

• Figure 1: Slice chart of the space-time EMCF ${\bf K}_{\rm EM}({\bf r}, t)$, where ${\bf r} = (x, y, z)$. The real part and the imaginary part of the EMCF are shown in two horizontal groups.
• Figure 2: Kernel entropy as a function of the concentration parameter $\mu$ with different antenna spacing settings.
• Figure 3: Schematic of the EIT-MMSE channel inference framework combined with EIT-Cov parameter tuning. The EMCF is first tuned by observed pilot vector ${\bf y}$. Then, the correlation matrix generated by EMCF is fed into the classical MMSE channel estimator to obtain the channel estimates.
• Figure 4: GPR objective function $\ln p({\bf y}|{\bm \mu}_{x,z})$ and trajectory of the Armijo-Goldstein backtracking optimizer (in red pentagrams).
• Figure 5: Average covariance estimation performance of different channel statistics estimators. The SNR is set to $\gamma = 10\,{\rm dB}$. Error bars are computed from 1000 Monte Carlo trials.

Theorems & Definitions (8)

• Remark 1
• Remark 2
• Remark 3
• Remark 4
• Remark 5
• Remark 6
• Remark 7
• Lemma 1: Derivative of the EMCF