Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Bayesian EM Digital Twins Channel Estimation

Lorenzo Del Moro, Francesco Linsalata, Marouan Mizmizi, Maurizio Magarini, Damiano Badini, Umberto Spagnolini

TL;DR

This work tackles CSI acquisition for future wireless networks by integrating Electromagnetic Digital Twins (EM-DT) as a priori priors to guide channel estimation. It introduces a Bayesian basis projection that leverages EM-DT-derived slow-varying spatial and temporal subspaces to project LS estimates, reducing pilot overhead while maintaining high spectral efficiency. Numerical results show about a $20$ dB NMSE improvement over LS at low $SNR$ and spectral efficiency close to ideal CSI, with substantial pilot reductions (e.g., from $N_p=18$ to $N_p=2$). The approach combines real-time EM-DT updates, ray-tracing-based propagation modes, and a low-complexity modal filter, offering a practical path toward environment-aware CSI in 6G and beyond.

Abstract

This letter proposes a Bayesian channel estimation method that leverages on the a priori information provided by the Electromagnetic Digital Twin's (EM-DT) representation of the environment. The proposed approach is compared with several conventional techniques in terms of Normalized Mean Square Error (NMSE), spectral efficiency, and number of pilots. Simulations prove more than $10\,$dB gain in NMSE and a spectral efficiency comparable to that of the ideal channel state information, for different signal-to-noise ratio (SNR) values. Additionally, the Bayesian EM-DT-empowered channel estimation enables a remarkable pilot reduction compared to maximum likelihood methods at low SNR.

Bayesian EM Digital Twins Channel Estimation

TL;DR

This work tackles CSI acquisition for future wireless networks by integrating Electromagnetic Digital Twins (EM-DT) as a priori priors to guide channel estimation. It introduces a Bayesian basis projection that leverages EM-DT-derived slow-varying spatial and temporal subspaces to project LS estimates, reducing pilot overhead while maintaining high spectral efficiency. Numerical results show about a dB NMSE improvement over LS at low and spectral efficiency close to ideal CSI, with substantial pilot reductions (e.g., from to ). The approach combines real-time EM-DT updates, ray-tracing-based propagation modes, and a low-complexity modal filter, offering a practical path toward environment-aware CSI in 6G and beyond.

Abstract

This letter proposes a Bayesian channel estimation method that leverages on the a priori information provided by the Electromagnetic Digital Twin's (EM-DT) representation of the environment. The proposed approach is compared with several conventional techniques in terms of Normalized Mean Square Error (NMSE), spectral efficiency, and number of pilots. Simulations prove more than dB gain in NMSE and a spectral efficiency comparable to that of the ideal channel state information, for different signal-to-noise ratio (SNR) values. Additionally, the Bayesian EM-DT-empowered channel estimation enables a remarkable pilot reduction compared to maximum likelihood methods at low SNR.
Paper Structure (9 sections, 14 equations, 5 figures)

This paper contains 9 sections, 14 equations, 5 figures.

Table of Contents

  1. Introduction
  2. System model
  3. Transmitted and Received Signal
  4. Channel Model
  5. Proposed Bayesian EM-DT-empowered channel Estimation Method
  6. Electromagnetic Digital Twin Framework
  7. Bayesian Basis Projection
  8. Numerical results
  9. Conclusion

Figures (5)

  • Figure 1: Reference EM-DT framework.
  • Figure 2: NMSE comparison with $N_p = 32$ among state-of-the-art approaches and our proposed method. Solid lines are the analytical results, while markers denote the simulated results. Dashed lines are the normalized error variance thresholds $\sigma^2_{N_{T_B}}$, $\sigma^2_{L_{\tau_\mathrm{max}}}$and $\sigma^2_{\bar{L}}$ for the BML, the denoising and the proposed method, respectively.
  • Figure 3: Spectral efficiency comparison when $N_p = 32$ among state-of-the-art approaches and our proposed method.
  • Figure 4: ECDF comparison for a SNR at the decision variable (after combing) when considering SNR $= -10$ dB and SNR $= 5$ dB among state-of-the-art approaches and our proposed method with $N_p=32$.
  • Figure 5: Comparison of a) MSE and b) spectral efficiency for the LS and the proposed approach, varying pilot subcarriers with respect to full piloting.