Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Towards Channel Charting Enhancement with Non-Reconfigurable Intelligent Surfaces

Mahdi Maleki, Reza Agahzadeh Ayoubi, Marouan Mizmizi, Umberto Spagnolini

TL;DR

The paper tackles channel charting for localization in dense urban environments by leveraging static, fully passive EMS to enrich propagation. It combines semi-supervised t-SNE and a semi-supervised Autoencoder to validate embedding quality under nonparametric and parametric mappings, respectively, and shows that a quantile-driven EMS phase design yields robust CC by balancing SNR and spatial dissimilarity. In a 3D ray-traced city at 30 GHz, optimized static EMS reduce the 90th-percentile localization error from over 50 m to below 25 m and markedly improve trajectory continuity under moderate supervision, demonstrating practical benefits without reconfiguration. The findings emphasize that static EMS can materially enhance CC performance and localization reliability, guiding future joint EMS-placement and codebook design for scalable, low-overhead deployment in 6G networks.

Abstract

We investigate how fully-passive electromagnetic skins (EMSs) can be engineered to enhance channel charting (CC) in dense urban environments. We employ two complementary state-of-the-art CC techniques, semi-supervised t-distributed stochastic neighbor embedding (t-SNE) and a semi-supervised Autoencoder (AE), to verify the consistency of results across nonparametric and parametric mappings. We show that the accuracy of CC hinges on a balance between signal-to-noise ratio (SNR) and spatial dissimilarity: EMS codebooks that only maximize gain, as in conventional Reconfigurable Intelligent Surface (RIS) optimization, suppress location fingerprints and degrade CC, while randomized phases increase diversity but reduce SNR. To address this trade-off, we design static EMS phase profiles via a quantile-driven criterion that targets worst-case users and improves both trustworthiness and continuity. In a 3D ray-traced city at 30 GHz, the proposed EMS reduces the 90th-percentile localization error from > 50 m to < 25 m for both t-SNE and AE-based CC, and decreases severe trajectory dropouts by over 4x under 15% supervision. The improvements hold consistently across the evaluated configurations, establishing static, pre-configured EMS as a practical enabler of CC without reconfiguration overheads.

Towards Channel Charting Enhancement with Non-Reconfigurable Intelligent Surfaces

TL;DR

The paper tackles channel charting for localization in dense urban environments by leveraging static, fully passive EMS to enrich propagation. It combines semi-supervised t-SNE and a semi-supervised Autoencoder to validate embedding quality under nonparametric and parametric mappings, respectively, and shows that a quantile-driven EMS phase design yields robust CC by balancing SNR and spatial dissimilarity. In a 3D ray-traced city at 30 GHz, optimized static EMS reduce the 90th-percentile localization error from over 50 m to below 25 m and markedly improve trajectory continuity under moderate supervision, demonstrating practical benefits without reconfiguration. The findings emphasize that static EMS can materially enhance CC performance and localization reliability, guiding future joint EMS-placement and codebook design for scalable, low-overhead deployment in 6G networks.

Abstract

We investigate how fully-passive electromagnetic skins (EMSs) can be engineered to enhance channel charting (CC) in dense urban environments. We employ two complementary state-of-the-art CC techniques, semi-supervised t-distributed stochastic neighbor embedding (t-SNE) and a semi-supervised Autoencoder (AE), to verify the consistency of results across nonparametric and parametric mappings. We show that the accuracy of CC hinges on a balance between signal-to-noise ratio (SNR) and spatial dissimilarity: EMS codebooks that only maximize gain, as in conventional Reconfigurable Intelligent Surface (RIS) optimization, suppress location fingerprints and degrade CC, while randomized phases increase diversity but reduce SNR. To address this trade-off, we design static EMS phase profiles via a quantile-driven criterion that targets worst-case users and improves both trustworthiness and continuity. In a 3D ray-traced city at 30 GHz, the proposed EMS reduces the 90th-percentile localization error from > 50 m to < 25 m for both t-SNE and AE-based CC, and decreases severe trajectory dropouts by over 4x under 15% supervision. The improvements hold consistently across the evaluated configurations, establishing static, pre-configured EMS as a practical enabler of CC without reconfiguration overheads.

Paper Structure

This paper contains 21 sections, 30 equations, 9 figures, 2 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. System Model
  3. Signal Model
  4. Channel Model
  5. Channel Charting Methods
  6. Feature Construction and Dissimilarity Metrics
  7. Nonlinear Dimensionality Reduction: t-Distributed Stochastic Neighbor Embedding
  8. Nonlinear Dimensionality Reduction: Fully Connected Autoencoder
  9. Channel Charting in Smart Radio Environment
  10. Evaluation Metrics
  11. Phase Profile Parameterization
  12. Optimization Problem: Continuous and Codebook-Based Formulation
  13. Codebook Construction for Phase Profiles
  14. Results and Discussion
  15. Scenario Description and Simulation Setup
  16. ...and 6 more sections

Figures (9)

  • Figure 1: Reference scenario and system, adapted from Maleki_CC_in_SRE
  • Figure 2: Structure of the used for channel charting. The orange and green colors represent the ReLU and tanh activation functions, respectively, used in the hidden layers.
  • Figure 3: Geometry of the considered scenario, adapted from Maleki_CC_in_SRE
  • Figure 4: Channel chart embeddings (, 15% supervision): (a) ground truth positions, (b) no , (c) specular , (d) best codebook s. Colors reflect $y$-coordinates. Only codebook-optimized s recover full spatial geometry, especially in .
  • Figure 5: Trajectory estimation using t-SNE (15 % supervision): (top) no EMS, (bottom) both EMSs with best codebook. Blue: ground truth; circles: inferred positions (red = error $>25$ m). Optimized cut severe outliers by over 4× and yield smooth, gap-free tracking.
  • ...and 4 more figures