Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

X-REFINE: XAI-based RElevance input-Filtering and archItecture fiNe-tuning for channel Estimation

Abdul Karim Gizzini, Yahia Medjahdi

TL;DR

Simulation results demonstrate that X-REFINE achieves a superior interpretability-performance-complexity trade-off, significantly reducing computational complexity while maintaining robust bit error rate (BER) performance across different scenarios.

Abstract

AI-native architectures are vital for 6G wireless communications. The black-box nature and high complexity of deep learning models employed in critical applications, such as channel estimation, limit their practical deployment. While perturbation-based XAI solutions offer input filtering, they often neglect internal structural optimization. We propose X-REFINE, an XAI-based framework for joint input-filtering and architecture fine-tuning. By utilizing a decomposition-based, sign-stabilized LRP epsilon rule, X-REFINE backpropagates predictions to derive high-resolution relevance scores for both subcarriers and hidden neurons. This enables a holistic optimization that identifies the most faithful model components. Simulation results demonstrate that X-REFINE achieves a superior interpretability-performance-complexity trade-off, significantly reducing computational complexity while maintaining robust bit error rate (BER) performance across different scenarios.

X-REFINE: XAI-based RElevance input-Filtering and archItecture fiNe-tuning for channel Estimation

TL;DR

Simulation results demonstrate that X-REFINE achieves a superior interpretability-performance-complexity trade-off, significantly reducing computational complexity while maintaining robust bit error rate (BER) performance across different scenarios.

Abstract

AI-native architectures are vital for 6G wireless communications. The black-box nature and high complexity of deep learning models employed in critical applications, such as channel estimation, limit their practical deployment. While perturbation-based XAI solutions offer input filtering, they often neglect internal structural optimization. We propose X-REFINE, an XAI-based framework for joint input-filtering and architecture fine-tuning. By utilizing a decomposition-based, sign-stabilized LRP epsilon rule, X-REFINE backpropagates predictions to derive high-resolution relevance scores for both subcarriers and hidden neurons. This enables a holistic optimization that identifies the most faithful model components. Simulation results demonstrate that X-REFINE achieves a superior interpretability-performance-complexity trade-off, significantly reducing computational complexity while maintaining robust bit error rate (BER) performance across different scenarios.
Paper Structure (12 sections, 8 equations, 4 figures, 1 algorithm)

This paper contains 12 sections, 8 equations, 4 figures, 1 algorithm.

Table of Contents

  1. Introduction
  2. System Model
  3. Proposed X-REFINE Framework
  4. X-REFINE Double Optimization Problem
  5. LRP XAI Scheme
  6. Computational Complexity Reduction
  7. Simulation Results
  8. X-REFINE Double Optimization
  9. Impact of Modulation Order
  10. Impact of Frequency Selectivity
  11. Computational Complexity Analysis
  12. Conclusion

Figures (4)

  • Figure 1: Block diagram of the proposed X-REFINE framework.
  • Figure 2: Relevance Scores distribution of XAI-CHEST, and the proposed X-REFINE framework considering the LF and HF channel models under different modulation schemes.
  • Figure 3: BER performance for the LF and HF channel models under different modulation schemes.
  • Figure 4: Complexity comparison across different scenarios. The baseline is the classical FNN model considering full inputs and full architecture.