An Efficient and Explainable KAN Framework for Wireless Radiation Field Prediction

TL;DR

This paper introduces a novel KAN framework with transformer modules for wireless radiation field prediction, capturing environmental context more effectively and providing explainability, leading to improved accuracy and efficiency.

Contribution

The paper proposes a new KAN-based neural network architecture with transformers that models complete radio rays, enhancing prediction accuracy and interpretability in wireless channel modeling.

Findings

Outperforms existing methods in realistic and synthetic scenes

Demonstrates improved accuracy with ablation studies confirming component contributions

Provides explainability for model performance

Abstract

Modeling wireless channels accurately remains a challenge due to environmental variations and signal uncertainties. Recent neural networks can learn radio frequency~(RF) signal propagation patterns, but they process each voxel on the ray independently, without considering global context or environmental factors. Our paper presents a new approach that learns comprehensive representations of complete rays rather than individual points, capturing more detailed environmental features. We integrate a Kolmogorov-Arnold network (KAN) architecture with transformer modules to achieve better performance across realistic and synthetic scenes while maintaining computational efficiency. Our experimental results show that this approach outperforms existing methods in various scenarios. Ablation studies confirm that each component of our model contributes to its effectiveness. Additional experiments provide clear explanations for our model's performance.