Physics-informed line-of-sight learning for scalable deterministic channel modeling

TL;DR

This paper introduces D$^2$LoS, a physics-informed neural network that efficiently predicts line-of-sight regions in complex environments, significantly reducing computation time while maintaining high accuracy.

Contribution

The paper presents a novel neural network approach that reformulates LoS prediction into sparse classification and regression, with a geometric post-processing step for physical accuracy.

Findings

Achieves 3.28 dB mean absolute error in received power

Accelerates visibility computation by over 25 times

Constructs RayVerse-100 dataset with detailed urban scenarios

Abstract

Deterministic channel modeling maps a physical environment to its site-specific electromagnetic response. Ray tracing produces complete multi-dimensional channel information but remains prohibitively expensive for area-wide deployment. We identify line-of-sight (LoS) region determination as the dominant bottleneck. To address this, we propose D$^2$LoS, a physics-informed neural network that reformulates dense pixel-level LoS prediction into sparse vertex-level visibility classification and projection point regression, avoiding the spectral bias at sharp boundaries. A geometric post-processing step enforces hard physical constraints, yielding exact piecewise-linear boundaries. Because LoS computation depends only on building geometry, cross-band channel information is obtained by updating material parameters without retraining. We also construct RayVerse-100, a ray-level dataset spanning 100 urban scenarios with per-ray complex gain, angle, delay, and geometric trajectory. Evaluated against rigorous ray tracing ground truth, D$^2$LoS achieves 3.28~dB mean absolute error in received power, 4.65$^\circ$ angular spread error, and 20.64~ns delay spread error, while accelerating visibility computation by over 25$\times$.