Learning Radio Environments by Differentiable Ray Tracing

TL;DR

This paper introduces a differentiable ray tracing method that enables calibration of radio environment models by computing derivatives of channel responses with respect to scene parameters, validated on synthetic and real data.

Contribution

It presents a novel gradient-based calibration approach with differentiable parametrizations, integrating with ray tracers to optimize scene properties for accurate channel modeling.

Findings

Effective calibration on synthetic data

Successful validation with real-world indoor measurements

Enhanced accuracy in modeling radio environments

Abstract

Ray tracing (RT) is instrumental in 6G research in order to generate spatially-consistent and environment-specific channel impulse responses (CIRs). While acquiring accurate scene geometries is now relatively straightforward, determining material characteristics requires precise calibration using channel measurements. We therefore introduce a novel gradient-based calibration method, complemented by differentiable parametrizations of material properties, scattering and antenna patterns. Our method seamlessly integrates with differentiable ray tracers that enable the computation of derivatives of CIRs with respect to these parameters. Essentially, we approach field computation as a large computational graph wherein parameters are trainable akin to weights of a neural network (NN). We have validated our method using both synthetic data and real-world indoor channel measurements, employing a distributed multiple-input multiple-output (MIMO) channel sounder.