Radio Map Prediction from Noisy Environment Information and Sparse Observations

TL;DR

This paper introduces a deep learning approach that compensates for noisy environment data in radio map prediction, demonstrating improved accuracy over traditional methods in indoor scenarios.

Contribution

The study shows that training CNNs with perturbed environment data enhances robustness to input noise, simplifying environment encoding without sacrificing accuracy.

Findings

Models trained with perturbed data reduce error by up to 25%.

Achieved 2.1 dB RMSE with noisy inputs on real data.

Binary occupancy encoding performs as well as detailed material info.

Abstract

Many works have investigated radio map and path loss prediction in wireless networks using deep learning, in particular using convolutional neural networks. However, most assume perfect environment information, which is unrealistic in practice due to sensor limitations, mapping errors, and temporal changes. We demonstrate that convolutional neural networks trained with task-specific perturbations of geometry, materials, and Tx positions can implicitly compensate for prediction errors caused by inaccurate environment inputs. When tested with noisy inputs on synthetic indoor scenarios, models trained with perturbed environment data reduce error by up to 25\% compared to models trained on clean data. We verify our approach on real-world measurements, achieving 2.1 dB RMSE with noisy input data and 1.3 dB with complete information, compared to 2.3-3.1 dB for classical methods such as ray-tracing and radial basis function interpolation. Additionally, we compare different ways of encoding environment information at varying levels of detail and we find that, in the considered single-room indoor scenarios, binary occupancy encoding performs at least as well as detailed material property information, simplifying practical deployment.