An Expedient, Generalizable and Realistic Data-Driven Indoor Propagation Model

Efficient and realistic indoor radio propagation modelling tools are inextricably intertwined with the design and operation of next generation wireless networks. Machine learning (ML)-based radio propagation models can be trained with simulated or real-world data to provide accurate estimates of the wireless channel characteristics in a computationally efficient way. However, most of the existing research on ML-based propagation models focuses on outdoor propagation modelling, while indoor data-driven propagation models remain site-specific with limited scalability. In this paper we present an efficient and credible ML-based radio propagation modelling framework for indoor environments. Specifically, we demonstrate how a convolutional encoder-decoder can be trained to replicate the results of a ray-tracer, by encoding physics-based information of an indoor environment, such as the permittivity of the walls, and decode it as the path-loss (PL) heatmap for an environment of interest. Our model is trained over multiple indoor geometries and frequency bands, and it can eventually predict the PL for unknown indoor geometries and frequency bands within a few milliseconds. Additionally, we illustrate how the concept of transfer learning can be leveraged to calibrate our model by adjusting its pre-estimate weights, allowing it to make predictions that are consistent with measurement data. <br>

An Expedient, Generalizable and Realistic Data-Driven Indoor Propagation Model

Efficient and realistic indoor radio propagation modelling tools are inextricably intertwined with the design and operation of next generation wireless networks. Machine learning (ML)-based radio propagation models can be trained with simulated or real-world data to provide accurate estimates of the wireless channel characteristics in a computationally efficient way. However, most of the existing research on ML-based propagation models focuses on outdoor propagation modelling, while indoor data-driven propagation models remain site-specific with limited scalability. In this paper we present an efficient and credible ML-based radio propagation modelling framework for indoor environments. Specifically, we demonstrate how a convolutional encoder-decoder can be trained to replicate the results of a ray-tracer, by encoding physics-based information of an indoor environment, such as the permittivity of the walls, and decode it as the path-loss (PL) heatmap for an environment of interest. Our model is trained over multiple indoor geometries and frequency bands, and it can eventually predict the PL for unknown indoor geometries and frequency bands within a few milliseconds. Additionally, we illustrate how the concept of transfer learning can be leveraged to calibrate our model by adjusting its pre-estimate weights, allowing it to make predictions that are consistent with measurement data. <br>

Geometric Simplifications of Natural Caves in Ray-Tracing-Based Propagation Modelling

Natural caves show some similarities to human-made tunnels, which have previously been the subject of radio-frequency propagation modelling using deterministic ray-tracing techniques. Since natural caves are non-uniform because of their inherent concavity and irregular limestone formations, detailed 3D models contain a large number of small facets, which can have a detrimental impact on the ray-tracing computational complexity as well as on the modelling accuracy. Here, we analyse the performance of ray tracing in repeatedly simplified 3D descriptions of two caves in the UK, i.e., Kingsdale Master Cave (KMC) Roof Tunnel and Skirwith Cave. The trade-off between the size of the reflection surface and the modelling accuracy is examined. Further, by reducing the number of facets, simulation time can be reduced significantly. Two simplification methods from computer graphics were applied: Vertex Clustering and Quadric Edge Collapse. We compare the ray-tracing results to the experimental measurements and to the channel modelling based on the modal theory. We show Edge Collapse to be better suited for the task than Vertex Clustering, with larger simplifications being possible before the passage becomes entirely blocked. The use of model simplification is predominantly justified by the computational time gains, with the acceptable simplified geometries roughly halving the execution time given the laser scanning resolution of 10 cm.