Spatial Signal Strength Prediction using 3D Maps and Deep Learning

Abstract

Machine learning (ML) and artificial neural networks (ANNs) have been successfully applied to simulating complex physics by learning physics models thanks to large data. Inspired by the successes of ANNs in physics modeling, we use deep neural networks (DNNs) to predict the radio signal strength field in an urban environment. Our algorithm relies on samples of signal strength collected across the prediction space and a 3D map of the environment, which enables it to predict the scattering of radio waves through the environment. While already extensive body of research exists in spatial signal strength prediction, our approach differs from most existing approaches in that it does not require the knowledge of the transmitter location, it does not require side channel information such as attenuation and shadowing parameters, and it is the first work, to the best of our knowledge, to use 3D maps to accomplish the task of signal strength prediction. This algorithm is developed for the purpose of placement optimization of a UAV or mobile robot to maximize the signal strength to or from a stationary transceiver but it also holds relevance to dynamic spectrum access networks, cellular coverage design, power control algorithms, etc.

A Unified Prediction Framework for Signal Maps

“…ML models for the signal strength likelihood for user localization have also been developed by prior art [22,31,38]. Krijestorac et al [28] propose state-of-the-art CNNs for estimating signal strength values and utilize a 3D map of the environment as features. In particular, they treat signal strength as a Gaussian random variable and predict its mean and variance.…”

“…We also did a straw man evaluation of a state-of-the-art CNN approach [28]. We usied a simplified implementation based on the source code kindly provided by the authors of [28], and tried two datasets: (i) the same synthetically generated data as in [28] and (ii) on own real-world UCI campus data.…”

“…We also did a straw man evaluation of a state-of-the-art CNN approach [28]. We usied a simplified implementation based on the source code kindly provided by the authors of [28], and tried two datasets: (i) the same synthetically generated data as in [28] and (ii) on own real-world UCI campus data. Specifically, we used the synthetic data to construct ground truth masks that matched the UCI campus masks and then trained the CNNs with the original masks and our realworld sparse masks.…”

RadioNet: Transformer based Radio Map Prediction Model For Dense Urban Environments

“…Ozyegen et al (2020) proposed to use Unet model with strided convolutions and inception modules for fast radio map prediction. Krijestorac et al (2020) employed Unet to interpolate radio map in urban environment using sparse signal samples collected across the prediction space and 3D map of the environment. Imai et al (2019) used CNN to extracted features for propagation loss prediction from spatial information.…”