Ray-Tracing Simulated Dataset in the KAUST Campus for the sub-6 GHz Band

The files here support the research work presented in the paper submitted for IEEE Transactions on Vehicular Technology, \Machine Learning-based Path Loss Prediction inSuburban Environment in the Sub-6 GHz Band\, which is currently under revision. This paper investigates the effectiveness of Machine Learning (ML) models in predicting path loss (PL), particularly for the sub-6 GHz band in a suburban campus of King Abdullah University of Science and Technology (KAUST). For training purposes, synthetic datasets using the ray-tracing simulation technique have been generated. The feasibility and accuracy of the ML-based PL models are verified and validated using both synthetic andmeasurement datasets. The random forest regression (RFR) and the k-nearest neighbors (KNN) algorithms provide the best PL prediction accuracy compared to other ML models. In addition, we compare the performance of the developed ML-based PL models with the traditional propagation models including COST-231 Hata, Longley-Rice, and Close-In models. The results show the superiority of the ML-based PL models compared to conventional models. Several simulations were conducted, including different transmitter (Tx) locations, transmitter power, and various frequencies in the sub-6 GHz band, namely 1.5 GHz, 2.3 GHz, 2.5 GHz, 3.5 GHz, and 6 GHz. This dataset contains, not only simulated path loss, but also useful features including the latitude\/longitude of the receiving point, the distance between the transmitter and the receiver, the azimuth\/elevation angles, the LoS status, the received power, as well as the simulation parameters: the transmitter location, height, power, and the frequency. This dataset can be used for further research works and benchmarking with other radiowave propagation models.