A site-specific channel large-scale parameter model with ray-based validation

In this article, a novel framework of large-scale parameter (LSP) predictions considering environmental factors is proposed, which covers path loss, the Ricean K-factor, root mean square (RMS) delay spread (DS), and angular spread (AS). In the proposed model, the considered scenario is classified into different region types based on a two-dimensional (2D) raster map. Each region type is then labeled with a large-scale parameter exponent (LSPE) matrix. To construct the model, a straight line is generated between the transmitter (Tx) and the receiver (Rx) to record the intersected region LSPE matrices and the corresponding region lengths. By computing the weighted LSP of each intersected region based on the straight-line information, the LSP for the link is calculated. Following that, we use Ranplan Professional, a measurement-verified commercial ray-based simulator, to simulate two outdoor scenarios and two indoor scenarios for extracting and validating the proposed LSP model. We then compare it with the Third Generation Partnership Project (3GPP) model and the Fifth Generation Channel Model Special Interest Group (5GCMSIG) model in both scenarios, and the results show that the proposed model outperforms these two channel models while maintaining similar model complexity.