This paper presents an empirical large-scale propagation analysis at 18 GHz within an indoor corridor environment featuring a 90Formula: see text junction. The results quantify the contrast between line-of-sight (LoS) and non-line-of-sight (NLoS) conditions through five path loss models: close-in (CI), floating-intercept (FI), alpha-beta-gamma (ABG), dual-slope (DS), and two-slope dual-slope (2S-DS). Under LoS conditions, all models yield a consistent path loss exponent close to 2.29, indicating a propagation regime slightly above free-space decay due to local corridor obstructions. Under NLoS conditions, the CI model produces an exponent of 5.79; however, the DS model reveals that this high value results from the combination of two distinct mechanisms: a distance-dependent decay-comparable to the LoS regime-and a discrete corner diffraction loss of 41.07 dB at the 90Formula: see text junction. Allowing independent pre- and post-corner slopes (2S-DS) yields a post-corner exponent of 0.72, confirming a waveguide-like regime after the corner and attributing an additional 1.13 dB to the diffraction event itself. This is consistent with the FI model, which absorbs the corner loss into a high intercept of 123.31 dB and yields a low distance slope. Comparison with the 3GPP InH-Office model shows that the standardized reference underestimates NLoS attenuation while overestimating shadow fading in this geometry. The median excess path loss under NLoS conditions exceeds the LoS case by approximately 43 dB. The study shows that receiver height significantly influences large-scale attenuation statistics, with measurable differences in dispersion and percentile behavior across 0.61 m, 1.30 m, and 1.91 m configurations. Additionally, varying the transmitter distance to the corridor junction (39.4 m, 19.7 m, and 9.9 m) demonstrates that proximity to the corner reduces the mean NLoS path loss by up to 8 dB. These findings quantify the strong interaction between geometry, receiver height, and transmitter placement in FR3 indoor environments, providing practical insights for reliable coverage design in corridor-based layouts.
Robles et al. (Thu,) studied this question.