Performance Analysis of Intelligent Reflecting Surface-Aided Vehicular Communications under Dynamic Blockage

Abstract

Intelligent reflecting surfaces (IRS) promise coverage enhancement for vehicular networks, but optimal deployment strategies remain unclear.This paper analyzes where and when an IRS provides the greatest value under dynamic line-of-sight (LOS) and non-line-of-sight (NLOS) blockage transitions. Using a blockage model motivated by published mmWave vehicular measurements (three spatial NLOS zones with 15 dB direct-path attenuation) and Monte Carlo simulations, we quantify regime-dependent IRS gains, CSI requirements, and scaling with blockage severity. The key findings are: (i) in NLOS the IRS yields about a 10.8-fold rate gain, compared with about a 1.2-fold gain in LOS (roughly a ninefold difference in gain); (ii) CSI requirements in NLOS can be relaxed by approximately 13-27% while preserving most of the IRS benefit; (iii) IRS-assisted NLOS performance remains stable across a wide range of scattering conditions (Rician K between 0.5 and 10); and (iv) for a 256-element IRS the gain increases superlinearly with blockage severity and exceeds a twofold improvement once the blockage level is around 20 dB. These results suggest deployment guidelines that prioritize IRS placement at severely blocked locations (e.g., urban canyons and intersections), where the IRS impact is most pronounced.