Impact of LOS/NLOS Propagation and Path Loss in Ultra-Dense Cellular\n Networks

Most prior work on performance analysis of ultra-dense cellular networks\n(UDNs) has considered standard power-law path loss models and non-line-of-sight\n(NLOS) propagation modeled by Rayleigh fading. The effect of line-of-sight\n(LOS) on coverage and throughput and its implication on network densification\nare still not fully understood. In this paper, we investigate the performance\nof UDNs when the signal propagation includes both LOS and NLOS components.\nUsing a stochastic geometry based cellular network model, we derive expressions\nfor the coverage probability, as well as tight approximations and upper bounds\nfor both closest and strongest base station (BS) association. Our results show\nthat under standard singular path loss model, LOS propagation increases the\ncoverage, especially with nearest BS association. On the contrary, using dual\nslope path loss, LOS propagation is beneficial with closest BS association and\ndetrimental for strongest BS association.\n