Stochastic Modeling of Interference in 5G mmWave Cellular Networks

Abstract

A simpler statistical interference model for millimeter-wave (mmWave) cellular networks for use in 5G is proposed in this work. The accuracy of this model is examined in an outdoor cellular network with different scenarios. The interference accuracy coefficient (IAC) is introduced to quantify the accuracy of predicting an outage event and the similarity between various interference models. In this paper, different signal-to-interference-plus-noise-ratio (SINR) distributions are used and a new framework has been proposed to quantify the similarity between SINR distributions of existing interference models. It is observed that the accuracy of millimeter-wave cellular networks is modeled by simpler interference models which are less erroneous compared to the models used for performance analysis in microwave networks. The results obtained in this work clearly show that in the mmWave networks scenarios with obstacles, the simple interference models are accurate enough. However, in omnidirectional communications, the interference models required are quite complex. Furthermore, in this work, the interference in mmWave networks with a deterministic channel is also modeled using the proposed Two-Ball interference model. This model gives the appropriate results of blockage events due to outages in mmWave networks.