Simplified expression of the average rate of cellular networks using stochastic geometry

Accurate modeling of network interference, deep understanding of its impact on the achievable performance, and development of efficient techniques to mitigate or exploit it are three important and fundamental research assets in current and next-generation cellular networks. In this context, Andrews, Baccelli, and Ganti [1] have recently introduced a new analytical approach to estimate coverage and rate of cellular networks subject to other-cell interference. In this paper, we move from the approach developed in [1], and propose an alternative analytical derivation to compute the rate of cellular networks. More specifically, by using stochastic geometry and Poisson point processes theory, we derive a simple and easy-to-compute expression of the rate, which can be used for arbitrary network and channel parameters, e.g., path-loss exponent, receiver noise, density of Base Stations (BSs), etc. Compared to [1], our framework has two main distinguishable features: i) the rate can be computed via a single numerical integral, rather than via a three-fold numerical integral; and ii) the formula is applicable to arbitrary fading distributions on the intended link, rather than being useful for Rayleigh fading only. Our analytical derivation is substantiated through extensive Monte Carlo simulations.