Modeling Non-Uniform UE Distributions in Downlink Cellular Networks

A recent way to model and analyze downlink cellular networks is by using random spatial models. Assuming user equipment (UE) distribution to be uniform, the analysis is performed at a typical UE located at the origin. At least one shortcoming of this approach is its inability to model non-uniform UE distributions, especially when there is dependence in the UE and the base station (BS) locations. To facilitate analysis in such cases, we propose a new tractable method of sampling UEs by conditionally thinning the BS point process and show that the resulting framework can be used as a tractable generative model to study current capacity-centric deployments, where the UEs are more likely to lie closer to the BSs.