Modeling and Analysis of User-Centric Collaborative Millimeter-Wave Cellular Networks

Abstract

This paper proposes a comprehensive framework to analyze the performance of the user-centric collaborative millimeter-Wave (mmWave) cellular networks, where multiple base stations (BSs) jointly transmit to form a cooperative cell. We model the locations of non-line-of-sight (NLOS) BSs and geographical centers as two independent Poisson point processes (PPPs). To capture the location correlation, users and line-of-sight (LOS) BSs are clustered around geographical centers. Applying the distributions of the distance sequences between the typical user and the interfering BSs, accurate formulas for the Laplace transform of interference are obtained for the case of Thomas cluster process (TCP) and Matern cluster process (MCP). Furthermore, we compute the signal to interference ratio (SIR) coverage probability and ergodic capacity under these two cluster processes, respectively. Our analysis and simulations show that the proposed user-centric collaborative mmWave network provides higher coverage probabilities and capacity gains compared with non-cooperation.