Huang, YifeiDurrani, SalmanZhou, Xiangyun2020-01-21June 4-7 29781509059324http://hdl.handle.net/1885/198783Increasing densification in future wireless networks means that user association will play an ever more critical role in the network decision process in order to manage the large number of base stations and users. Though conventional user association aims to maximize a sum rate or capacity related objective, user rate fairness could become a more important consideration for dense networks. In this paper, we propose a downlink base station preference association scheme where users connect to the base station where it is most preferred in terms of the maximum received power. We prove analytically that this scheme results in roughly the same number of users associated to each base station regardless of base station transmit power, and will result in high user rate fairness in dense networks. In addition, we study how the associations change with network dynamics, i.e., users entering and exiting the network (e.g., due to users crossing boundaries of small cells) or base stations entering and exiting the network (e.g., due to base station switching ON or OFF to reduce energy consumption). Our results show that there exists a type of user most likely to re-associate, and that a shrinking network leads to more re-association than a growing one.5 pagesapplication/pdfen-AU© 2017 IEEEUser associationheterogeneous networksnetwork dynamics201710.1109/VTCSpring.2017.81084732019-11-25