Multi-Cell Flow-Level Performance of Traffic-Adaptive Beamforming under Realistic Spatial Traffic Conditions

Upcoming 5G wireless technology will enable various new applications and is going to support a massive amount of devices per cell. In this regard, ultra-dense small cell deployments are a means to cope with the need of extremely high data rates of several tens of Gbps and with the increasing data traffic demand. However, denser deployments may lead to more severe inter-cell interference, which is strongly connected to the actual spatial distribution of the mobile traffic demand. Traffic-adaptive beamforming using phased antenna arrays can be an attractive solution for concentrating capacity at desired traffic hot spot locations while inter-cell interference is reduced. In this paper, we propose a flexible and holistic model, which describes flow-level performance of networks, which consist of base stations equipped with phased antenna arrays, accurately and considers dynamic inter-cell interference. Moreover, we present a configurable spatial traffic model to generate data traffic maps with various statistical properties. We use these traffic maps to evaluate the performance of a traffic-adaptive beamforming algorithm proposed and compare it the performance of a stateof- the-art antenna down-tilt algorithm.