A Joint Optimization Framework for Network Deployment and Adaptive User Assignment in Indoor Millimeter Wave Networks

Abstract

Millimeter wave (mmW) systems typically use beamforming techniques to compensate for the high pathloss. However, directional communications in the presence of uncertainty in user equipment (UE) locations and channel conditions make maintaining coverage and connectivity challenging. In this context, we propose a joint optimization framework to determine the minimum number of required access points (APs), their optimal locations, their optimal beam directions, and their optimal assignments to individual UEs in order to maintain a network-wide signal-to-noise ratio (SNR) coverage and stable connections. The network deployment decisions (i.e., the required number of APs, their placements, and their beam directions) are static and are taken before UE locations and channel conditions are revealed. The UE assignment decisions are taken under each realization of UE locations and channel conditions considering the availability and stability of the mmW beams. We develop our joint optimization framework following a two-stage chance-constrained stochastic optimization model. Our numerical results demonstrate the gains brought by our proposed framework in terms of reducing cost of network deployment while ensuring a network-wide SNR coverage and stable connections under various UE distributions and system parameters.