Load Balancing User Association in Millimeter Wave MIMO Networks

TL;DR

This paper proposes a load balancing user association scheme for mmWave MIMO networks that improves network performance by efficiently balancing base station loads and interference, using a polynomial-time algorithm.

Contribution

Introduces a novel load balancing user association scheme considering interference and rate dependencies, with a polynomial-time algorithm for near-optimal solutions.

Findings

Significant network performance improvement with the new scheme.

WCS algorithm outperforms genetic algorithms in accuracy and speed.

WCS reaches optimal solutions in small networks.

Abstract

User association is necessary in dense millimeter wave (mmWave) networks to determine which base station a user connects to in order to balance base station loads and maximize throughput. Given that mmWave connections are highly directional and vulnerable to small channel variations, user association changes these connections and hence significantly affects the user's instantaneous rate as well as network interference. In this paper, we introduce a new load balancing user association scheme for mmWave MIMO networks which considers this dependency on user association of user's transmission rates and network interference. We formulate the user association problem as mixed integer nonlinear programming and design a polynomial-time algorithm, called Worst Connection Swapping (WCS), to find a near-optimal solution. Simulation results confirm that the proposed user association scheme improves network performance significantly by moving the traffic of congested base stations to lightly-loaded ones and adjusting the interference accordingly. Further, the proposed WCS algorithm outperforms other generic algorithms for combinatorial programming such as the genetic algorithm in both accuracy and speed at several orders of magnitude faster, and for small networks where exhaustive search is possible it reaches the optimal solution.