Improved Load Balancing for LTE-A Heterogeneous Networks using Particle Swarm Optimization

Abstract

Heterogeneous networks (HetNets) are a promising means of meeting the requirements of Long Term Evolution-Advanced (LTE-A) in terms of data traffic, coverage and capacity. In HetNets, power disparities arise between base stations in different tiers. The use of existing user association schemes will lead to load imbalances between these base stations, thus affecting network performance. Biased user association has been widely studied to improve load balancing in HetNets. Static biasing has been the focus of most existing work but this approach does not yield optimized performance because the optimal biasing values vary with user location. In this paper, we investigate the use of the Particle Swarm Optimization (PSO) algorithm to conduct dynamic user association by finding the optimal bias values. The simulation results demonstrate that the proposed scheme achieves better load balancing performance in terms of the network balance index compared to a baseline scheme.