Access Point Selection in Small Cell Ultra-Dense Network, With Load Balancing

Abstract

Ultra dense network of small cells (SCs) was introduced to support high data rate services and increase network capacity. Owing to user equipment (UE) mobility and limited coverage of SCs, the load across a SC network recurrently becomes unbalanced. Such unbalanced loads result in performance degradation of network in terms of throughput, inefficient utilization of network resources and link failure. This paper proposes and evaluates a novel load balancing methodology based on UE-SC association, which incorporated two main factors:(i)load of SCs in UE’s neighborhood (ii)the received Signal to Interference and Noise Ratio (SINR)/Channel Quality Indicator (CQI) and Quality of Service (QoS) requirements of user. The proposed algorithm ensures that the standard deviation of load in SCs in the neighborhood of a given UE remains minimum when the UE is associated to a potential SC. Consequently, achieving load balancing among SCs proactively while maintaining high SINR/CQI levels for UE, with each UE-SC pairing. System-level simulations show that the proposed algorithm provides a more balanced load across network, which is quantified by a smaller standard deviation of load across the SCs and higher network throughput when compared to other load balancing algorithms.