IEEE Access (Jan 2018)

Load-Dependent Handover Margin for Throughput Enhancement and Load Balancing in HetNets

  • Mohanad Alhabo,
  • Li Zhang

DOI
https://doi.org/10.1109/ACCESS.2018.2878489
Journal volume & issue
Vol. 6
pp. 67718 – 67731

Abstract

Read online

Inbound handover (HO) or hand-in is done when the user equipment (UE) performs HO from a macrocell (MC) to a small cell (SC), while the outbound HO or handout is done when a UE hands over from SC to MC. On the other hand, the inter-SC HO is done when a UE performs HO between two SCs. The outbound HO is not as complex as the other two types, because the UE has only one HO target base station, i.e., the MC. Therefore, in this paper, we only consider the inbound and inter-SC HO types. The user may associate with a small cell for a very short time of stay (ToS), smaller than a short time of stay threshold, and this may cause frequently unnecessary HOs and result in service interruption causing degradation in the quality of service (QoS). In this paper, we propose a novel HO method for the purpose of load balancing and throughput improvement in heterogeneous networks (HetNets). The influence of interference from both MC and SC base stations is taken into account so as to offloaded the user from the congested cell and forced it to HO to the SC that gives a good data rate by choosing the best SC, which has the highest signal to interference plus noise ratio (SINR), from a reduced neighbor cell list (NCL). The NCL is optimized utilizing the SINR threshold and ToS. The proposed method utilizes a modified A3 HO initiation event taking into account the cell load and the interference. Results show that the proposed method can perform HO while maintaining the throughput to a good level. In addition, the proposed method has significantly reduced the inter-SC HOs and inbound HO and radio link failures compared to the existing methods. Under different network conditions, load factors, and call arrival rates, results show that the proposed method can give significantly better performance, thereby producing higher throughput for the user and the network.

Keywords