• Cell identity allocation and optimisation of handover parameters in self-organised LTE femtocell networks

      Zhang, Xu (University of BedfordshireUniversity of Bedfordshire, 2013-09-16)
      Femtocell is a small cellular base station used by operators to extend indoor service coverage and enhance overall network performance. In Long Term Evolution (LTE), femtocell works under macrocell coverage and combines with the macrocell to constitute the two-tier network. Compared to the traditional single-tier network, the two-tier scenario creates many new challenges, which lead to the 3rd Generation Partnership Project (3GPP) implementing an automation technology called Self-Organising Network (SON) in order to achieve lower cost and enhanced network performance. This thesis focuses on the inbound and outbound handovers (handover between femtocell and macrocell); in detail, it provides suitable solutions for the intensity of femtocell handover prediction, Physical Cell Identity (PCI) allocation and handover triggering parameter optimisation. Moreover, those solutions are implemented in the structure of SON. In order to e ciently manage radio resource allocation, this research investigates the conventional UE-based prediction model and proposes a cell-based prediction model to predict the intensity of a femtocell's handover, which overcomes the drawbacks of the conventional models in the two-tier scenario. Then, the predictor is used in the proposed dynamic group PCI allocation approach in order to solve the problem of PCI allocation for the femtocells. In addition, based on SON, this approach is implemented in the structure of a centralised Automated Con guration of Physical Cell Identity (ACPCI). It overcomes the drawbacks of the conventional method by reducing inbound handover failure of Cell Global Identity (CGI). This thesis also tackles optimisation of the handover triggering parameters to minimise handover failure. A dynamic hysteresis-adjusting approach for each User Equipment (UE) is proposed, using received average Reference Signal-Signal to Interference plus Noise Ratio (RS-SINR) of the UE as a criterion. Furthermore, based on SON, this approach is implemented in the structure of hybrid Mobility Robustness Optimisation (MRO). It is able to off er the unique optimised hysteresis value to the individual UE in the network. In order to evaluate the performance of the proposed approach against existing methods, a System Level Simulation (SLS) tool, provided by the Centre for Wireless Network Design (CWiND) research group, is utilised, which models the structure of two-tier communication of LTE femtocell-based networks.