Optimization of Soft Handover for Addis Ababa UMTS Radio Access Network

Handover is an essential radio management technique of mobile communication to enable seamless communication while users move from one cell to another one. To mitigate impacts of interference on cell-edge users, third Generation (3G) supports soft handover where users can be simultaneously served by two or more cells, in addition to hard handover, where a user is served by only one cell. The number of soft handover users and their active cells determine the degree of soft handover overhead. The soft handover overhead data for the year 2018 of Addis Ababa 3G network is 87% average overhead, which is considerably beyond the maximum recommended value. This figure indicates the requirement to optimize the soft handover overhead using the soft handover related network parameters' configurations. In this thesis, soft handover overhead performance of the Addis Ababa 3G network is analyzed. The spatial distribution of the overhead is visualized using data collected from the network management system. Furthermore, optimization of soft handover overhead is done through fine-tuning window add, pilot power, and electrical antenna down tilt. We apply a heuristic algorithm for the optimization considering potential sets of values for the optimization variables. Moreover, obtained results are compared with a result obtained using existing default network configuration parameters. For the sample 3G sites network simulation, we use WinProp while we also use MATLAB and Google Earth for soft handover performance analysis and visualization. Soft handover overhead performance analysis of 5 Radio Network controllers (RNCs) shows that the Addis Ababa network is within the range of 111 to 115% for the 90 th percentile. Sites in RNC4 using carrier three are selected for the simulation and optimization as 5 to 21 more number of radio links used comparing with the other three carriers. After optimization, results show that soft handover overhead and network capacity are considerably improved by fine-tuning the window add, pilot power, and antenna tilt parameters. For instance, soft handover overhead is reduced from 85.4 % to 57.7% and network capacity gain by 3.76% when using window add of 2 dB, pilot power of 7.5%, and electrical antenna down tilt by 2 degrees. Soft handover overhead is reduced from 85.4 % to 46.6% and network capacity gain by 5.72% when using window add of 1.5 dB, pilot power of 5% and electrical antenna down tilt by 3 degrees.