The number of mobile subscribers has increased in recent years. The first billion milestone was exceeded in 2002 and the fifth billion was reached in 2010. In the mean time the average data consumption exceeded the hundreds of Megabytes per user per month. With the evolution of wireline networks, the users expect similar perormance to the fixed lines. To meet these targets 3GPP introduced the long term evolution (LTE) network in 2004. The operators to keep up with the increasing number of subscribers, have to deploy new LTE networks and have to optimize their existing LTE networks. To reduce the operational efforts of LTE networks, 3GPP has worked on the self optimization (SON) aspects of LTE.
In this thesis I present a system level simulation results of self-optimzing load balancing algorithm in LTE network. I evaluated the performance and impact of this algorithm on end-user experience. I compared the results for different traffic mixes, using real-time streaming applications to determine the improvement of QoS (quality of service) parameters and the percevied video quality (MOS).
The evaluation of the simulations had shown that there are possibilites to improve the performance of the load balancing algorithm so I introduced a new load calculation method. With this load calculation the performance of load balancing had greatly increased.