الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 68 |  |  |
| | | from | 68 | | |
Abstract
The deployment of 4 th Generation (4G) or Long-Term Evolution (LTE) Mobile Networks, while old news to developed countries, is ongoing in developing countries with some starting LTE deployments only a year ago. LTE is essential where broadband access is mostly available through wireless connection. The process of LTE deployment is costly and time consuming. Network operators attempt to minimize both overheads. With the current challenges facing network operators to acquire new sites, 4G deployment always depends on reusing the already existing 2G/3G macro layer. We study the approach of rapid 4G deployment by analyzing data from a major network operator in the Middle East. In particular, we look at network measurements from a single large cluster right after LTE deployment. We analyze the impact of the deployment >approach on user throughput and find that LTE throughput could be improved further for the majority of LTE users in more than 50% of the studied cells by improving either interference or coverage. We find that this is mainly because the LTE system is more sensitive to interference when compared to 3G. In this thesis, we propose a new approach for mobile network performance optimization to mitigate the pitfalls of new LTE network deployments using machine learning. We present a data-driven methodology to detect the affected radio cells and mitigate the issue through physical optimization. We showcase some practical cases that had obvious performance enhancement following our proposed approach.