Estimation of intra-site inter-sector coverage overlapping using user data for LTE network

Abstract

The increase in user demand has created challenges for network capacity implementation. Due to the limited available LTE spectrum, capacity upgrades can take different directions such as adding new sites, new sectors, or using multi-beam antennas for cell splitting instead of adding new frequencies. However, the introduction of higher order sectorization has led to coverage overlaps and increased interference towards adjacent sectors, which limits the performance of Long term Evolution (LTE) networks and degrades spectral efficiency. Although resource scheduling methods have been introduced to cancel inter-cell interference, they limit resource utilization and impact network capacity. Therefore, network operators need to minimize coverage overlap to an acceptable level to mitigate the negative impact on spectral efficiency. Physical optimization has played a key role in achieving this, but traditional methods of identifying inter-cell overlapping have been inefficient due to the lack of geographical view and the time taken to locate the actual issue. The introduction of minimization of drive test (MDT) has moved optimization activities to a data-driven perspective. This thesis proposes a method of estimating intra-site inter-sector overlapping using MDT data and introduces three metrics to identify the granularity of the overlapping and its impact on Signal to interference and noise ratio (SINR) . These metrics can be used to prioritize the worst-affected sites and required azimuth changes to avoid overlapping. The proposed approach can overcome limitations such as false detection and time taken for identification in current methodologies. It accurately identifies issues with over 95% confidence, where spectrum efficiency can be improved, resulting in savings on network operating expenses.