Mobility management for heterogeneous long term evolution mobile networks based on drive test measurement

Abstract

Mobility for high-speed users has been a significant challenge for all mobile networks such as long-term evolution (LTE). Heterogeneous Network (HetNet) comprises numerous cells of varying sizes, various carrier frequency band and cell coverage. These differences make mobility difficult as its procedure time will increase due to a different level of required measurements. User equipment (UE) with a high speed of up to 140 km/h causes a high risk of disconnection to the handover (HO) procedure timing which is longer than UE with realistic speed. Several field measurements were conducted using the drive test (DT) tool to obtain real measurements of the LTE network deployed with different cell sizes and different operating frequency bands across highways. During the field measurements, there are frequent HOs and HO failures occurred, especially when UE with increasing speed up to 140 km/h in different HO environments in terms of peak hours or, site locations and LTE cell configuration. The handover preparation stage is affected by the UE speed increase. Moreover, the analyzed results show that HO preparation time increases as UE speed increases and HO execution time is slightly faster than the HO preparation time. The overall HO time was found to be less than 1 s for high-speed UE, indicating that the network environment influences the HO timing as compared to other studies conducted in a similar research area. In this thesis, UE speed is added as one of the main criteria to determine whether the network has to proceed HO with a routine UE HO or enforce UE to perform HO due to high speed which combined with regular HO parameters. Therefore, the matrix is defined with UE speed and reference signal received power (RSRP) as main conditions for the forcing handover knowing that RSRP is mainly used by mobile network operators (MNO). The algorithm calculates the HO waiting time once the speed and RSRP exceed the defined thresholds. The overall HO procedure is improved by 11% by using the forcing HO algorithm.