Data Analysis for Hearability and Synchronization in Telecom Systems

Sammanfattning

In a wireless communication system, synchronization is one of the most critical functions. Good timing and synchronization can prevent messages of the communication system from interfering with each other and enable a smooth message transfer. For 5G (the fifth generation of telecommunication network), the timing and synchronization requirements are even more strict. To synchronize the communication system, the first thing is to detect the connection quality and find the faults’ positions. This thesis aims to investigate a machine learning or deep learning approach to predict the connection quality called ”Hearability” between cells and to find the base stations whose cells have an absolute time error larger than a configurable threshold. The results show that the random forest classifier can obtain the highest accuracy while requiring less computing time than k-Nearest-Neighbors, Support Vector Machine, and Graph Neural Network. Additionally, we found the base stations with faults and offered suggestions for detecting the faults in complicated cases.