Clustering Based Interference Analysis of GSM-R Network on Drive Test Data

The traditional method of analyzing and processing the drive test (DT) data to find the interferences in the Global System for Mobile Communications-Railway (GSM-R) network is often a manual method, which is not only inefficient, but also difficult to find the correlation and regularity among the DT data. In order to improve the efficiency of capturing interferences in GSM-R network and to discover the regularity of the DT data, we propose a method to capture and analyze the interferences in GSM - R network based on DT data clustering. Firstly, we design and implement a software system of DT terminal to acquire DT data sets. Then, we use the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering algorithm to analyze the DT data and initially determine the location of the interferences in GSM-R network. Finally, according to the analyzed results, we conduct an interference check and successfully find the interference source. The method provides a reference for the future research of drive test data.