Advanced performance monitoring for self-healing cellular mobile networks

Abstract

This dissertation is devoted to development and validation of advanced per- formance monitoring system for existing and future cellular mobile networks. Knowledge mining techniques are employed for analysis of user specific logs, collected with Minimization of Drive Tests (MDT) functionality. Ever increas- ing quality requirements, expansion of the mobile networks and their extend- ing heterogeneity, call for effective automatic means of performance monitoring. Nowadays, network operation is mostly controlled manually through aggregated key performance indicators and statistical profiles. These methods are are not able to fully address the dynamism and complexity of modern mobile networks. Self-organizing networks introduce automation to the most important network functions, but the opportunity of processing large arrays of user reported perfor- mance data is underutilized. Advanced performance monitoring system developed in the presented re- search considers both numerical and sequential properties of the MDT data for detection of faults. Network malfunctions analyzed in this study are sleeping cells in either physical or medium access layer. A full data mining cycle is em- ployed for identification of problematic regions in the network. Pre-processing with statistical normalization and sliding window methods, both linear and non- linear transformation and dimensionality reduction algorithms, together with clustering and classification methods are used in the discussed research. Sev- eral post-processing and detection quality evaluation methods are proposed and applied. The developed system is capable of fast and accurate detection of non- trivial network dysfunctions and is suitable for future mobile networks, even in combination with cognitive self-healing. As a result, operation of modern mo- bile networks would become more robust, increasing quality of service and user experience.