Alarm and service monitoring of large scale multi-service mobile networks

Sammanfattning: Two of the most important challenges in network service assurance are ' an overwhelming flow of low-quality alarms  ' to understand the structure and quality of the delivered services This thesis proposes solutions for alarm and service monitoring that addresses monitoring of large scale multi-service mobile networks.  The work on alarms are based on statistical analysis of data collected from a real-world alarm flow and an associated trouble ticket database containing the network administrators expert knowledge. Using data from the trouble ticketing system as a reference, we examine the relationship between the original alarm severity and the human perception of the alarm priority. Using this knowledge, we suggest a neural network-based approach for alarm prioritization. Tests using live data show that our prototype assigns the same severity as a human expert in 50% of all cases, compared to 17% for a na¨ıve approach. In order to model and monitor the services this thesis propose a novel domain specific language called SALmon, which allows for efficient representation of service models, together with a computational engine for evaluation of service models. We show that the proposed system is a good match against real-world scenarios with special requirements around service modeling.