Semantic-Based Service Analysis and Optimization

In management systems, the focus is moving from managing the Quality of Service (QoS) of end-user service sessions to managing their Quality of Experience (QOE). To do this, the user experience and context of those sessions much be modelled, analysed, and optimised to ensure end user experience matches expectations as closely as possible. In this thesis, we introduce the Aesop approach, which uses semantic-based techniques to autonomically optimize end-user service delivery. The Aesop knowledge base models the end-user service management domain in a manner that is aware of the temporal properties of concepts. The autonomic Aesop Engine runs efficient semantic algorithms that implement the MAPE functions using those temporal properties to operate on small partitioned subsets of the knowledge base. The effectiveness and performance of the Aesop approach was evaluated on a HAN test bed. A significant improvement was observed on the compliance levels of high priority sessions in all experimental scenarios, with compliance levels more than doubled in some cases. A case study demonstrated that Aesop was also applicable in the Mobile Broadband Access domain.