Ranking of lte cells based on key performance indicators using mcdm methods

Abstract

The growth in worldwide data traffic and user subscriptions in mobile telecommunication networks makes it increasingly difficult to manage network performance in an environment already containing multiple radio access technologies. Despite the rise of 5G, LTE remains the dominant technology, and new cells are installed daily to support traffic growth and new services such as voice over LTE. Detecting faulty cells in the network is one of the main concerns of operators. Self-organizing networks have been introduced to deal with this problem, and their self-healing function- ality has improved cell fault management. Nonetheless, faulty cell detec- tion remains challenging, and most of the tasks involved are still done manually. This paper introduces a new method of faulty cell detection in an LTE radio access network, applying multiple criteria methods to this problem. The cells are ranked based on selected key performance indica- tors, using the multi-attribute utility theory to construct a utility function. The analytic hierarchy process is used to define weights for the criteria. Keywords: long-term evolution, multiple criteria methods, radio access network perfor- mance management, self-organizing networks.