PROJECT TITLE :

Automatic Feature Selection Technique for Next Generation Self-Organizing Networks - 2018

ABSTRACT:

Despite self-organizing networks (SONs) pursue the automation of management tasks in current cellular networks, the selection of the foremost helpful performance indicators (PIs), used as inputs for SON functions, is still performed by network consultants. During this letter, a completely unique supervised technique for the automated selection of PIs for self-healing functions is proposed, hoping on the dissimilarity of their statistical behavior beneath different network states. Results using knowledge from a live network show that the proposed methodology outperforms an professional's selection, permitting the degree and complexity of both network databases and SON functions to be reduced without an professional's intervention.