Using Fuzzy Clustering for Analysis State of Network Elements of a Mobile Operator

Abstract

The article proposes a method for classifying state of network elements of a telecommunications operator. The method is based on use of the principles of fuzzy clustering of objects, which allows you to distribute many of analyzed objects into groups that have closest characteristics. Proximity of an object to a specific group is determined using value of the degree of belonging. For each group of objects, a scenario is formed - a list of measures necessary to restore performance of a given group of objects or to carry out preventive work to prevent emergency situations. Assignment of an object to a specific group is carried out based on the maximum value of the degree of belonging. The implementation of the proposed provisions allows you to automate the decision-making process to maintain the network of the telecom operator in working condition and reduce the time costs of operator’s personnel.