Simplifying Network Management with Fuzzy Logic

Abstract

This paper advocates the use of machine learning for complex network management tasks. Specifically, this paper describes a novel idea of using fuzzy logic for effective network management, as the use of imprecise and linguistic information for intelligent decision making, can reduce the overhead of extensive network measurements for management purpose. It also describes the distinguishing features of network management that makes fuzzy logic suitable to be used in this field. Among a huge number of possible applications of this idea in broad area of network management, this paper briefly discusses few options that we are planning to explore further. In addition, it discusses one such option in detail.