Comparative study of machine learning technics for mobile ad hoc networks

The routing optimization in MANET is a topical issue. Nodes in MANET are mobile, and therefore the optimal routing protocol should first of all be able to update the list of routes in use in real time. In the last decade, machine learning has become increasingly popular in solving various challenges in MANET. One of the areas of using machine learning in MANET is the improving of existing routing protocols or developing new ones. The purpose of this work is to study machine learning techniques and their application possibilities in routing protocols for MANETs. The paper also discusses several examples of routing protocols that are based on machine learning methods developed over the past 2 years and provides a comparative analysis of popular supervised machine learning methods that are used for classification in various fields, including node classification for optimization of routing processes in MANET.