Supervised vector machine learning with brown boost energy efficient data delivery in MANET

Abstract

A mobile ad hoc network (MANET) is an auto-configuring, infrastructure-free network in which nodes move independently in any direction. Owing to the fact that a MANET necessitates minimum cost for the design of networks, infrastructure construction, and management, fault tolerance by possessing a distributed control, its popularity has risen in the recent few years. However, it is a laborious and cumbersome process of MANET for ensuring consistent communication between nodes because of unstable network topology. MANET necessitates a genuine data delivery method that redesigns the variable network topology for ensuring consistent communication. Several methods were proposed in the literature, however, causing a significant time and energy during data delivery. To address these issues, we plan to develop a data delivery method called, Supervised Vector Machine with BrownBoost Classification (SVM-BBC) for efficient data delivery with minimal time consumption and higher data delivery rate. In the SVM-BBC method, distinct numbers of mobile nodes are considered as training data samples. Moreover, Support Vector Machine (SVM) classifiers are considered weak classifiers. Here Supervised SVM is employed for classifying neighboring nodes with better link quality and lesser energy consumption. Next, weak learners outputs are aggregated to obtain strong classifier output results, therefore ensuring efficient data delivery. Here, aggregation is carried out by means of BrownBoost Node Classification to produce a higher data delivery rate during data delivery. Experimental evaluation is performed on various factors, namely energy consumption, data delivery rate,data delivery time, data drop rate, and throughput with varying numbers of mobile nodes.