Improvıng the Network Throughout ın IOT-Manets usıng Deep Learnıng Routıng Algorıthm

Abstract

The Internet of Things (IoT) based mobile ad‐hoc network (MANET) offer the mobile user with greater flexibility and lower network deployment costs and enjoy a power balance problem among sensor nodes, because IoT mainly operates on fixed devices or sensors, while MANETs are operated on mobile sensor nodes. The choice of efficient and short paths by the MANET protocol thus becomes an increasingly difficult task, or it may lose focus on the shortest path selection. The correct use of battery power is therefore necessary for maintaining the network connectivity in multi-hop transmission. In this paper, we utilize Deep Neural Network (DNN) routing on IoT-MANETs. The IoT nodes helps in collection and acquisition of data, and MANETs are responsible for data routing and transmitting effectively the packets between the source and sink nodes. The simulation results are estimated in terms of average delay, throughput and network energy efficiency. The results of proposed DNN achieves higher network throughout than the existing machine learning algorithm