An Analytical Model for QoS in Mobile Ad Hoc Network using Artificial Neural Network (ANN)

This study examines the challenge of predicting end-to-end packet delay in mobile ad hoc networks for Quality of Service (QoS) routing. Factors like node neighbors, interference, and medium access control impact delay unpredictability. We explore the potential of artificial neural networks to enhance accuracy in forecasting these delays, aiming to improve QoS within the dynamic Mobile Ad Hoc Network (MANET) context. We calculated delay using sourcedestination path length and average neighbors. Two models were built: Regression Neural Networks (RNN) and Radial Basis Function (RBF) networks. Using a network simulator, we gathered data sets for three routing protocols: AODV, DSDV, and DSR. These sets include delay, path length, and average neighbors. RNN outperformed RBF in predicting delay, as determined by various performance metrics