End to end delay using Aodv-artificial neural networks (Ann) to improve performance of manets

Abstract

MANETs consist of the nodes that move continuously in the random directions. The frequent topology changes leads to broken links which increases the delay in sending the data to the end node. In the traditional direction-finding protocols such as AODV, DSR or DSDV, do not focus on reducing the back-to-back stoppage of the transmitted packets and remaining energy of the network. In healthcare applications where the urgency of the data is on the highest priority, the routing protocol that can reduce the back-to-back delay is required. To get better quality of service (Qos). We have implemented the AODV-ANN predicts the delay dependencies based on distance between two nodes, relative mobility and congestion index and it helps in choosing the energy efficient and delay aware optimized path to send data from starting place to end node. The presentation of the system has been computed based in back-to-back stoppage, remaining energy of the system. The proposed AODV-ANN predicts the enhancement of the (quality of service) networks. The AODV-ANN simulation is better results and increase the network lifetime.