RAD: reinforcement authentication model based on DYMO protocol for MANET
Rushdi A. Hamamreh
Mohammed R. Ayyad
Inderscience Enterprises Ltd.
Abstract: This article aimed to develop a new model based on DYMO protocol where a modification was proposed to route discovery and route maintenance processes. In route discovery process we made an authentication process between the nodes by using MD5 hashing algorithm, then we used reinforcement learning to improve the route maintenance process based on machine learning approach. At the end we used Diffie-Hellman key management to exchange the secret key to encrypt and decrypt the data between source S and destination D. When we tested the proposed protocol, the results show improvement in the performance of MANETs, despite the little increased in the end to end delay in comparison with DYMO protocol. This is due to the overheads in authentication and encryption processes.