Congestion Aware Genetic Q-learning Based RPL Routing Protocol for Vehicle Ad-Hoc Networks

Vehicular Ad-Hoc Network (VANET) is the one among the prominent research area. On behalf of its eminent behaviour link dynamically varying topology, huge mobility, irregular connectivity and density it is still in the open research area. Its dynamic nature is the biggest challenge to do data transmission among the network. This research work introduced a novel protocol which is Congestion Aware Genetic Q-Learning (CAGQ) based RPL routing protocol for VANET networks. The major sub sections of the protocol are system model, mobility model, and preeminent path selection and trickle timer based genetic Q-Learning. The extensive evaluation is done through simulation which analysed the performance of RPL routing protocol in extremely dynamic vehicular environment. The contiki based Cooja simulator is employed for the process of simulation. The calculated results of the protocol CAGQ model is compared with the earlier works such as MRHOFL and EMCC. The analysis is done by calculating the parameters such as packet delivery ratio and Routing overhead. This result indicates that the proposed CAGQ method performed better and produced good results in data transmission even in huge mobility based VANETs.