Reverse AODV routing protocol extension using learning Automata in ad hoc networks

This paper proposes a novel reserve on-demand routing protocol for mobile ad hoc networks based on best route selection with learning autonoma. Our proposed protocol capable of operating efficiently under bursty traffic conditions, is introduced. We applied our method in an optimized version of ad hoc on-demand distance vector (AODV) routing algorithm, namely Reverse AODV (R-AODV) routing algorithm. In path discovery phase, the source node select best route that has high stability between available routes. The main idea that used in proposed protocol is changing fitness of routes passing time Also with the awareness of fitness of routes, source node can select best route in sets of available route, when active route fails. For do this we changed R-AODV algorithm in some ways. In our algorithm namely Reverse AODV with learning Automata (RAODVA) routing algorithm, the route request packet not changed for estimation route stability and building multiple routes. Computer simulation using ns-2 simulator is performed with competition to other methods and effectiveness of the proposed method is quantitatively validated.