Design of Opportunistic Routing Based on Markov Decision Process

The instability of wireless links makes the choice of forwarding nodes have too many possibilities in opportunistic routing. In this paper, the Markov decision process (MDP) is used to model the whole packet forwarding process from the source node to the destination node, which facilitates making reasonable decisions when the sender selects the forwarding nodes. By solving a finite-state MDP problem, an optimal forwarding strategy is obtained to minimize the expected number of transmissions for each node. Further more, the properties of expected any-path transmissions (EAX) are used for the selection of candidate forwarder set, which can significantly reduce the feasible optimal solution space. Finally, An opportunistic routing protocol based on Markov decision process (MDP-OR) is designed, whose effectiveness is verified by an example and its simulation.