The Propagation Delay Evaluation And Prediction Model Of Probabilistic Routing Algorithms For Opportunistic Networks

Opportunistic Networks is a kinds of new Ad hoc networks that relies on the nodes’moving to bring communication opportunities. It is different from the traditional networks, because there is not a fixed end-to-end path between the source node and the destination node. Since the routing mode of Opportunistic Networks using is different from the mode of traditional networks, the traditional routing protocols do not apply to Opportunistic Networks. The research of routing algorithms is always a hot topic in Opportunistic Networks. Although there is much research of the routing algorithms, the research about the performance evaluation of routing algorithms is still relatively few. In this paper, we first study the movement law of human society by combining complex networks and sociological knowledge to improve the application of the inter-contact time. We find that the inter-contact time between humans follows a power-law distribution. Based on these characteristics, we use Markov chains to simulate the forwarding process of probabilistic routing algorithms. We propose separately a propagation delay evaluation model and a TTL prediction model of probabilistic routing algorithms.In this paper, the two models are all based on the hypothesis that the inter-contact time between humans follows a power-law distribution other than the former hypothesis that the inter-contact time between humans follows exponential distribution, because the power-law distribution is closer to the real laws of human movement. The propagation delay evaluation model of probabilistic routing algorithms uses the performance indicators of propagation delay to evaluate the probabilistic routing algorithms. And this model is suitable both for the single-copy and multi-copy schemes of the probabilistic routing. At the same time, our model gives out a specific way to calculate the messages propagation delay for the multi-copy schemes of probabilistic routing. And our work can make the research of the multi-copy schemes no longer remains on the stage of finding the convergence condition of messages propagation delay. Based on the first model, this paper proposes a TTL prediction model with guarantee of success rate for probabilistic routing algorithms, this model increases the transmission success rate of routing algorithms to model the forwarding process of probabilistic routing algorithms. Our model not only can calculate the propagation delay of messages, but also can predict the TTL under the given conditions of transmission success rate. Our work can make the TTL of probabilistic routing algorithms no longer relies on the experience to set, but it can be predicted by our model.