Link Modeling and Delay Analysis in Networks with Disruptive Links

Abstract

Delay- and Disruption-Tolerant Networks (DTNs) refer to a range of networks with link intermittency that is mainly driven by mobility, predictable or unpredictable network environmental conditions. Examples of DTNs include interplanetary networks, battlefield networks, smart highways, remote sensing, and animal-movement outposts. There exist a number of mobility models describing the operation of various DTNs. One common characteristic that all mobility models share is the distribution of contact time and inter-contact time between nodes. Predicting an end-to-end delay in networks with disruptive links is more complicated than predicting the delay in connected networks. Disruptive patterns and underlying routing algorithms play a major role in an end-to-end delay modeling. In this article, we introduce a new model that can be used to estimate the end-to-end delay in networks with intermittent links. The model incorporates the two non-deterministic delay distributions, namely link intermittency and tandem queuing delay distributions. The model is based on an open queuing system with exponentially distributed link intermittency. The model gives a close approximation of the average end-to-end delay and the delay variance in closed forms. Simulation results on various networks and under different traffic conditions confirm the accuracy of the model within the conventional bounds of statistical significance.