Modeling link lifetime data with parametric regression models in MANETs

Lifetime of a link defines the amount of time the link is available for transmission. In this work, we associate statistical parametric models to link lifetime distributions of commonly used mobility patterns. The study reveals interesting properties about link lifetimes, including findings showing that lifetimes of the mobility patterns studied are of wear-out type, rather than random failure. Such knowledge is useful in exploring different ways in which link lifetimes can be used to improve performance across various networking layers.