Connectivity Models : A New Approach to Modeling Contacts in Opportunistic Networks. ∗
Thabotharan Kathiravelu, Arnold Pears, Nalin Ranasinghe · 2006
The influence of node mobility in Mobile Ad hoc NETworks (MANETs) is significant and has significant implications for system performance. The difficulty and cost associated with collecting real movement traces has resulted in the use of synthetic mobility models in many simulation studies. Simulations use mobility models to describe how devices move in the geographical (coordinate) space of the simulated deployment environment. Typical mobility models describe the location, speed, direction of movement etc., of mobile entities over time [1]. It has been observed by many researchers that synthetic mobility models fail to capture common features of real world movement of humans and the connectivity patterns they establish [4, 16]. This has prompted the development of a range of more complex mobility models which address deficiencies in earlier models [7,8,11,14]. However, the increase in the fidelity of mobility models comes at a price. There is a corresponding increase in simulation storage and time overheads. This paper argues that for Opportunistic networking simulations high fidelity mobility models are both too costly and inappropriate. We summarize the early proposals in mobility models for MANETs, then look at the relevance of recent advances. We introduce connectivity mod