Mobility Models for Next Generation Wireless Networks

Chapter 6

Group Mobility and Other Synthetic Mobility Models

In the previous chapters, we introduced synthetic mobility models for MANETs sharing the property of being entity-based: that is, mobility of a network composed of n nodes is modeled as n independent stochastic processes with the same statistical properties, each modeling the movement of a single node. In many situations, though, movement of mobile entities displays spatial correlation, that is, the movement of an entity is influenced by the movement patterns of other entities in the surroundings. A typical such example is movement of vehicles along a road: if vehicle A follows vehicle B on a single-lane road where overtaking is forbidden, the velocity of A is clearly upper bounded by the velocity of the preceding vehicle B.

Among mobility models taking spatial correlation of movement into account, an important class is that of group mobility models, whose purpose is to model situations where the movement patterns of the members of a subset of the network entities (called a group) are highly correlated. Examples of group mobility in real-world scenarios are soldiers moving on the battlefield, disaster recovery and law enforcement operations, movement of tourist groups in a museum or a city, etc.

It is important to observe that group mobility can be considered as an instance of the more general class of spatially correlated mobility models. In fact, in the former class of models nodes are explicitly and statically partitioned ...