Abstract—Locational data are extremely useful resource to
study customer behavior and mobility patterns. In this paper,
beyond directly measuring how their location, velocity and
acceleration change over time, we extend our discussion to
construct a data driven model to quantitatively evaluate the
moving objects’ interests and intentions, which are represented
by their waypoints distributions. Waypoints are defined with
the Random Waypoint (RWP) mobility model, which is one of
the most commonly used models in mobility management. To
effectively deploy RWP model, the detection of accurate
waypoint distribution is crucial and, however, challenging in
most practical situations. Moreover, to understand the how and
why an object moves in a its specific pattern, the knowledge of
waypoint distribution could be valuable in many use cases. In
this work, we analytically derive the relationship between
waypoint distribution and the locational data that could be
obtained directly from sensors, such as the number of objects’
arrivals to a particular area. An estimation scheme using
supervised learning algorithm is proposed to simplify the
evaluation of our model. Simulations are carried out to verify
the correctness and accuracy of our proposed scheme.
Index Terms—Locational data, mobility management,
waypoint distribution, supervised learning.
Wang Ting is now with SAP Asia Ptd Ltd, Singapore (e-mail:
dean.wang@sap.com).
Chor Ping Low is with School of Electrical and Electronic Engineering,
Nanyang Technological University, Singapore (e-mail: icplow@ntu.edu.sg).
Cite:Ting Wang and Chor Ping Low, "A Data Driven Model for the Detection of Random Waypoint," International Journal of Information and Education Technology vol. 3, no. 4, pp. 417-423, 2013.