Telemetry Data Mining Techniques, Applications, and Challenges

Trajectory Data Model

A spatial trajectory is a trace generated by a moving object in geographical spaces, which is consisting of an ordered set of spatiotemporal points (Frentzos, 2007). This can be defined for any trajectory “T” which can be seen as an ordered set of spatiotemporal points consisting of 3 dimensions: location in terms of x-coordinate “x”, y-coordinate “y” and temporal dimension in terms of time “t” This is formally defined as, and can be seen in Figure1, (Al-Serafi, 2015)

Figure 1.

Trajectory consisting of “n” points

Application Of Trajectory Data Mining

There exists a wide spectrum of applications driven and improved by trajectory data mining, such as; knowing moving objects locations in advance can be substantial. Discovery of behavioral patterns and prediction of future movement can greatly influence different fields, such as analysis of the wild animals’ movement in order to predict their migrations, monitoring and analysis of vehicle movement in order to predict traffic congestions, mobile user movement and access point availability prediction in order to assure the requested level of quality of service or analysis and location prediction of the movement of aircrafts or space crafts (Ivana, 2006). In the section, we classify these applications based on the derivation of trajectories categories. The derivation of trajectories can be classified into four major categories, mobility of people, mobility of animals, mobility of vehicles and mobility of natural phenomena.