Piecewise Linear Representation Segmentation in Noisy Domains With a Large Number of Measurements: The Air Traffic Control Domain

Abstract

The importance of time series segmentation techniques is rapidly expanding, due to the growth in collection and storage technologies. Among them, one of the most used ones is Piecewise Linear Representation, probably due to its ease of use. This work tries to determine the difficulties faced by this technique when the analyzed time series shows noisy data and a large number of measurements and how to introduce the information about the present noise in the segmentation process. Both difficulties are met in the Air Traffic Control domain, which exhibits position measurements of aircraft's trajectories coming from sensor devices (basically surveillance radar and aircraft-derived data), being used as the motivating domain. Results from the three main traditional techniques are presented (sliding window, top down and bottom up approaches) and compared with a new introduced approach, the Hybrid Local Residue Analysis technique.

Initialization Procedures for Multiobjective Evolutionary Approaches to the Segmentation Issue

“…Secondly, the configuration of such processes can be problem dependent due to the required parameters (total error and max segment error) and, thus, a general technique may be hard to apply to a set of problem instances with consistent results. The consideration for the measurement of several objectives in conflict to test the quality of a segmentation process was faced in [8] with the use of multiobjective quality indicators [20]. Given the multiobjective nature of this process, in [7] the segmentation issue was formalized according to equation 2, which explicitly presents this nature.…”

A multi-objective approach for the segmentation issue

“…The basis for this formalization can be found in Guerrero et al (2011), where it is applied for comparison issues over single-objective algorithms, with specific quality measures for the air traffic control domain. Once the problem has been formalized, it is interesting to analyse the way in which this multi-objective formulation has been tackled in the available algorithms.…”