Data Reduction of Spatio-temporal Trajectories using a Modified Online Compression Algorithm

Abstract

With development of mobile devices equipped with a global positioning system, such as smartphones, large amounts of spatial information are generated. These data, which are often stored and modeled as a sequence of spatial locations over time, are called trajectory. The large amount of trajectory data has increased the cost of transferring, storing and processing such data. To overcome these problems, a number of compression algorithms have been proposed for reducing the size of trajectory data. In this paper, seven algorithms including uniform sampling, Douglas Poker, TD-TR, Opening Window, OPW-TR, TD-SB and SQUISH-E algorithms are being discussed and the advantages and disadvantages of these algorithms are investigated as well. The SQUISH-E algorithm can create a balance between the compression rate and the Synchronized Euclidean Distance error, but has a high compression rate than other compression algorithms. To solve mentioned problem, this paper proposed a method for changing the priority window of the SQUISH-E algorithm, which improves the compression rate of this algorithm. In order to evaluate the performance of the proposed method, all algorithms are implemented on six trajectories of varying complexity and compared with each other in terms of criteria such as compression rate, run-time, and concurrency Euclidean distance errors. The results of implementation of the proposed method indicate the improvement of the proposed algorithm at the compression rate, computation time, and Synchronized Euclidean Distance error. In compare to SQUISH-E algorithm, the computation time and compression rate of proposed algorithm is decreased about 130 millisecond and 0.015, respectively.

A Flexible Trajectory Compression Algorithm for Multi-Modal Transportation

“…A problem that immediately arises is that raw trajectories contain a large amount of duplicate data. Therefore, storing, retrieving, managing, processing, and querying these trajectories is computationally expensive and requires large storage spaces (Nasiri, Azimi, and Abbaspour, 2018;Zhao and Shi, 2019). A critical step in trajectory representation and preprocessing is to reduce incoming data volumes while preserving the main properties and semantics associated with these trajectories and resulting movement patterns (Makris, Kontopoulos, Alimisis, and Tserpes, 2021).…”