Modeling synchronized time series

Abstract. We consider the class of applications that manage time series (TS) and propose a data model and a query language that let these applications manipulate TS data sets at a logical level. We introduce the concept of synchronized time series (STS) to model the alignment of several time series in a common time space. We show how this concept helps to analyze and compare information extracted from time series, and how it constitutes a convenient tool that supports an extended set of operations. The main contribution of the paper is a formal query language that combines generic operators on TS with user-defined functions. The language can be evaluated in closed form over STS data sets and gives rise to rewriting and optimization techniques. Throughout the paper, we illustrate our approach with several examples drawn from a few representative TS applications.

• Type: Journal
• Year: 2010
• Where: IDEAS
• Authors: Zoé Faget, Philippe Rigaux, David Gross-Amblard, Virginie Thion-Goasdoué
• Keywords: Common Time Space, Database, IDEAS 2010, Query Language, Time Series