MTSC: An Effective Multiple Time Series Compressing Approach

Abstract

As the volume of time series data being accumulated is likely to soar, time series compression has become essential in a wide range of sensor-data applications, like Industry 4.0 and Smart grid. Compressing multiple time series simultaneously by exploiting the correlation between time series is more desirable. In this paper, we present MTSC, a novel approach to approximate multiple time series. First, we define a novel representation model, which uses a base series and a single value to represent each series. Second, two graph-based algorithms, $MTSC_{mc}$ and $MTSC_{star}$, are proposed to group time series into clusters. $MTSC_{mc}$ can achieve higher compression ratio, while $MTSC_{star}$ is much more efficient by sacrificing the compression ratio slightly. We conduct extensive experiments on real-world datasets, and the results verify that our approach outperforms existing approaches greatly.

The work is supported by the Ministry of Science and Technology of China, National Key Research and Development Program (No. 2016YFB1000700), National Key Basic Research Program of China (No. 2015CB358800), NSFC (61672163, U1509213), Shanghai Innovation Action Project (No. 16DZ1100200).