A comparison study for traffic flow data compression

With the rapid development of detecting technology, the amount and scale of detected traffic data are increasing in a fast speed. Lots of methods have been proposed for traffic data compression, which contributes to saving storage space. This paper employed robust principal component analysis (RPCA), Kronecker product and tensor decomposition into traffic network flow data compression, and compared them with the compression method based on principal component analysis (PCA). The data points from PeMS are compressed and recovered by the four methods respectively. Results show that each method has its own characteristics and advantages.