Spectral analysis of Floating Car Data

Abstract. Floating Car Data (FCD) are one important data source in traffic telematic systems. The original variable in these systems is the vehicle velocity. The paper analyses the measured value “vehicle velocity" by methods of information technology. Consequences for processing, transmission and storage of FCD under condition of limited resources are discussed.

Starting point of the investigation is the analysis of spectral characteristics of velocity-time-profiles. The spectra are determined by the Discrete Fourier Transform (DFT) from measurement data and simulation data of a microscopic traffic model. One essential property of velocity-time-profiles is their low-pass characteristic. The resulting cut-off-frequency is comparatively small and depends on the traffic scenario. Conclusions concerning the necessary sample rate in FCD systems and the processing of raw data are discussed. Finally the transinformation of velocity-time-profiles is analysed. This results in similar values for an optimal sample rate of FCD systems under condition of limited transmission capacity.