Analysis of Trend in Vehicular Traffic Flow Data by Wavelets

This paper proposes a method to model the daily trend in vehicular traffic flow data using a wavelet transform technique. The approach in this paper first decomposes the historical traffic volume to an approximate part associated with low frequency fluctuations and several detailed parts associated with high frequency fluctuations by means of discrete wavelet transform. The approximate part, averaged over a few days, represents the daily trend of traffic flow. The high frequency parts are separately modelled and added to the trend to get an approximate nature of the traffic flow over a day. A case study is performed at a traffic junction in Dublin with the proposed approach