IDENTIFICATION OF PARAMETERS FOR ESTIMATION OF FREEWAY TRAFFIC USING INFORMATION FROM A MOBILE CELLULAR NETWORK

Vehicular traffic models are the basis for many activities in the field of transportation system control, such as vehicle routing, traffic estimation, safety, etc. Typically, these models have the common characteristic of depending on a large number of parameters that must be accurately tuned, often for different road morphologies and/or traffic patterns.In this paper, we propose a parameter identification mechanism based on stochastic approximation applied to a vehicular model which exploits data coming from a mobile cellular network; information available from the wireless network is used to feed both the vehicular model and the identification algorithm. Preliminary simulation results are reported and discussed.