Technical report bds : 00 - 02 Data acquisition , interfacing and pre - processing of highway traffic data ∗

If you want to cite this report, please use the following reference instead: Data acquisition, interfacing and pre-processing of highway traffic data, " Synopsis Data acquisition and pre-processing are important steps in any traffic data collection, analysis or simulation project. In this paper we present the results of a case study we have done on traffic simulation for a stretch of the E17 highway near Antwerp, Belgium. We will especially focus on the data collection and processing part. The objective of the work presented here is to illustrate how traffic data can be collected and processed in such a way that it can be used for modeling and prediction purposes. The data collection and processing process consists of three steps: data acquisition, interfacing, and pre-processing. During this process a central database with raw sensor data is created. In general the raw sensor data will contain errors or missing values due to sensor failures, data link errors, biases, and other measurement errors. Therefore, we introduce a pre-processing step in order to deal with data that is known to be corrupt or for which the value is temporarily not available (many simulation packages require data records that are complete and contain no gaps). After the three steps have been carried out, we have a database of " clean and complete " traffic data that can be used for further analysis or that can be used in other applications such as highway traffic simulation or traffic forecasts.gium. He is currently a research assistant funded by the Belgian federal government (Federal Office for Scientific, Technical and Cultural Affairs–OSTC) and is working towards a doctoral degree in Applied Sciences at the ESAT-SISTA research group of K.U.Leuven. The topic of his research is modeling and control of traffic flows on highways. De Moor has published more than 200 papers in international journals and conference proceedings and received several national and international awards for his work. He is also one of the founders of two spin-offs: ISMC NV (specialized in modeling and control of industrial processes), and Data4S NV (active in data-mining and services in bio-informatics and telecommunications). His research interests include numerical linear algebra, system identification, control theory, and data-mining.