Influence of particle filter parameters on error correction accuracy in traffic surveillance using sensor networks

Location estimation is an important part of a traffic surveillance system. Markov chain Monte Carlo methods based on particle filters have proved to be an effective solution in sensing error correction. We investigate in this paper the influence of particle filter parameters variation on sensing errors correction accuracy. Considered traffic surveillance system is based on a wireless sensor network. Several forms of probability density matrix and various methods for particle weight computation where considered, allowing us to find the dependencies between parameters. Finally, we use simulation to find optimal solutions in different traffic conditions.

[1]  M. Potkonjak,et al.  On-line fault detection of sensor measurements , 2003, Proceedings of IEEE Sensors 2003 (IEEE Cat. No.03CH37498).

[2]  Christian P. Robert,et al.  Monte Carlo Statistical Methods , 2005, Springer Texts in Statistics.

[3]  Kostia Robert.  Night-Time Traffic Surveillance: A Robust Framework for Multi-vehicle Detection, Classification and Tracking , 2009, 2009 Sixth IEEE International Conference on Advanced Video and Signal Based Surveillance.

[4]  Dieter Fox,et al.  Real-time particle filters , 2004, Proceedings of the IEEE.

[5]  Mark Coates,et al.  Distributed particle filters for sensor networks , 2004, Third International Symposium on Information Processing in Sensor Networks, 2004. IPSN 2004.

[6]  Philip Resnik,et al.  GIBBS SAMPLING FOR THE UNINITIATED , 2010 .

[7]  Amol Deshpande,et al.  Online Filtering, Smoothing and Probabilistic Modeling of Streaming data , 2008, 2008 IEEE 24th International Conference on Data Engineering.

[8]  K. Riedel.  Numerical Bayesian Methods Applied to Signal Processing , 1996 .

[9]  Christian P. Robert,et al.  Monte Carlo Statistical Methods (Springer Texts in Statistics) , 2005 .

[10]  Sebastian Thrun,et al.  Decentralized Sensor Fusion with Distributed Particle Filters , 2002, UAI.

[11]  Suchendra M. Bhandarkar,et al.  Fast and Robust Background Updating for Real-time Traffic Surveillance and Monitoring , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Workshops.

[12]  K OrJ.  Numerical Bayesian methods applied to signal processing , 1996 .

[13]  Sujit Dey,et al.  Model-Based Techniques for Data Reliability in Wireless Sensor Networks , 2009, IEEE Transactions on Mobile Computing.

[14]  Dan Pescaru,et al.  Influence of driver behavior patterns in correcting video sensing errors in traffic surveillance applications , 2009, 2009 IEEE 5th International Conference on Intelligent Computer Communication and Processing.