Skip to Main Content
Skip Nav Destination
ASME Press Select Proceedings
Proceedings of the International Conference on Internet Technology and Security
Hao Xie
Hao Xie
Huazhong University of Science and Technology Mehmet OZKAN,
Sevilla University
Search for other works by this author on:
No. of Pages:
ASME Press
Publication date:

Particle filter (PF) has been widely used in target tracking for its advantage of not being restricted by the assumption of linearity and Gaussian approximation. But due to the degeneracy problem of PF, its performance will decline greatly in target tracking. A novel PF algorithm is proposed in this paper. In this algorithm, we firstly use the Maximum Likelihood (ML) method to estimate the latest location of mobile targets in Wireless Sensor Network (WSN), and then apply the Kalman filter (KF) to getting the targets' states (their instantaneous locations and velocities) and co-variances according to the location. Finally, the new targets' state and covariance are used to generate proposal distribution for PF with higher accuracy. This new algorithm is called MKPF. The simulation results indicate that MKPF is satisfactory both in accuracy and real-time.

I. Introduction
II. Multiple Targets Tracking Description
III. Multi-target data association algorithm
IV. An improved algorithm--mkpf
V. Simulations
VI. Conclusion
This content is only available via PDF.
You do not currently have access to this chapter.
Close Modal

or Create an Account

Close Modal
Close Modal