Increasingly, for many application areas, it is becoming important to include elements of nonlinearity and non-Gaussianity in order to model accurately the underlying dynamics of a physical system. Moreover, it is typically crucial to process data on-line as it arrives, both from the point of view of storage costs as well as for rapid adaptation to changing signal characteristics. In this paper, we review both optimal and suboptimal Bayesian algorithms for nonlinear/non-Gaussian tracking problems, with a focus on particle filters. Particle filters are sequential Monte Carlo methods based on point mass (or “particle”) representations of probability densities, which can be applied to any state-space model and which generalize the traditional Kalman filtering methods. Several variants of the particle filter such as SIR, ASIR, and RPF are introduced within a generic framework of the sequential importance sampling (SIS) algorithm. These are discussed and compared with the standard EKF through an illustrative example.

Abstract not found

Abstract- The mobile usage of Internet is characterized by frequent changes of the access network and consequently, changes of the application’s IP address or port. Intermediate NAT devices can exchange additionally the transport and network headers. Not using the current IP and port parameter leads to lost packets and service interruption. To overcome these problems, the applications send updates or keep-a-lives in regular basis, for example Dead-Peer-Detection in IKE. These messages inform the communication participants that the host is reachable. The main shortcoming is that the updates are performed at constant intervals regardless of the network properties. The problem oscillates in mobile environment where frequent network changes are expected. The result is wasted resources and long disconnection intervals. The key idea in this work is to set the update intervals proportional to the probability for network change. The probability density function is built using the past disconnections, thus the history is used to optimize the update intervals. Novel framework based on Particle filter is derived and simulated in this paper. The new method outperforms significantly the classical constant updates. Many protocols in mobile environment can profit from the new framework, like SIP, IKE, Routing protocols etc. Index Terms- update interval, keep-a-live, Particle filter, Sequential Monte Carlo, NAT, mobile environment

Abstract. A basic requirement for a practical tracking system is to adjust the tracking model in real time when the appearance of the tracked object changes. However, since the scale of the targets often varied irregularly, systems with fixed-size tracking window usually could not accommodate to these scenarios. In present paper, a new multi-scale information measure for image was introduced to probe the size-changes of tracked objects. An automatic window-size updating method was then proposed and integrated into the classical color histogram based meanshift and particle filtering tracking frameworks. Experimental results demonstrated that the improved algorithms could select the proper size of tracking window not only when the object scale increases but the scale decreases as well with minor extra computational overhead. 1

In the last decade alpha-stable distributions have become a standard model for impulsive data. Especially the linear symmetric alpha-stable processes have found applications in various fields. When the process parameters are timeinvariant, various techniques are available for estimation. However, time-invariance is an important restriction given that in many communications applications channels are time-varying. For such processes, we propose a relatively new technique, based on particle filters which obtained great success in tracking applications involving non-Gaussian signals and nonlinear systems. Since particle filtering is a sequential method, it enables us to track the time-varying autoregression coefficients of the alpha-stable processes. The method is tested both for abruptly and slowly changing autoregressive parameters of signals, where the driving noises are symmetric-alpha-stable processes and is observed to perform very well. Moreover, the method can easily be extended to skewed alpha-stable distributions. 1.