Improved pruning algorithm for Gaussian mixture probability hypothesis density filter

With the increment of the number of Gaussian components, the computation cost increases in the Gaussian mixture probability hypothesis density(GM-PHD) filter. Based on the theory of Chen et al, we propose an improved pruning algorithm for the GM-PHD filter, which utilizes not only the Gaussian components’ means and covariance, but their weights as a new criterion to improve the estimate accuracy of the conventional pruning algorithm for tracking very closely proximity targets. Moreover, it solves the end-less while-loop problem without the need of a second merging step. Simulation results show that this improved algorithm is easier to implement and more robust than the formal ones.

An evolutionary particle filter based EM algorithm and its application

In this paper, an evolutionary recursive Bayesian estimation algorithm is presented, which incorporates the latest observation with a new proposal distribution, and the posterior state density is represented by a Gaussian mixture model that is recovered from the weighted particle set of the measurement update step by means of a weighted expectation-maximization algorithm. This step replaces the resampling stage needed by most particle filters and relieves the effect caused by sample impoverishment. A nonlinear tracking problem shows that this new approach outperforms other related particle filters.

Mapping alteration minerals using sub-pixel unmixing of ASTER data in the Sarduiyeh area,SE Kerman,Iran

This paper is an attempt to introduce the role of earth observation technology and a type of digital earth processing in mineral resources exploration and assessment.The sub-pixel distribution and quantity of alteration minerals were mapped using linear spectral unmixing(LSU)and mixture tuned matched filtering(MTMF)algorithms in the Sarduiyeh area,SE Kerman,Iran,using the visible-near infrared(VNIR)and short wave infrared(SWIR)bands of the Advanced Spaceborne Thermal Emission and Reflection Radiometer(ASTER)instrument and the results were compared to evaluate the efficiency of methods.Three groups of alteration minerals were identified:(1)pyrophylite-alunite(2)sericite-kaolinite,and(3)chlorite-calcite-epidote.Results showed that high abundances within pixels were successfully corresponded to the alteration zones.In addition,a number of unreported altered areas were identified.Field observations and X-ray diffraction(XRD)analysis of field samples confirmed the dominant mineral phases identified remotely.Results of LSU and MTMF were generally similar with overall accuracy of 82.9 and 90.24%,respectively.It is concluded that LSU and MTMF are suitable for sub-pixel mapping of alteration minerals and when the purpose is identification of particular targets,rather than all the elements in the scene,the MTMF algorithm could be proposed.

A multiple maneuvering targets tracking algorithm based on a generalized pseudo-Bayesian estimator of first order

We describe the design of a multiple maneuvering targets tracking algorithm under the framework of Gaussian mixture probability hypothesis density(PHD) filter.First,a variation of the generalized pseudo-Bayesian estimator of first order(VGPB1) is designed to adapt to the Gaussian mixture PHD filter for jump Markov system models(JMS-PHD).The probability of each kinematic model,which is used in the JMS-PHD filter,is updated with VGPB1.The weighted sum of state,associated covariance,and weights for Gaussian components are then calculated.Pruning and merging techniques are also adopted in this algorithm to increase efficiency.Performance of the proposed algorithm is compared with that of the JMS-PHD filter.Monte-Carlo simulation results demonstrate that the optimal subpattern assignment(OSPA) distances of the proposed algorithm are lower than those of the JMS-PHD filter for maneuvering targets tracking.

Monte Carlo Likelihood Estimation of Mixed-Effects State Space Models with Application to HIV Dynamics

The statistical inference for generalized mixed-effects state space models （MESSM） are investigated when the random effects are unknown. Two filtering algorithms are designed both of which are based on mixture Kalman filter. These algorithms are particularly useful when the longitudinal ts are sparse. The authors also propose a globally convergent algorithm for parameter estimation of MESSM which can be used to locate the initial value of parameters for local while more efficient algorithms. Simulation examples are carried out which validate the efficacy of the proposed approaches. A data set from the clinical trial is investigated and a smaller mean square error is achieved compared to the existing results in literatures.