Mixture models have become more popular in modelling compared to standard distributions. The mixing distributions play a role in capturing the variability of the random variable in the conditional distribution. Studie...Mixture models have become more popular in modelling compared to standard distributions. The mixing distributions play a role in capturing the variability of the random variable in the conditional distribution. Studies have lately focused on finite mixture models as mixing distributions in the mixing mechanism. In the present work, we consider a Normal Variance Mean mix<span>ture model. The mixing distribution is a finite mixture of two special cases of</span><span> Generalised Inverse Gaussian distribution with indexes <span style="white-space:nowrap;">-1/2 and -3/2</span>. The </span><span>parameters of the mixed model are obtained via the Expectation-Maximization</span><span> (EM) algorithm. The iterative scheme is based on a presentation of the normal equations. An application to some financial data has been done.展开更多
本文针对杂波条件下多扩展目标的状态估计,目标个数估计,扩展目标形状估计问题,提出了一种基于标签随机有限集(Labelled random finite sets,L-RFS)框架下多扩展目标跟踪学习算法,该学习算法主要包括两方面:多扩展目标动态建模和多扩展...本文针对杂波条件下多扩展目标的状态估计,目标个数估计,扩展目标形状估计问题,提出了一种基于标签随机有限集(Labelled random finite sets,L-RFS)框架下多扩展目标跟踪学习算法,该学习算法主要包括两方面:多扩展目标动态建模和多扩展目标的跟踪估计.首先,结合广义标签多伯努利滤波器(Generalized labelled multi-Bernoulli,GLMB)建立了扩展目标的量测有限混合模型(Finite mixture models,FMM),利用Gibbs采样和贝叶斯信息准则(Bayesian information criterion,BIC)准则推导出有限混合模型的参数来对多扩展目标形状进行学习,然后采用等效量测方法来替代扩展目标产生的量测,对扩展目标形状采用椭圆逼近建模,实现扩展目标形状与状态的估计.仿真实验表明本文所给的方法能够有效跟踪多扩展目标,并且在目标个数估计方面优于CBMeMBer算法.此外,与标签多伯努利滤波(LMB)计算比较表明:GLMB和LMB算法滤波估计精度接近,二者精度高于CBMeMBer算法.展开更多
文摘Mixture models have become more popular in modelling compared to standard distributions. The mixing distributions play a role in capturing the variability of the random variable in the conditional distribution. Studies have lately focused on finite mixture models as mixing distributions in the mixing mechanism. In the present work, we consider a Normal Variance Mean mix<span>ture model. The mixing distribution is a finite mixture of two special cases of</span><span> Generalised Inverse Gaussian distribution with indexes <span style="white-space:nowrap;">-1/2 and -3/2</span>. The </span><span>parameters of the mixed model are obtained via the Expectation-Maximization</span><span> (EM) algorithm. The iterative scheme is based on a presentation of the normal equations. An application to some financial data has been done.
基金Supported by National Natural Science Foundation of China(11471104)Natural Science Foundation of Henan Educational Committee(2011B110018)Program for Innovative Research Team(in Science and Technology)in University of Henan Province(14IRTSTHN023)
文摘本文针对杂波条件下多扩展目标的状态估计,目标个数估计,扩展目标形状估计问题,提出了一种基于标签随机有限集(Labelled random finite sets,L-RFS)框架下多扩展目标跟踪学习算法,该学习算法主要包括两方面:多扩展目标动态建模和多扩展目标的跟踪估计.首先,结合广义标签多伯努利滤波器(Generalized labelled multi-Bernoulli,GLMB)建立了扩展目标的量测有限混合模型(Finite mixture models,FMM),利用Gibbs采样和贝叶斯信息准则(Bayesian information criterion,BIC)准则推导出有限混合模型的参数来对多扩展目标形状进行学习,然后采用等效量测方法来替代扩展目标产生的量测,对扩展目标形状采用椭圆逼近建模,实现扩展目标形状与状态的估计.仿真实验表明本文所给的方法能够有效跟踪多扩展目标,并且在目标个数估计方面优于CBMeMBer算法.此外,与标签多伯努利滤波(LMB)计算比较表明:GLMB和LMB算法滤波估计精度接近,二者精度高于CBMeMBer算法.
