The mixture of factor analyzers （MFA） can accurately describe high resolution range profile （HRRP） statistical charac- teristics. But how to determine the proper number of the models is a problem. This paper devel...The mixture of factor analyzers （MFA） can accurately describe high resolution range profile （HRRP） statistical charac- teristics. But how to determine the proper number of the models is a problem. This paper develops a variational Bayesian mixture of factor analyzers （VBMFA） model. This procedure can obtain a lower bound on the Bayesian integral using the Jensen＇s inequality. An analytical solution of the Bayesian integral could be obtained by a hypothesis that latent variables in the model are indepen- dent. During computing the parameters of the model, birth-death moves are utilized to determine the optimal number of model au- tomatically. Experimental results for measured data show that the VBMFA method has better recognition performance than FA and MFA method.展开更多
The theory of slow backward-wave amplifications is developed based on electron cyclotron maser （ECM） mechanism employing an initially rectilinear beam, A nonlinear evolution equation is derived to describe the elect...The theory of slow backward-wave amplifications is developed based on electron cyclotron maser （ECM） mechanism employing an initially rectilinear beam, A nonlinear evolution equation is derived to describe the electron energy. Numerical calculations show that the saturated interaction efficiency in this system may exceed 20~, and the saturated interaction length spans 3-6 centimeters. The distinctive interaction mechanism is promising for the design of compact backward microwave amplification devices, Numerical studies are also presented for the slow-wave ECM efficiency with inclusion of Gaussian beam electron velocity spread. It is shown that the velocity spread reduces the interaction e^ciency.展开更多
基金supported in part by the National Natural Science Foundation of China(60772140)the Program for Cheung Kong Scholarsand Innovative Research Team in University(IRT0645)
文摘The mixture of factor analyzers （MFA） can accurately describe high resolution range profile （HRRP） statistical charac- teristics. But how to determine the proper number of the models is a problem. This paper develops a variational Bayesian mixture of factor analyzers （VBMFA） model. This procedure can obtain a lower bound on the Bayesian integral using the Jensen＇s inequality. An analytical solution of the Bayesian integral could be obtained by a hypothesis that latent variables in the model are indepen- dent. During computing the parameters of the model, birth-death moves are utilized to determine the optimal number of model au- tomatically. Experimental results for measured data show that the VBMFA method has better recognition performance than FA and MFA method.
基金supported by National Natural Science Foundation of China(Nos.11275007 and 11175023)the Program for Liaoning Excellent Talents in University(LJQ2012098)
文摘The theory of slow backward-wave amplifications is developed based on electron cyclotron maser （ECM） mechanism employing an initially rectilinear beam, A nonlinear evolution equation is derived to describe the electron energy. Numerical calculations show that the saturated interaction efficiency in this system may exceed 20~, and the saturated interaction length spans 3-6 centimeters. The distinctive interaction mechanism is promising for the design of compact backward microwave amplification devices, Numerical studies are also presented for the slow-wave ECM efficiency with inclusion of Gaussian beam electron velocity spread. It is shown that the velocity spread reduces the interaction e^ciency.
