prior estimation
1.THE PRIOR ESTIMATION FOR COVARIANCE MATRIX OF STRUCTURAL PARAMETERS IN SYSTEM IDENTIFICATION PROCEDURES
3.1. When the number of groups is assumed known, firstly, under particularly prior of parameter, we prove that the bayesian estimation of the normal mixture model parameters is admissibility.
5.SEVERAL SORTS OF STOCHASTIC LINEAR PROGRAMMING AND ESTIMATION OF PRIOR DISTRIBUTION
7.Objective To study the methods of estimation of prior parameter and compare their application condition in the absence of the gold standard based on the Bayesian theory.
8.the other is that the components are independent each other. First, Bayesian Reliability estimation for component with exponential distribution is studied in various prior pdf: The failure rate function X is assumed as follows: Gamma prior conjugate distribution, Bayes hypothesis and Jeffrey norm. And I calculate the posterior pdf of reliability function R in Bayes theorem.
9.An SNR estimation algorithm was d eveloped based on eigenvalue decomposition of the correlation matrix of the rece ived signals and the principle of minimum description length (MDL) in informatio n theory. The algorithm can estimate the SNR of digital modulation signals commo nly used in additional white Gaussian noise (AWGN) channels and multipath channe ls without prior information of the modulation type, baud rate or carrier freque ncy of the signals.
10.Abstract: The conventional variable structure control technique for uncertain system requires that the uncertainty bound is known as a premise to assure robustness.The requirement creates an over-conservative controller and enlarges chattering.The proposed controller regards the influence of unknown disturbances and parameter uncertainties as an equivalent disturbance and generates an on-line estimation used in SMC to cancel the slowly varying uncertainties by the mechanism of time delay.The reaching law approach is used to get the conditions and band of quasi-sliding mode.The new methodology offers a robust feedback control with much lower gains and reduces chattering without a prior knowledge of the uncertainty bounds or matched conditions.
