Multi-Parameter Estimation in Compound Gaussian Clutter by Variational Bayesian
| Autor: | Yuanwei Jin, Anish C. Turlapaty |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
020301 aerospace & aeronautics
Estimation theory business.industry Gaussian Bayesian probability 020206 networking & telecommunications Pattern recognition 02 engineering and technology symbols.namesake 0203 mechanical engineering Signal Processing Expectation–maximization algorithm 0202 electrical engineering electronic engineering information engineering symbols Variational message passing Applied mathematics Clutter Artificial intelligence Electrical and Electronic Engineering Bayesian linear regression business Recursive Bayesian estimation Mathematics |
| Zdroj: | IEEE Transactions on Signal Processing. 64:4663-4678 |
| ISSN: | 1941-0476 1053-587X |
| Popis: | In this paper, we consider the problem of multi-parameter estimation in the presence of compound Gaussian clutter for cognitive radar by the variational Bayesian method. The advantage of variational Bayesian is that the estimation of multi-variate parameters is decomposed to problems of estimation of univariate parameters by variational approximation, thus enabling analytically tractable approximate posterior densities in complex statistical models consisting of observed data, unknown parameters, and hidden variables. We derive the asymptotic Bayesian Cramer–Rao bounds and demonstrate by numerical simulations that the proposed approach leads to improved estimation accuracy than the expectation maximization method and the exact Bayesian method in the case of non-Gaussian nonlinear signal models and small data sample size. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|