Maximum Likelihood Source
نویسنده
چکیده
This paper deals with the source separation problem which consists in the separation of a mixture of independent sources without a priori knowledge on the mixing matrix. When the source distributions are known in advance, this problem can be solved via the Maximum Likelihood (ML) approach by maximizing the data likelihood function using (i) the Expectation-Maximization (EM) algorithm and (ii) a stochastic version of it, the SEM. Two important features of our algorithm are that (a) the covariance of the additive noise can be estimated as a regular parameter, (b) in the case of discrete sources, it is possible to separate more sources than sensors. The eeectiveness of this method is illustrated by numerical simulations.
منابع مشابه
A maximum-likelihood parametric approach to source localizations
Source localization using passive sensor arrays has been an active research problem for many years. Most near-field source localization algorithms involve two separate estimations, namely, relative time-delay estimations and source location estimation. In this paper, a one-step maximum-likelihood parametric source localization algorithm is proposed based on the maximum correlation between time ...
متن کاملApproximate Maximum Likelihood Blind Source Separation with Arbitrary Source Pdfs
We present a quasi-maximum likelihood approach to blind source separation (BSS) which is based on approximating the source distributions by their truncated Edgeworth expansions. The paper focuses on the 22 case, for which the problem is known to reduce to the estimation of a single rotation angle. Unlike existing maximum likelihood BSS techniques , the proposed algorithm is consistent for any s...
متن کاملMaximum likelihood estimation of the position of a radiating source in a waveguide
An array of sensors is receiving radiation from a source of interest. The source and the array are in a oneor two-dimensional waveguide. The maximum-likelihood estimators of the coordinates of the source are analysed under the assumptions thatthe noise field is Gaussian. The Cramer-Rao lower bound is of the order of the number of modes which define the source excitation function. The results sh...
متن کاملMinimum Entropy Algorithms for Source Separation
The minimum entropy or maximum likelihood estimation can be utilized in blind source separation problem. Based on the local generalized Gaussian probability density model, a set of general anti-Hebbian rule can be derived. This set of adaptation rules give promising results when we test the real recordings.
متن کاملChange Point Estimation of a Process Variance with a Linear Trend Disturbance
When a change occurs in a process, one expects to receive a signal from a control chart as quickly as possible. Upon the receipt of signal from the control chart a search for identifying the source of disturbance begins. However, searching for assignable cause around the signal time, due to the fact that the disturbance may have manifested itself into the rocess sometimes back, may not always l...
متن کاملMaximum Likelihood and Semiparametric Estimation in Logistic Models with Incomplete Covariate Data
Maximum likelihood estimation of regression parameters with incomplete covariate information usually requires a distributional assumption about the concerned covariates which implies a source of misspeciication. Semiparametric procedures avoid such assumptions at the expense of eeciency. A simulation study is carried out to get an idea of the performance of the maximum likelihood estima-tor und...
متن کامل