A unifying maximum-likelihood view of cumulant and polyspectral measures for non-Gaussian signal classification and estimation.
Georgios B. GiannakisMichail K. TsatsanisPublished in: IEEE Trans. Inf. Theory (1992)
Keyphrases
- maximum likelihood
- gaussian distribution
- parameter estimation
- maximum likelihood estimation
- estimation algorithm
- log likelihood function
- blind source separation
- em algorithm
- maximum likelihood estimator
- expectation maximization
- multiple views
- mixture model
- continuous valued
- independent component analysis
- non stationary
- likelihood function
- robust estimation
- markov networks
- maximum a posteriori
- gaussian mixture model
- density estimation
- estimation accuracy
- heavy tailed
- maximum likelihood estimates
- quantitative measures
- hyperparameters
- probabilistic mixture model