Title :
Blind deconvolution of sparse spike trains using stochastic optimization
Author_Institution :
Signal & Image Process. Inst., Univ. of Southern California, Los Angeles, CA, USA
Abstract :
The author proposes an approach to blind deconvolution of Bernoulli-Gaussian processes based upon true maximum-likelihood estimation of fixed-dimension quantities. The maximum likelihood estimator is implemented using a stochastic version of the expectation-maximization (EM) algorithm. Practical results are in accordance with the known behavior of maximum-likelihood estimates, and the algorithm presents a moderate numerical complexity
Keywords :
maximum likelihood estimation; signal processing; stochastic processes; Bernoulli-Gaussian processes; blind deconvolution; expectation-maximization; maximum-likelihood estimation; numerical complexity; sparse spike trains; stochastic optimization; Additive noise; Deconvolution; Image processing; Image restoration; Linear systems; Maximum likelihood estimation; Signal processing; Speech processing; Stochastic processes; Stochastic resonance;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1992. ICASSP-92., 1992 IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0532-9
DOI :
10.1109/ICASSP.1992.226378
