论文部分内容阅读
When the distribution of the sources cannot be e stimated accurately, the ICA algorithms failed to separate the mixtures blindly. The generalized Gaussian model (GGM) is presented in ICA algorithm since it can model non-Gaussian statistical structure of different source signals easily. B y inferring only one parameter, a wide class of statistical distributions can be characterized. By using maximum likelihood (ML) approach and natural gradient d escent, the learning rules of blind source separation (BSS) based on GGM are pre sented. The experiment of the ship-radiated noise demonstrates that the GGM can model the distributions of the ship-radiated noise and sea noise efficiently, and the learning rules based on GGM gives more successful separation results aft er comparing it with several conventional methods such as high order cumulants a nd Gaussian mixture density function.
The generalized Gaussian model (GGM) is presented in ICA algorithm since it can model non-Gaussian statistical structure of different source signals easily. B By using maximum likelihood (ML) approach and natural gradient d escent, the learning rules of blind source separation (BSS) based on GGM are pre sented. The experiment of the ship-radiated noise demonstrates that the GGM can model the distributions of the ship-radiated noise and sea noise efficiently, and the learning rules based on GGM gives more successful separation results aft er it with several conventional methods such as high order cumulants a nd Gaussian mixture density function.