Sound Compression Model Based on Overcomplete ICA
基于Overcomplet ICA的声音压缩模型

ICA is a statistical method for transforming an observed multidimensional random vector into components that are mutually independent as possible. In this paper we introduce Algebra ICA algorithm (AICA) which is the ef- ficient approach to solve Overcomplete ICA algorithm problem. Then we put forward a new sound compression struc- ture based on the Overcomplete ICA algorithm(SCO). Finally we present an experiment on sound signal compression and decompression.