查看文章

The problem of nonlinear blind source separation (B55) is a challenging research task, and several methods have been proposed in the literature (Burel, 1992; Valpola, Giannakopoulos, Honkela, & Karhunen, 2000; Lee, Koehler, & Orglmeister, 1997; Lin, Grier, & Cowan, 1997; Yang, Amari, & Cichocki, 1998; Marques & Almeida, 1999; Pajunen, Hyvarinen, & Karhunen, 1996; Pajunen & Karhunen, 1997; Fyfe & Lai, 2000; Taleb & Jutten, 1999; Hyvarinen, Karhunen, & Oja, 2001). Fruitful applications are conceivable in, among others, the fields of telecommunications, array processing, Vision, biomedical data analysis, and acoustic source separation where nonlinearities can occur in the mixing process. Nonlinear BSS has recently been applied to data from industrial pulp processing (Hyvarinen et al., 2001). In nonlinear BSS we observe a mixed signal,