Separation of a mixture of independent signals using time delayed correlations

Abstract

The problem of separating n linearly superimposed uncorrelated signals and determing their mixing coe cents is reduced to an Eigenvalue problem which involves the simultaneous diagonalisation of two symmetric matrices whose elements are measureable time delayed correlation functions. The diagonalisation matrix can be determined from a cost function whose number of minima is equal the number of degenerate solutions. Our approach o ers the possibility to separate also nonlinear mixtures of signals.

Long Duration Stimuli and Nonlinearities in the Neural–Haemodynamic Coupling

“…In all, 30 trials were collected at 15 Hz, each of 12 secs duration, stimulus onset at 8 secs, 15 secs inter-stimulus interval. Images were analysed using a variation of a signal-source separation algorithm (Molgedey and Schuster, 1994) exploiting a weak temporal model, as previously described (Zheng et al, 2001). The resultant barrel activation maps were registered with images of cortical surface to guide insertion of the electrophysiology probe.…”

Speech intelligibility improvement using convolutive blind source separation assisted by denoising algorithms

“…In such a case less restrictive conditions than statistical independence can be used such as second order statistics. Several approaches are based on this assumption (Molgedey and Schuster, 1994;Ziehe et al, 2000;Cichocki and Belouchrani, 2001;Choi et al, 2002;Choi et al, 2003).…”

A Rescue Radar System for the Detection of Victims Trapped Under Rubble Based on the Independent Component Analysis Algorithm

“…ICA algorithm has been chosen to accomplished this task. The ICA is a method for separating mixed data (such as MRI images [13], biomedical data [15], sounds, telecommunication channels or signals) into underlying informational components. The ICA belongs to a class of methods called blind sources separation (BBS).…”

“…The aim of this work is the design and development of a lightweight microwave detection system based on a compact X-band radar to detect victims trapped under rubble and other obstacles. The life signals are extracted from the modulated backscattered wave by the independent component analysis (ICA) algorithm [12][13][14], a powerful processing technique successfully adopted to analyze mixed signals and adopted recently in order to analyze brain signals and in particular electroencephalographic (EEG) data [15]. The preliminary experimental results, obtained considering a realistic scenario, are quite promising.…”