Abstract | ||
---|---|---|
We introduce a new blind source separation (BSS) algorithm for correlated noncircular sources that uses only second-order statistics and fully takes the correlation structure into account. We propose a parametric entropy rate estimator that uses a widely linear autoregressive (AR) model for the sources, and derive the BSS algorithm by minimizing the mutual information of separated time series. We compare the performance of the new algorithm with competing algorithms and demonstrate its superior separation performance as well as its effectiveness in separation of non-Gaussian sources when the identification conditions are met. |
Year | DOI | Venue |
---|---|---|
2011 | 10.1109/TSP.2011.2114653 | IEEE Transactions on Signal Processing |
Keywords | Field | DocType |
correlation structure,gaussian entropy rate stationary gaussian source,second-order statistics,linear autoregressive model,new blind source separation,noncircular correlated sources,linear autoregressive,identification condition,independent component analysis,parametric entropy rate estimator,blind source separation algorithm,autoregressive processes,correlated noncircular source,blind separation,gaussian entropy rate,blind source separation,gaussian processes,non-gaussian source,mutual information,bss algorithm,new algorithm,noncircular sources,time series,superior separation performance,noncircular correlated source,complex gaussian sources,indexing terms,entropy rate,covariance matrix,correlation,cost function,approximation algorithms,entropy,ar model | Artificial intelligence,Gaussian process,Blind signal separation,Source separation,Mathematical optimization,Entropy rate,Pattern recognition,Algorithm,Gaussian,Parametric statistics,Independent component analysis,Mutual information,Mathematics | Journal |
Volume | Issue | ISSN |
59 | 6 | 1053-587X |
Citations | PageRank | References |
16 | 0.75 | 12 |
Authors | ||
2 |