Abstract
Independent component analysis (ICA) is a popular technique for separating sources from observed linear mixtures of the sources. If the measured signals are corrupted by noise, then they are generally preprocessed before applying ICA. We make use of a recently developed technique known as the iterative principal component analysis (IPCA) to preprocess the noisy signals and estimate the signal subspace prior to application of ICA. This preprocessing technique is consistent with the assumptions made in ICA, is invariant to any scaling of the data, and accounts for heteroscedastic errors. Through a simulated example, we show that if the measured signals are contaminated with a high-level of additive noise and outliers, the use of the proposed preprocessing technique results in more precise extraction of the independent sources as compared to the use of PCA as a preprocessing method. Application of the technique to an experimental chemometric data set shows that pure species spectra are more accurately extracted from mixture spectra using ICA, if the data is preprocessed using IPCA.
Similar content being viewed by others
References
Jutten, C., Herault, J.: Separation of sources-part I. Signal Process. 24, 1–10 (1991)
Vigário, R., Särelä, J., Jousmäki, V., Hämäläinen, M., Oja, E.: Independent component approach to the analysis of EEG and MEG recordings. IEEE Trans. Biomed. Eng. 47, 589–593 (2000)
Chen, J., Wang, X.Z.: A new approach to near-infrared spectral data analysis. J. Chem. Inf. Comput. Sci. 41, 992–1001 (2001)
Cichocki, A., Amari, S.: Adaptive Blind Signal and Image Processing: Learning Algorithms and Applications. Wiley, Chichester (2002)
Lee, D.D., Seung, H.S.: Algorithms for non-negative matrix factorization. Advances in neural information processing (13), MIT press, Cambridge (2001)
Comon, P.: Independent component analysis, a new concept? Signal Process. 36, 287–314 (1994)
Hyvärinen, A.: A survey on independent component analysis. Neural Comput Surv 2, 294–328 (1999)
Hyvärinen, A., Karhunen, J., Oja, E.: Independent Component Analysis. Wiley, Chichester (2001)
Hyvärinen, A.: Gaussian moments for noisy independent component analysis. IEEE Signal Process. Lett. 6, 145–147 (1999)
Hyvärinen, A.: Independent component analysis in the presence of Gaussian noise by maximum joint likelihood. Neurocomputing 22, 49–67 (1998)
Joho, M., Mattias, H., Lambert, R.H.: Overdetermined blind source separation using more sensors than source signals in a noisy mixture, pp. 81–86. Proceedings of ICA, Helsinki (2000)
Mathis, H., Joho, M.: Blind signal separation in noisy environments using a three step quantizer. Neurocomputing 49, 61–78 (2002)
Cichocki, A., Douglas, S.C., Amari, S.: Robust techniques for independent component analysis (ICA) with noisy data. Neurocomputing 22, 113–129 (1998)
Vorobyov, S., Cichocki, A.: Blind noise reduction for multisensory signals using ICA and subspace filtering with applications to EEG analysis. Biol. Cybern. 86, 293–303 (2002)
Attias, H.: Independent factor analysis. Neural Comput. 11, 803–851 (1999)
Ikeda, S., Toyoma, K.: Independent component analysis for noisy data: MEG data analysis. Neural Netw. 13, 1063–1074 (2000)
Narasimhan, S., Shah, S.L.: Model identification and error covariance matrix estimation from noisy data using PCA. Control Eng. Pract. 16, 146–155 (2008)
Fuller, W.A.: Measurement Error Models. Wiley, Chichester (1979)
Chan, N.N., Mak, T.F.: Estimating of linear functional relationships. Biometrika 70, 263–267 (1983)
Davies, M.: Identifiability issues in noisy ICA. IEEE Signal Process. Lett. 11, 470–473 (2004)
Gävert, H., Hurri, J., Särelä, J., Hyvärinen, A.: The FastICA Matlab Package, http://research.ics.tkk.fi/ica/fastica (2005)
Cichocki, A., Amari, S., Siwek, K., Tanaka, T. Phan, A. H. et al.: ICALAB Toolboxes, http://www.bsp.riken.jp/ICALAB (2007)
Lv, Q., Zhang, X.D.: A unified method for blind separation of sparse sources with unknown source number. IEEE Signal Process. Lett. 13, 49–51 (2006)
Wentzell, P.D., Andrews, D.T.: Maximum likelihood multivariate calibration. Anal. Chem. 69, 2299–2311 (1997)
Bhatt, N.P., Mitna, A., Narasimhan, S.: Multivariate calibration of non-replicated measurements for heteroscedastic errors. Chemom. Intell. Lab. Syst. 85, 70–81 (2007)
Author information
Authors and Affiliations
Corresponding author
Appendix: The IPCA algorithm
Appendix: The IPCA algorithm
The algorithm for estimating the source subspace using IPCA is given below. We use [U,S,V] = svd(X,m) to denote the truncated singular value decomposition of a m × N data matrix X, where U is the m × m matrix of left singular vectors corresponding to the non-zero singular values, S is a m × m diagonal matrix containing the non-zero singular values ordered from the largest to the smallest along the diagonal, and V is a m × N matrix of right singular vectors.
-
Step 1. Set iteration counter k = 1 and λ 0 to be zero.
-
Step 2. Set estimates of the non-zero elements of \( \Upsigma_{\varepsilon }^{k} \) to be a small fraction (say 0.0001) of the corresponding elements of S x .
-
Step 3. Obtain the transformed matrix X s = L −1 X where LL T = \( \Upsigma_{\varepsilon }^{k} \) .
-
Step 4. Let [U,S,V] = svd(X s ,m). Obtain estimate, \( M^{k} = U_{n + 1 \ldots \ldots m}^{T} L, \) where \( U_{n + 1 \ldots \ldots m} \) is the sub-matrix of U corresponding last m–n columns.
-
Step 5. Let λ k be the sum of the last m–n singular values. Stop if relative change in λ is less than the specified tolerance; else continue.
-
Step 6. Obtain the solution for the non-zero elements of the error covariance matrix by solving the optimization problem given by (14). Denote the solution as \( \Upsigma_{\varepsilon }^{k + 1} \).
-
Step 7. Increment iteration counter k and return to step 3.
It is to be noted that in order to apply above algorithm, an initial guess of the number of sources has to be specified. In order to determine the actual number of sources, we can iteratively apply above algorithm for different guesses of the number of sources as described in the text and check whether the converged singular values satisfy the required theoretical condition.
Rights and permissions
About this article
Cite this article
Ramesh Babu, P., Narasimhan, S. Multivariate techniques for preprocessing noisy data for source separation using ICA. Int J Adv Eng Sci Appl Math 4, 32–40 (2012). https://doi.org/10.1007/s12572-012-0065-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12572-012-0065-z