Abstract
The goal of independent component analysis (ICA) is to decompose observed signals into components as independent as possible. In linear instantaneous blind source separation, ICA is used to separate linear instantaneous mixtures of source signals into signals that are as close as possible to the original signals. In the estimation of the so-called demixing matrix one has to distinguish two different factors:
-
1.
Variance of the estimated inverse mixing matrix in the noiseless case due to randomness of the sources.
-
2.
Bias of the demixing matrix from the inverse mixing matrix:
This chapter studies both factors for circular and noncircular complex mixtures. It is important to note that the complex case is not directly equivalent to the real case of twice larger dimension. In the derivations, we aim to clearly show the connections and differences between the complex and real cases. In the first part of the chapter, we derive a closed-form expression for the CRB of the demixing matrix for instantaneous noncircular complex mixtures. We also study the CRB numerically for the family of noncircular complex generalized Gaussian distributions (GGD) and compare it to simulation results of several ICA estimators. In the second part, we consider a linear noisy noncircular complex mixing model and derive an analytic expression for the demixing matrix of ICA based on the Kullback-Leibler divergence (KLD). We show that for a wide range of both the shape parameter and the noncircularity index of the GGD, the signal-to-interference-plus-noise ratio (SINR) of KLD-based ICA is close to that of linear MMSE estimation. Furthermore, we show how to extend our derivations to the overdetermined case (\(M>N\)) with circular complex noise.
Sections 3.1.1 and 3.2 of this chapter are based on our previous journal publication [35]. © 2013 IEEE. Reprinted, with permission, from Loesch and Yang [35].
Sections 3.3.1–3.3.3 of this chapter are based on our previous conference publication [34]. First published in the Proceedings of the 20th European Signal Processing Conference (EUSIPCO-2012) in 2012, published by EURASIP.
An erratum to this chapter is available at 10.1007/978-3-642-55016-4_3
An erratum to this chapter can be found at http://dx.doi.org/10.1007/978-3-642-55016-4_20
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
See Sect. 3.1.1 for a definition.
- 2.
Examples of digital modulation schemes are phase shift keying (PSK), pulse amplitude modulation (PAM) or quadrature amplitude modulation (QAM).
- 3.
See Sect. 3.2.4 for a definition.
- 4.
For a large noise variance \(\sigma ^2\) the theoretical analysis cannot fully describe the behavior of KLD-based ICA since we only take into account terms of order \(\sigma ^2\). However, simulation results show that KLD-based ICA still performs similarly to linear MMSE estimation.
- 5.
Due to the inherent scaling ambiguity between the mixing matrix \({\mathbf A}\) and the source signals \(\mathbf{s}\), without loss of generality, we can scale \(\mathbf{s}\) and accordingly \({\mathbf A}\) such that \(\mathrm{E}\left[ |s_i|^2\right] = 1\) and \(\gamma _i \in [0,1]\).
- 6.
Some authors [5, 15, 47] prefer the so-called expected interference-to-source ratio (ISR) matrix whose elements \(\overline{\text {ISR}}_{ij}\) are defined (for \(i\ne j\) and unit variance sources) as \(\overline{\text {ISR}}_{ij}=\mathrm{E}\left[ \frac{\left| G_{ij}\right| ^2}{\left| G_{ii}\right| ^2}\right] \), where \(G_{ii}\) denotes the diagonal elements and \(G_{ij}\) the off-diagonal elements of \({\mathbf G}\). To compute \(\overline{\text {ISR}}_{ij}\), usually \(G_{ii} \approx 1\) (i.e., \({{\mathrm{var}}}(G_{ii}) \ll 1\)) is assumed such that \(\overline{\text {ISR}}_{ij} \approx {{\mathrm{var}}}(G_{ij})\). In this section, we do not use the ISR matrix but instead directly derive the iCRB for \({\mathbf G}\).
- 7.
References
Adali, T., Li, H.: A practical formulation for computation of complex gradients and its application to maximum likelihood ICA. In: Proc. IEEE Conference on Acoustics, Speech, and Signal Processing (ICASSP), vol. 2, pp. II-633-II-636 (2007)
Adali, T., Li, H.: Complex-valued adaptive signal processing, ch. 1. In: T. Adali, S. Haykin (eds.) Adaptive Signal Processing: Next Generation Solutions, pp. 1–85. Wiley, New York (2010)
Adali, T., Li, H., Novey, M., Cardoso, J.F.: Complex ica using nonlinear functions. IEEE Trans. Signal Process. 56(9), 4536–4544 (2008)
Adali, T., Schreier, P., Scharf, L.: Complex-valued signal processing: the proper way to deal with impropriety. IEEE Trans. Signal Process. 59(11), 5101–5125 (2011)
Anderson, M., Li, X.L., Rodriquez, P.A., Adali, T.: An effective decoupling method for matrix optimization and its application to the ICA problem. In: Proceedings of the IEEE Conference on Acoustics, Speech, and Signal Processing (ICASSP), pp. 1885–1888 (2012)
Brandwood, D.H.: A complex gradient operator and its application in adaptive array theory. IEE Proc. 130, 11–16 (1983)
Cardoso, J., Souloumiac, A.: Blind beamforming for non-gaussian signals. Radar Signal Process. IEE Proc. F 140(6), 362–370 (1993)
Cardoso, J.F.: On the performance of orthogonal source separation algorithms. In: Proceedings of the European Signal Processing Conference (EUSIPCO), pp. 776–779 (1994)
Cardoso, J.F.: Blind signal separation: statistical principles. Proc. IEEE 86(10), 2009–2025 (1998)
Cardoso, J.F., Adali, T.: The maximum likelihood approach to complex ICA. In: Proceedings of the IEEE Conference on Acoustics, Speech, and Signal Processing (ICASSP), vol. 5, pp. 673–676 (2006)
Cichocki, A., Sabala, I., Choi, S., Orsier, B., Szupiluk, R.: Self adaptive independent component analysis for sub-gaussian and super-gaussian mixtures with unknown number of sources and additive noise. In: Proceedings of 1997 International Symposium on Nonlinear Theory and its Applications (NOLTA-97), vol. 2, pp. 731–734 (1997)
Comon, P., Jutten, C. (eds.): Handbook of Blind Source Separation: Independent Component Analysis and Applications, 1st edn. Elsevier, Amsterdam (2010)
Davies, M.: Identifiability issues in noisy ica. IEEE Signal Process. Lett. 11(5), 470–473 (2004)
De Lathauwer, L., De Moor, B.: On the blind separation of non-circular sources. In: Proceedings of the European Signal Processing Conference (EUSIPCO), vol. 2, pp. 99–102. Toulouse, France (2002)
Doron, E., Yeredor, A., Tichavsky, P.: Cramér-Rao-induced bound for blind separation of stationary parametric gaussian sources. IEEE Signal Process. Lett. 14(6), 417–420 (2007)
Douglas, S., Cichocki, A., Amari, S.: A bias removal technique for blind source separation with noisy measurements. Eletron. Lett. 34(14), 1379–1380 (1998)
Douglas, S.C.: Fixed-point algorithms for the blind separation of arbitrary complex-valued non-gaussian signal mixtures. EURASIP J. Appl. Signal Process. 2007(1), Article ID 36,525 (2007)
Eriksson, J., Koivunen, V.: Complex-valued ICA using second order statistics. pp. 183–192 (2004)
Eriksson, J., Koivunen, V.: Complex random vectors and ICA models: identifiability, uniqueness, and separability. IEEE Trans. Inf. Theory 52(3), 1017–1029 (2006)
Fiori, S.: Neural independent component analysis by maximum-mismatch learning principle. Neural Netw. 16(8), 1201–1221 (2003)
Hjørungnes, A.: Complex-Valued Matrix Derivatives. Cambridge University Press, Cambridge (2011)
Horn, R.A., Johnson, C.R.: Matrix analysis, 1st publis. (1985), 10th print edn. Cambridge University Press, Cambridge (1999)
Hyvärinen, A.: Independent component analysis in the presence of gaussian noise by maximizing joint likelihood. Neurocomputing 22, 49–67 (1998)
Jagannatham, A., Rao, B.: Cramér-Rao lower bound for constrained complex parameters. IEEE Signal Process. Lett. 11(11), 875–878 (2004)
Joho, M., Mathis, H., Lambert, R.H.: Overdetermined blind source separation: using more sensors than source signals in a noisy mixture. In: Proceedings of the International Conference on Independent Component Analysis and Blind Source Separation (ICA), pp. 81–86 (2000)
Koldovsky, Z., Tichavsky, P.: Methods of fair comparison of performance of linear ICA techniques in presence of additive noise. In: Proceedings of the IEEE Conference on Acoustics, Speech, and Signal Processing (ICASSP), vol. 5, pp. 873–876 (2006)
Koldovsky, Z., Tichavsky, P.: Asymptotic analysis of bias of Fast ICA-based algorithms in the presence of additive noise. Technical Report 2181, UTIA, AV CR (2007a)
Koldovsky, Z., Tichavsky, P.: Blind instantaneous noisy mixture separation with best interference-plus-noise rejection. In: Proceedings of the International Conference on Independent Component Analysis and Blind Source Separation (ICA), pp. 730–737 (2007b)
Li, H., Adali, T.: Algorithms for complex ML ICA and their stability analysis using Wirtinger calculus. IEEE Trans. Signal Process. 58(12) (2010)
Li, X.L., Adali, T.: Complex independent component analysis by entropy bound minimization. IEEE Trans. Circ. Syst. I Regul. Pap. 57(7), 1417–1430 (2010)
Loesch, B.: Complex blind source separation with audio applications. Ph.D. thesis, University of Stuttgart (2013). http://www.hut-verlag.de/9783843911214.html
Loesch, B., Yang, B.: On the relation between ICA and MMSE based source separation. In: Proceedings of the IEEE Conference on Acoustics, Speech, and Signal Processing (ICASSP), pp. 3720–3723 (2011)
Loesch, B., Yang, B.: Cramér-Rao bound for circular complex independent component analysis. In: Proceedings of the International Conference on Latent Variable Analysis and Signal Separation (LVA/ICA), pp. 42–49 (2012a)
Loesch, B., Yang, B.: On the solution of circular and noncircular complex KLD-ICA in the presence of noise. In: Proceedings of the European Signal Processing Conference (EUSIPCO), pp. 1479–1483 (2012b)
Loesch, B., Yang, B.: Cramér-Rao bound for circular and noncircular complex independent component analysis. IEEE Trans. Signal Process. 61(2), 365–379 (2013)
Mandic, D.P., Goh, V.S.L.: Complex Valued Nonlinear Adaptive Filters: Noncircularity, Widely Linear, and Neural Models, 1st edn. Wiley, Chichester (2009)
Novey, M., Adali, T.: ICA by maximization of nongaussianity using complex functions. In: Proceedings of the IEEE Workshop on Machine Learning for Signal Processing (MLSP), pp. 21–26 (2005)
Novey, M., Adali, T.: Adaptable nonlinearity for complex maximization of nongaussianity and a fixed-point algorithm. In: Proceedings of the IEEE Workshop on Machine Learning for Signal Processing (MLSP), pp. 79–84 (2006)
Novey, M., Adali, T.: On extending the complex FastICA algorithm to noncircular sources. IEEE Trans. Signal Process. 56(5), 2148–2154 (2008)
Novey, M., Adali, T., Roy, A.: A complex generalized gaussian distribution—characterization, generation, and estimation. IEEE Trans. Signal Process. 58(3), 1427–1433 (2010)
Ollila, E., Kim, H.J., Koivunen, V.: Compact Cramér-Rao bound expression for independent component analysis. IEEE Trans. Signal Process. 56(4), 1421–1428 (2008)
Ollila, E., Koivunen, V., Eriksson, J.: On the Cramér-Rao bound for the constrained and unconstrained complex parameters, pp. 414–418 (2008)
Remmert, R.: Theory of Complex Functions. Graduate Texts in Mathematics. Springer, New York (1991)
Schreier, P.J., Scharf, L.L.: Statistical signal processing of complex-valued data: The theory of improper and noncircular signals. Cambridge University Press, Cambridge (2010)
Tichavsky, P., Koldovsky, Z., Oja, E.: Performance analysis of the Fast ICA algorithm and Cramér-Rao bounds for linear independent component analysis. IEEE Trans. Signal Process. 54(4) (2006)
Wirtinger, W.: Zur formalen theorie der funktionen von mehr komplexen veränderlichen. Math. Ann. 97(1), 357–375 (1927)
Yeredor, A.: Blind separation of gaussian sources with general covariance structures: bounds and optimal estimation. IEEE Trans. Signal Process. 58(10), 5057–5068 (2010)
Yeredor, A.: Performance analysis of the strong uncorrelating transformation in blind separation of complex-valued sources. IEEE Trans. Signal Process. 60(1), 478–483 (2012)
Zhang, L.Q., Cichocki, A., Amari, S.: Natural gradient algorithm for blind separation of overdetermined mixture with additive noise. IEEE Signal Process. Lett. 6(11), 293–295 (1999)
Zhu, X.L., Zhang, X.D., Ye, J.M.: A generalized contrast function and stability analysis for overdetermined blind separation of instantaneous mixtures. Neural Comput. 18(3), 709–728 (2006)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendices
Appendix 1
1.1 Values of \(\kappa \), \(\xi \), \(\beta \), \(\eta \) for Complex GGD
The pdf of a noncircular complex GGD with zero mean, variance \(\mathrm{E}[|s|^2]=1\) and noncircularity index \(\gamma \in [0,1]\) is given by
where \(\alpha = \varGamma (2/c)/\varGamma (1/c)\) and \(\varGamma (\cdot )\) is the Gamma function. The function \(\varphi (s,s^*) = - \frac{\partial }{\partial s^*} \ln p(s,s^*)\) is then given by
By integration in polar coordinates, it can be shown that \(\kappa \), \(\xi \), \(\beta \) and \(\eta \) are given by:
1.2 Induced CRB for Real ICA
Here, we briefly review the iCRB for real ICA [41, 45]. In the following, all real quantities \(q\) are denoted as \(\mathring{q}\). In the derivation of the iCRB for the real case \(\mathring{\varphi }(\mathring{s}) = - \partial \ln p(\mathring{s}) / \partial \mathring{s}\) and the parameters \(\mathring{\kappa }=E [ \mathring{\varphi }^2(\mathring{s})]\), \(\mathring{\eta }=\mathrm{E}[\mathring{s}^2 \mathring{\varphi }^2(\mathring{s})] = 2+\mathrm{E}\left[ \mathring{s}^2 \frac{\partial \mathring{\varphi }(\mathring{s})}{\partial \mathring{s}}\right] \) are defined using real derivatives. In [41, 45] it was shown that
Appendix 2
Here we derive an analytic expression for \({\mathbf W}_{\text {ICA}}\) in the presence of noise by using a perturbation analysis. Motivated by \({\mathbf W}_{\text {ICA}} \mathop {=}\limits ^{\sigma ^2=0} {\mathbf A}^{-1}\), we assume that \({\mathbf W}_{\text {ICA}}\) can be written as \({\mathbf W}_{\text {ICA}} = {\mathbf A}^{-1} + \sigma ^2 {\mathbf B} + \fancyscript{O}(\sigma ^4)\) and derive \({\mathbf B}\) by a two-step perturbation analysis:
-
1.
Taylor series approximation of \(\mathrm{E}({\pmb {\varphi }}^*({\mathbf{y}}) {\mathbf{y}}^T)\) in (3.51) at \({\mathbf{y}}={\hat{\varvec{y}}} = {{\mathbf W}}_{\text {ICA}} {{\mathbf A}} {\mathbf{s}}\),
-
2.
Taylor series approximation of the result of the above step by exploiting \({\mathbf W}_{\text {ICA}} = {\mathbf A}^{-1}+\sigma ^2 {\mathbf B} + \fancyscript{O}(\sigma ^4)\) and \(\hat{\mathbf{y}} = {\mathbf{s}} + \sigma ^2 {\mathbf B} {\mathbf A} {\mathbf{s}}+\fancyscript{O}(\sigma ^4) = {\mathbf{s}} + \sigma ^2 {\mathbf C} {\mathbf{s}}+\fancyscript{O}(\sigma ^4)={\mathbf{s}}+\sigma ^2 {\mathbf{b}} + \fancyscript{O}(\sigma ^4)\) with \({\mathbf C}={\mathbf B}{\mathbf A}\) and \(\mathbf{b} = {\mathbf C} \mathbf{s} = [b_1, \dots , b_N]^T\).
In this way, we determine explicitely the deviation \(\sigma ^2{\mathbf B}\) of \({\mathbf W}_{\text {ICA}}\) from the inverse solution \({\mathbf A}^{-1}\).
The general Taylor series expansion of \(\varphi ^*(y) \mathop {\widehat{=}} \varphi ^*(y,y^*)\) is given as
with \(\varpi (y,y^*) = \frac{\partial \varphi ^*}{\partial y}\), \(\vartheta (y,y^*) = \frac{\partial \varphi ^*}{\partial y^*}\), \(\nu (y,y^*) = \frac{\partial ^2 \varphi ^*}{(\partial y)^2}\), \(\zeta (y,y^*) = \frac{\partial ^2 \varphi ^*}{(\partial y^*)^2}\) and \(\epsilon (y,y^*) = \frac{\partial ^2 \varphi ^*}{\partial y \partial y^*}\). To simplify notation, we will drop the dependence of \(\varphi ^*(\cdot )\), \(\varpi (\cdot )\), \(\vartheta (\cdot )\), \(\nu (\cdot )\), \(\zeta (\cdot )\), \(\epsilon (\cdot )\) on \(y^*\) and keep only the dependence on \(y\) in the following.
Let
As shown in [31, 34], \({\mathbf W}_{\text {ICA}} = {\mathbf A}^{-1} + \sigma ^2 {\mathbf C}\), where the elements of \({\mathbf C}\) can be computed from
and
with the transformed noise covariance matrix \({\mathbf R}_{-1} = {\mathbf W} {\mathbf R}_{\mathbf{v}} {\mathbf W}^H ={\mathbf A}^{-1} {\mathbf R}_{\mathbf{v}} {\mathbf A}^{-H} + \fancyscript{O}(\sigma ^2)\) and the transformed noise pseudo-covariance matrix \(\bar{{\mathbf R}}_{-1}\!=\! {\mathbf W} \bar{{\mathbf R}}_{\mathbf{v}} {\mathbf W}^T = {\mathbf A}^{-1} \bar{{\mathbf R}}_{\mathbf{v}} {\mathbf A}^{-T} + \fancyscript{O}(\sigma ^2)\). Note that \({\mathbf R}_{-1}^H = {\mathbf R}_{-1}\) and \(\bar{{\mathbf R}}_{-1}^T = \bar{{\mathbf R}}_{-1}\).
If \(p(s,s^*)\) is symmetric in the real part \(\mathfrak {R}s\) or imaginary part \(\mathfrak {I}s\) of \(s\), i.e., \(p(-\mathfrak {R}s, \mathfrak {I}s)=p(\mathfrak {R}s, \mathfrak {I}s)\) or \(p(\mathfrak {R}s, -\mathfrak {I}s)=p(\mathfrak {R}s, \mathfrak {I}s)\), the parameters \(\kappa _i\), \(\rho _i\), \(\delta _i\), \(\lambda _i\), \(\xi _i\), \(\omega _i\), \(\tau _i\) are real. For \(\rho _i + 1 \pm \delta _i \ne 0\), we then get from (3.97)
For \((\gamma _j \xi _i+\kappa _i)(\gamma _i \xi _j + \kappa _j) \ne 1\) and \((\gamma _j \xi _i-\kappa _i)(\gamma _i \xi _j - \kappa _j) \ne 1\), we obtain from (3.98)
Rights and permissions
Copyright information
© 2014 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Loesch, B., Yang, B. (2014). Performance Study for Complex Independent Component Analysis. In: Naik, G., Wang, W. (eds) Blind Source Separation. Signals and Communication Technology. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-55016-4_3
Download citation
DOI: https://doi.org/10.1007/978-3-642-55016-4_3
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-55015-7
Online ISBN: 978-3-642-55016-4
eBook Packages: EngineeringEngineering (R0)