 Research
 Open Access
CP decomposition approach to blind separation for DSCDMA system using a new performance index
 Awatif Rouijel^{1}Email author,
 Khalid Minaoui^{1},
 Pierre Comon^{2} and
 Driss Aboutajdine^{1}
https://doi.org/10.1186/168761802014128
© Rouijel et al.; licensee Springer. 2014
 Received: 15 January 2014
 Accepted: 6 August 2014
 Published: 15 August 2014
Abstract
In this paper, we present a canonical polyadic (CP) tensor decomposition isolating the scaling matrix. This has two major implications: (i) the problem conditioning shows up explicitly and could be controlled through a constraint on the socalled coherences and (ii) a performance criterion concerning the factor matrices can be exactly calculated and is more realistic than performance metrics used in the literature. Two new algorithms optimizing the CP decomposition based on gradient descent are proposed. This decomposition is illustrated by an application to directsequence code division multiplexing access (DSCDMA) systems; computer simulations are provided and demonstrate the good behavior of these algorithms, compared to others in the literature.
Keywords
 CP decomposition
 Tensor
 DSCDMA
 Blind separation
 Optimization
1 Introduction
Blind source separation consists in estimating unknown signals observed from their mixture without knowing any information about them, except mild properties such as their independence. Early work on blind source separation was initiated by Jutten and Hérault [1, 2] in the case of an instantaneous mixture. More recently, the use of multilinear algebra methods has attracted attention in several areas such as data mining, signal processing, and particularly in wireless communication systems, among others. Wireless communication data can sometimes be viewed as components of a highorder tensor (order strictly larger than 2). Solving the problem of source separation then means finding a decomposition of this tensor and determining its parameters. One of the most popular tensor decompositions is the canonical polyadic decomposition (CP), also known as parallel factor analysis (PARAFAC), which can be seen as an analog of the matrix singular value decomposition (SVD), since it decomposes the tensor into a sum of rankone components [3–5]. This decomposition has been exploited and generalized in several works for solving different signal processing problems [6, 7] such as multiarray multisensor signal processing. The interest of the CP decomposition lies in its uniqueness under certain conditions. Typical algorithms for finding the CP components include alternating least squares (ALS) and descent algorithms [8, 9], which do not isolate the scaling factor matrix. Herein, we propose two new optimization algorithms for CP tensor decomposition, which isolate the scaling matrix in the optimization process and offers the possibility to monitor the conditioning.
It is well known that loading matrices are identified up to column scaling. This indeterminacy is complicated to take into account, given that the product of all scaling matrices must be equal to the identity. For this reason, only approximate performance indices have been used so far by ignoring the last constraint. However, one can ask oneself whether it is possible to calculate the exact performance index: this is our second contribution. The present paper develops preliminary results appeared in [10] and includes performances obtained in the frame of directsequence code division multiplexing access (DSCDMA) blind multiuser detection and estimation.
The rest of this paper is organized as follows. Section 2 presents notation, definitions, and properties of thirdorder tensors, and the exact CP decomposition problem is then stated. In Section 3, the lowrank approximation is formulated. Existence and uniqueness of this decomposition are also investigated. ALS and the two proposed algorithms are presented in Section 4. Section 5 is dedicated to the new performance criterion with a focus on an exact performance index calculation. In Section 6, we show the usefulness of our algorithms and the performances obtained. An application of these algorithms to CDMA transmission is then provided to illustrate the effectiveness of the latter.
2 Notation and preliminaries
2.1 Notations and definitions
Let us first introduce some essential notation. Scalars are denoted by lowercase letters, e.g., a. Vectors are denoted by boldface lowercase letters, e.g., a; matrices are denoted by boldface capital letters, e.g., A. Higherorder tensors are denoted by boldface Euler script letters, e.g., . The p th column of a matrix A is denoted a_{ p }, the (i,j) entry of a matrix A is denoted by A_{ ij }, and element (i,j,k) of a thirdorder tensor is denoted by T_{ ijk }. 1 will represent a vector containing ones, and I the identity matrix.
Definition 2.1
where ${X}_{{i}_{1},\cdots \phantom{\rule{0.3em}{0ex}},\phantom{\rule{0.3em}{0ex}}{i}_{N}}$ is the (i_{1},⋯, i_{ N }) elements of the N^{ t h } order of tensor.
Definition 2.2.
The symbol ⊗ represents the tensor outer product. The outer product of two tensors is another tensor, the order of which is given by the sum of the orders of the two former tensors. Equation 2 is a generalization of the concept of outer product of two vectors, which yields itself a matrix (secondorder tensor). The outer product of three vectors $\mathbf{a}\in {\mathbb{C}}^{I}$ and $\mathbf{b}\in {\mathbb{C}}^{\phantom{\rule{0.3em}{0ex}}J}$ and $\mathbf{c}\in {\mathbb{C}}^{K}$ yields a thirdorder decomposable tensor $\mathcal{Z}={\mathbf{a}}_{{}^{\otimes}}{\mathbf{b}}_{{}^{\otimes}}\mathbf{c}\in {\mathbb{C}}^{I\times J\times K}$ where Z_{ ijk }=a_{ i }b_{ j }c_{ k }.
Definition 2.3.
The rank of an arbitrary tensor $\mathcal{T}\in {\mathbb{C}}^{{I}_{1}\times {I}_{2}\times \dots \times {I}_{N}}$, denoted by R = r a n k(X), is the minimal number of rank1 tensors that yield in a linear combination.
Decomposable tensors have thus a rank equal to 1.
Definition 2.4.
The Kruskal rank, or krank , of a matrix is the largest number j such that every set of j columns is linearly independent.
Definition 2.5.
Definition 2.6.
where ⊠ is the Kronecker product.
Definition 2.7.
Definition 2.8.
where K = I J.
2.2 Preliminaries
where ${\mathbf{\text{T}}}_{1}^{I,\phantom{\rule{0.3em}{0ex}}\mathit{\text{KJ}}}\left(\text{resp.}\phantom{\rule{2.77626pt}{0ex}}{\mathbf{\text{T}}}_{2}^{J,\phantom{\rule{0.3em}{0ex}}\mathit{\text{KI}}}\phantom{\rule{2.77626pt}{0ex}}\text{and}\phantom{\rule{2.77626pt}{0ex}}{\mathbf{\text{T}}}_{3}^{K,\phantom{\rule{0.3em}{0ex}}\mathit{\text{JI}}}\right)$ is the matrix of size I × K J (resp. J × K I and K × J I) obtained by unfolding the array of size I × J × K in the first mode (resp. the second mode and the third mode), and Λ is the R × R diagonal matrix defined as Λ= D i a g{λ_{1},…, λ_{ R }}; see [5] for further details on matrix unfoldings.
The explicit handwriting of decomposable tensors as given in (8) is subject to scale indeterminacies. In the tensor literature, optimization of the CP decomposition (8) has been carried out without isolating the scaling factor Λ, which is generally included in one of the loading matrices, so that Λ= I. In [5], Kolda and Bader proposed to reduce the indeterminacies by normalizing the vectors and storing the norms in Λ. Our first proposal is to pull the factors λ_{ r } outside the product and calculate the optimal value of the scaling factor, which permits to monitor the conditioning of the problem. Scaling indeterminacies are then clearly reduced to unit modulus but are not completely fixed, hence the difficulty in estimating the identification error of loading matrices A, B and C. Our second proposal (Section 5) is to calculate the 3R complex phases (reducing to signs in the real case).
3 Existence and uniqueness
For uniqueness, Harshman has shown that is sufficient to have A and B of full rank, and C of krank ≥ 2 [3]. When 1 < R ≤ 2, the Kruskal and Harshman conditions are equivalent. For R > 2, Kruskal’s condition may be satisfied even when Harshman’s are not, and this condition is claimed to be only sufficient for R > 3 [14]. However, observations are actually corrupted by noise, so that (8) does not hold exactly.
3.1 Lowrank approximation
3.2 Conditioning of the problem
In view of matrix G, we can see that coherences play a role in the conditioning of the problem. From Equations 11 to 12, and since diagonal entries of G are equal to 1, it is indeed clear that imposing cross scalar products of the form ${\mathbf{a}}_{p}^{H}{\mathbf{a}}_{q}$ to have a modulus strictly smaller than 1 will lead with greater chances to an acceptable conditioning. Also note that scalar products do not appear individually in (12) but through their products, since entries of G can also be written as ${G}_{\mathit{\text{pq}}}={\mathbf{a}}_{p}^{H}{\mathbf{a}}_{q}\phantom{\rule{0.3em}{0ex}}{\mathbf{b}}_{p}^{H}{\mathbf{b}}_{q}\phantom{\rule{0.3em}{0ex}}{\mathbf{c}}_{p}^{H}{\mathbf{c}}_{q}$. This statement has deeper implications, particularly in existence and uniqueness of the solution to Problem (10), as subsequently elaborated.
3.3 Existence
According to the results in [15, 16], the infimum of (10) may not be reachable. In fact, the set of tensors of rank at most ξ is not closed if ξ > 1. Examples of the lack of closeness have been provided in the literature [15, 16], which suffice to prove it. In other words, it may happen that for a given tensor, and for any rankr approximation of it, there always exists another better rankr approximation.
the infimum of (10) is reached. It may be seen that this condition already gives a quantitative bound to the conditioning of (11) because coherences bound extradiagonal entries of matrix G, which has ones on its diagonal.
3.4 Uniqueness
The uniqueness of the tensor decomposition can be ensured by using a sufficient condition based on Kruskal’s theorem (9), previously mentioned.
4 Optimization for CP decomposition
In Section 3, we presented CP for threeway tensors. Various optimization algorithms exist to compute CP decomposition without constraint, as ALS or descent algorithms [7, 8, 19]. We subsequently present optimization algorithms to compute the CP decomposition (10), under the constraints of unit norm columns of loading matrices.
4.1 Alternating least squares algorithm
The ALS algorithm was proposed for CP computation by Carroll and Chang in [20] and Harshman in [3] and still stays the workhorse algorithm today, mainly owing to its ease of implementation [21]. ALS is hence the classical solution to minimize the cost function (10), despite its lack of convergence proof. This iterative algorithm alternates among the estimation of matrices A, B, and C.
where T^{I, K J} is the unfolding matrix of size I × J K defined in Section 2.2, and ()^{ † } is the MoorePenrose pseudo inverse. By symmetry, the expressions are similar for $\hat{\mathbf{B}}$ and $\hat{\mathbf{C}}$.
4.2 Proposed algorithms
Therefore, we need to find three matrices A, B, and C of unit norm columns which minimize (16). Stack these three matrices in a I+J+K by R matrix denoted by X. The objective can now be also written Υ(X,Λ), for the sake of convenience.
where M of size R × R and N of size I × R. The gradient of Υ with respect to B and C is similar, taking into account the fact that matrices M and N need to be defined accordingly (for the gradient of Υ with respect to B and C, the dimension of matrix N is J × R and K × R, respectively, while the dimension of M is always R × R).
The difficulty that arises in constrained optimization is to make sure that the move remains within the feasible set, , defined by the constraints. In the following subsections, we propose two versions of our algorithm, with two different ways of calculating scale factor Λ.
Descent algorithms are also determined by the steps that will be executed in the chosen direction. There are various methods for the step selection, and the most widely used are Backtracking and Armijo[22]. To compute the stepsize ℓ(k) in Algorithm 1 and Algorithm 2, we use a popular inexact line search method, very simple and quite effective, which is the backtracking line search. It depends on two fixed constants α, β with 0 < α < 0.5 and 0 < β < 1.
 1.
Given a descent direction D for Υ, α ∈ (0,0.5), β ∈ (0,1).
 2.
Initialization: ℓ = 1.
 3.
while Υ (X +ℓ D;Λ) > Υ(X;Λ)+α ℓ∇Υ ^{ T }D
 4.
ℓ = β ℓ.
4.2.1 Algorithm 1
4.2.2 Algorithm 2
4.2.2.0 Stopping criterion
The convergence is usually considered to be obtained at the k th iteration when the error between tensor , and the tensor reconstructed from the estimated loading matrices, does not significantly change between iterations k and k + 1.
However, in the complex case, the phase of the entries of loading matrices found at the end of the algorithm – as defined by the stopping criterion above  is different from the original. To remedy this problem, we propose a new performance criterion in order to minimize the distance between the original and the estimated matrices. Although this criterion is not usable when actual loading matrices are unknown, it still permits to assess the performances effectively attained.
5 Performance criterion
Our contribution herein consists in finding the exact minimum distance (23) under this angular constraint, by calculating the 3R optimal phases affecting the columns of the estimated loading matrices.
where Π is the set of permutations of {1,2,⋯, R}. When the permutation acts in too large dimension, greedy versions are possible to limit the exhaustive search in the permutation set.
 1.
For 1 ≤ i ≤ R! do
 2.
Calculate the 3R optimal phases affecting the columns of the estimated loading matrices:
 (a)
Permute the columns of the three estimated matrices according to the permutation π(i): ${\hat{\mathbf{A}}}_{\pi \left(i\right)}$, ${\hat{\mathbf{B}}}_{\pi \left(i\right)}$, and ${\hat{\mathbf{C}}}_{\pi \left(i\right)}$;
 (b)
For each r th columns of loading matrices and estimated matrices, do:

Set x = φ −α and y = ψ −β and solve the polynomial of degree 6 in a single variable φ:$\begin{array}{lcr}{c}_{0}& +\hfill & \phantom{\rule{0.3em}{0ex}}{c}_{1}cos\left(2x\right)+{c}_{2}\stackrel{2}{cos}\left(2x\right)+{c}_{3}\stackrel{3}{cos}\left(2x\right)\hfill \\ +& \phantom{\rule{0.3em}{0ex}}{c}_{4}\stackrel{4}{cos}\left(2x\right)\phantom{\rule{0.3em}{0ex}}+\phantom{\rule{0.3em}{0ex}}{c}_{5}\stackrel{5}{cos}\left(2x\right)\phantom{\rule{0.3em}{0ex}}+\phantom{\rule{0.3em}{0ex}}{c}_{6}\stackrel{6}{cos}\left(2x\right)=0.\end{array}$

Replace x in $siny=\frac{{\rho}_{\mathbf{a}}}{{\rho}_{\mathbf{b}}}sinx$, obtain y and consequently ψ;

Calculate χ in: exp(ȷ(φ +ψ +χ)) = 1;

Calculate the minimum distance δ:$\begin{array}{l}\delta \phantom{\rule{0.3em}{0ex}}\left(\mathbf{\text{x}};\widehat{\mathbf{\text{x}}}\right)=\underset{\phi ,\psi ,\chi}{min}\left\{{\u2225\mathbf{a}{e}^{\mathrm{\u0237\phi}}\widehat{\mathbf{a}}\u2225}^{2}+{\u2225\mathbf{b}{e}^{\mathrm{\u0237\psi}}\widehat{\mathbf{b}}\u2225}^{2}\right.\\ \phantom{\rule{6em}{0ex}}\left(\right)close="\}">+{\u2225\mathbf{c}{e}^{\mathrm{\u0237\chi}}\widehat{\mathbf{c}}\u2225}^{2}& .\end{array}$

Save the results: distance(i) = δ, phase_{ φ }(i) = φ, phase_{ ψ }(i) = ψ and phase_{ χ }(i) = χ;
 3.
End do.
 4.
Choose the 3R angles which return the smaller distance δ.
6 Simulation results
To evaluate the efficiency and behavior of the proposed algorithms with new performance criterion, two experiments are made: the first one for random loading matrices and the second one for DSCDMA system. In all experiments, the results are obtained from 100 Monte Carlo runs. At each iteration and for every SNR value, a new noise realization is drawn. The stopping criterion chosen for all experiments is Υ^{(n)}<ε and Υ^{(n)}−Υ^{(n−1)}<ε, where ε is a threshold by which the global minimum is considered to be reached, and n is the current iteration. In the following simulations, we take: ε=10^{−6}.
6.1 Example 1: random loading matrices
6.2 Example 2: application To DSCDMA system
In this example, we place ourselves in a blind context. We assume that the receiver has no knowledge neither on the spreading codes nor on symbol sequences. Classically, telecommunications blind techniques are based on some a priori knowledge, such as temporal properties of transmitted signals or the spatial properties of the receiver [25–27].
Recently, algebraic tensor methods have received considerable attention in signal processing [2]. It also turns out that multilinear algebra tools are often more powerful than their matrix equivalent. Sidiropoulos et al. are the first to adopt tensor approaches in the telecommunications field in 2000 [12]. They observed that the samples of a CDMA signal received by an array of antennas can be stored in a cube, each dimension corresponding to a diversity (coding diversity, temporal diversity, and spatial diversity). Thus, they showed that the deterministic blind separation problem of CDMA signals can be solved by the CP decomposition [28].
In this example, we propose to apply the CP decomposition algorithms as detailed in Section 4 with the new performance criterion on the DSCDMA technique. A comparison with the ALS algorithm is then made.
6.2.1 Tensor modeling
We consider R users with one transmitting antenna, transmitting simultaneously their signals to an array of K receiving antennas. In other words, we consider a communication system of type ‘multiuser SIMO’.
where the complex scalar A_{ kr } = β_{ r }a_{ k } (θ_{ r }), with β_{ r } is the fading coefficient of the r th user and a_{ k } (θ_{ r }) the response of the antenna k at the angle of arrival θ_{ r }.
6.2.2 Simulation
In this experiment, we present the performance of the receiver algorithms (Algorithm 1 and Algorithm 2 with new performance criterion) which estimate blindly the symbol S.
Angles of arrival for four users
Angles of arrival  

Figure 4  (60, 30, 0, 20) 
Angles of arrival for three scenarios
Angles of arrival  

Curve 1  (10, 20) 
Curve 2  (60, 40, 0, 20) 
Curve 3  (60,50, 10, 40, 80) 
7 Conclusions
In this paper, we have shown in Section 3.2 that, in CP tensor decompositions, the scale matrix Λ takes as optimal value a Gram matrix controlling the conditioning of the problem. This shows that bounding coherences would allow to ensure a minimal conditioning. We have described several algorithms able to compute the minimal polyadic decomposition of threeway arrays. The two proposed algorithms Algorithm 1 and Algorithm 2 have been described and tested, which involve a separate explicit calculation of the scale matrix Λ. Contrary to the performance measures used in the literature, which are optimistic by construction, the performance index calculated herein is more realistic by taking into account the angular constraint. An application of the CP decomposition with exact performance criterion to DSCDMA system has been presented. Finally, computer simulations have been performed in the context of SIMOCDMA system, in order to demonstrate both the good performances of the proposed algorithms, compared to ALS one and their usefulness in CDMA system. The judgment of our algorithms do not solely rely on the reconstruction error and the convergence speed, but it also takes into account the error in the loading matrices obtained and the BER in the case of the CDMA application.
Appendices
Appendix 1
Detailed Λ optimal calculation
Appendix 2
Performance criterion details
Solving the sixth degree equation yields x. Replacing x in $siny=\frac{{\rho}_{\mathbf{a}}}{{\rho}_{\mathbf{b}}}sinx$ yields y.
Declarations
Authors’ Affiliations
References
 Jutten C, Hérault J: Blind separation of sources, part I: an adaptive algorithm based on neuromimetic architecture. Signal Process 1991, 24: 120. 10.1016/01651684(91)90079XView ArticleMATHGoogle Scholar
 Comon P, Jutten C: Handbook of Blind Source Separation, Independent Component Analysis and Applications. Academic Press, Oxford UK, Burlington USA; 2010. ISBN: 9780123747266Google Scholar
 Harshman RA: Foundations of the Parafac procedure: models and conditions for an “explanatory” multimodal factor analysis. UCLA working papers in phonetics, (University Microfilms, Ann Arbor, Michigan, No. 10,085) 1970, 16: 184.Google Scholar
 Comon P: Tensors: a brief introduction. IEEE Sig. Proc. Mag 2014, 31(3):4453. Special issue on BSS. hal00923279View ArticleGoogle Scholar
 Kolda TG, Bader BW: Tensor decompositions and applications. SIAM Rev 2009, 51(3):455500. 10.1137/07070111XMathSciNetView ArticleMATHGoogle Scholar
 Comon P: Tensor decompositions, state of the art and applications. In IMA Conf. Mathematics in Signal Processing. Clarendon press,, Oxford, UK; 2000:124.Google Scholar
 Almeida ALFD, Favier G, Mota JCM: The constrained trilinear decomposition with application to MIMO wireless communication systems. In GRETSI’07. Colloque GRETSI. Troyes; 2007.Google Scholar
 Comon P, Luciani X, De Almeida ALF: Tensor decompositions, alternating least squares and other tales. J. Chemometrics 2009, 23: 393405. 10.1002/cem.1236View ArticleGoogle Scholar
 Sorber L, Van Barel M, De Lathauwer L: Optimizationbased algorithms for tensor decompositions: canonical polyadic decomposition, decomposition in ranks terms and a new generalization. SIAM J. Optimization 2013, 23(2):695720. 10.1137/120868323MathSciNetView ArticleMATHGoogle Scholar
 Comon P, Minaoui K, Rouijel A, Aboutajdine D: Performance index for tensor polyadic decompositions. In 21th EUSIPCO Conference. IEEE,, Marrakech, Morocco; 2013.Google Scholar
 Berge JMFT, Sidiropoulos N: On uniqueness in candecomp/parafac. Psychometrika 2002, 67(3):399409. 10.1007/BF02294992MathSciNetView ArticleMATHGoogle Scholar
 Sidiropoulos ND, Bro R, Giannakis GB: Parallel factor analysis in sensor array processing. IEEE Trans. Sig. Proc 2000, 48(8):23772388. 10.1109/78.852018View ArticleGoogle Scholar
 Kruskal JB: Threeway arrays: rank and uniqueness of trilinear decompositions. Linear Algebra Appl 1977, 18: 95138. 10.1016/00243795(77)900696MathSciNetView ArticleMATHGoogle Scholar
 ten Berge JMF, Sidiropoulos ND: On uniqueness in candecomp/parafac. Psychometrika 2002, 67(3):399409. 10.1007/BF02294992MathSciNetView ArticleMATHGoogle Scholar
 Lim LH, Comon P: Blind multilinear identification. IEEE Trans. Inf. Theory 2014, 60(2):12601280.MathSciNetView ArticleGoogle Scholar
 Comon P, Lim LH: Sparse representations and lowrank tensor approximation. Research Report ISRN I3S//RR201101FR, I3S, SophiaAntipolis, France (February 2011.) http://hal.archivesouvertes.fr/docs/00/70/34/94/PDF/RR1102P.COMON.pdf
 Lim LH, Comon P: Multiarray signal processing: tensor decomposition meets compressed sensing. CompteRendus Mécanique de l’Academie des Sci 2010, 338(6):311320.View ArticleMATHGoogle Scholar
 Gribonval R, Nielsen M: Sparse representations in unions of bases. IEEE Trans. Inf. Theory 2003, 49(13):33203325.MathSciNetView ArticleMATHGoogle Scholar
 Acar E, Dunlavy DM, Kolda TG: A scalable optimization approach for fitting canonical tensor decompositions. J. Chemometrics 2011, 25: 6786. 10.1002/cem.1335View ArticleGoogle Scholar
 Carroll J, Chang J: Analysis of individual differences in multidimensional scaling via an nway generalization of “eckartyoung” decomposition. Psychometrika 1970, 35: 283319. 10.1007/BF02310791View ArticleMATHGoogle Scholar
 Smilde A, Bro R, Geladi P: MultiWay Analysis. (Wiley, Chichester UK; 2004.Google Scholar
 Boyd S, Vandenberghe L: Convex Optimization. Cambridge University Press, the United States of America, New York; 2004. ISBN: 978521833783 hardbackView ArticleMATHGoogle Scholar
 Comon P: Estimation multivariable complexe. Traitement du Signal 1986, 3(2):97101.Google Scholar
 Hjorungnes A, Gesbert D: Complex valued matrix differentiation: techniques and key results. IEEE Trans. Signal Process 2007, 55(6):27402746.MathSciNetView ArticleGoogle Scholar
 Moulines E, Duhamel P, Cardoso JF, Mayrargue S: Subspace methods for the blind identification of multichannel fir filters. IEEE Trans. Signal Process 1995, 43: 516525. 10.1109/78.348133View ArticleGoogle Scholar
 Godard DN: Selfrecovering equalization and carrier tracking in twodimensional data communication systems. IEEE Trans. Commun 1980, 28: 18671875. 10.1109/TCOM.1980.1094608View ArticleGoogle Scholar
 der Veen AJV, Paulraj A: An analytical constant modulus algorithm. IEEE Trans. Signal Proc 1996, 44: 11361155. 10.1109/78.502327View ArticleGoogle Scholar
 de Almeida ALF, Favier G, Mota JCM: Parafacbased unified tensor modeling for wireless communication systems with application to blind multiuser equalization. Signal Process 2007, 87(2):337351. 10.1016/j.sigpro.2005.12.014View ArticleMATHGoogle Scholar
Copyright
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.