Blind Identification and Equalization of CDMA Signals Using the Levenvberg-Marquardt Algorithm

In this paper we describe the Levenvberg-Marquardt
(LM) algorithm for identification and equalization of CDMA
signals received by an antenna array in communication channels.
The synthesis explains the digital separation and equalization of
signals after propagation through multipath generating intersymbol
interference (ISI). Exploiting discrete data transmitted and three
diversities induced at the reception, the problem can be composed
by the Block Component Decomposition (BCD) of a tensor of
order 3 which is a new tensor decomposition generalizing the
PARAFAC decomposition. We optimize the BCD decomposition by
Levenvberg-Marquardt method gives encouraging results compared to
classical alternating least squares algorithm (ALS). In the equalization
part, we use the Minimum Mean Square Error (MMSE) to perform
the presented method. The simulation results using the LM algorithm
are important.

• Mohammed Boutalline
• Imad Badi
• Belaid Bouikhalene
• Said Safi

• Identification and equalization
• communication channel
• Levenvberg-Marquardt
• tensor decomposition

[1] L. D. Lathauwer, “Decompositions of a higher-order tensor in block
terms - part 1 : Lemmas for partitioned matrices,” SIAM J. Matrix Anal.
Appl, USA, 2006.
[2] R. A. Harshman, “Foundations of the parafac procedure : Model and
conditions for an ’explanatory’ multi-mode factor analysis,” UCLA
Working, vol. Papers in Phonetics, 16 :, pp. 1–84, 1970.
[3] N. D. Sidiropoulos, G. B. Giannakis, and R. Bro, “Blind parafac
receivers for ds-cdma systems,” Signal Processing, IEEE Transactions
on, vol. 48, no. 3, pp. 810–823, 2000.
[4] D. Nion and L. D. Lathauwer, “A block factor analysis based receiver
for blind multi-user access in wireless communications,” In Proc. IEEE
Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP), pp.
825–828, 2006.
[5] L. D. Lathauwer, “Decompositions of a higher-order tensor in block
terms - part 2 : Definitions and uniqueness,” SIAM J. Matrix Anal. Appl,
USA, 2006.
[6] L. D. Lathauwer and D. Nion, “Decompositions of a higher-order tensor
in block terms - part 3 : Alternating least squares algorithms,” SIAM J.
Matrix Anal. Appl, USA, 2006.
[7] D. Marquardt, “An algorithm for least-squares estimation of non-linear
parameters,” Technical University of Denmark, pp. 431–441, 1963.
[8] C. F. V. L. Gene H. Golub, Matrix Computationsn. Johns Hopkins
University Press, 3rd edition, October 15, 1996.
[9] N. J. Higham, Accuracy and Stability of Numerical Algorithms. Soc
for Industrial & Applied Math, 1 edition, 1996.
[10] J. J. More, “The levenberg-marquardt algorithm: implementation and
theory,” Springer, Berlin, vol. 630, pp. 105–116, 1977.[11] K. Levenberg, “A method for the solution of certain problems in least
squares,” Quart. Appl. Math., vol. 2, pp. 164–168, 1944.
[12] P. E. Gill and W. Murray, “Algorithms for the solution of the nonlinear
least-squares problem,” SIAM J. Numer. Anal, vol. vol. 15, no 5, pp.
977–992, 1978.
[13] H. Gavin, “The levenberg-marquardt method for nonlinear least
squares curve-fitting problems,” Department of Civil and Environmental
Engineering, Duke University, 2011.
[14] M. K. Transtrum and J. P. Sethna, “Improvements to the
levenberg-marquardt algorithm for nonlinear least-squares
minimization,” Journal of Computational Physics, January 30,
2012.