NonStationary CMA for Decision Feedback Equalization of Markovian Time Varying Channels

In this paper, we propose a modified version of the Constant Modulus Algorithm (CMA) tailored for blind Decision Feedback Equalizer (DFE) of first order Markovian time varying channels. The proposed NonStationary CMA (NSCMA) is designed so that it explicitly takes into account the Markovian structure of the channel nonstationarity. Hence, unlike the classical CMA, the NSCMA is not blind with respect to the channel time variations. This greatly helps the equalizer in the case of realistic channels, and avoids frequent transmissions of training sequences. This paper develops a theoretical analysis of the steady state performance of the CMA and the NSCMA for DFEs within a time varying context. Therefore, approximate expressions of the mean square errors are derived. We prove that in the steady state, the NSCMA exhibits better performance than the classical CMA. These new results are confirmed by simulation. Through an experimental study, we demonstrate that the Bit Error Rate (BER) is reduced by the NSCMA-DFE, and the improvement of the BER achieved by the NSCMA-DFE is as significant as the channel time variations are severe.

Authors:

• S. Cherif
• M. Turki-Hadj Alouane

Keywords:

• Time varying channel
• Markov model
• Blind DFE
• CMA
• NSCMA.

References:

[1] N. Yousef and A. Sayed, "Ability of adaptive filters to track carrier
offsets and channel non stationarities," IEEE Transactions on Signal
Processing, vol. 50, no. 7, pp. 1433-1544, July 2002.
[2] R. A. Ziegler and J. Cioffi, "Estimation of time-varying digital radio
channels," IEEE Transactions on Vehicular Technology, vol. 41, no. 2,
pp. 134-151, May 1992.
[3] B. C. Ng, D. Gesbert, and A. Paulraj, "A semi-blind approach to
structured channel equalization," in Proc. IEEE Int. Conference on ASSP,
vol. VI, 1998, pp. 3385-3388.
[4] R. Kennedy, B. Anderson, and R. Bitmead, "Blind adaptation of decision
feedback equalizers : Gross convergence properties," International
Journal of Adaptive Control and Signal Processing, vol. 7, pp. 497-523,
1993.
[5] S. Marcos, S. Cherif, and M. Jaidane, "Blind cancellation of intersymbol
interference in decision feedback equalizers," in Proc. IEEE Int.
Conference on ASSP, vol. II, 1995, pp. 1073-1076.
[6] C. Johnson, J. Schniter, and al., "Blind equalization using the constant
modulus criterion: A review," Proc. IEEE Special Issue on Blind
Identification and Equalization, 1998.
[7] G. Giannakis, C. Tepedelenlioglu, and H. Liu, "Adaptive blind equalization
of time-vatying channels," in Proc. IEEE Int. Conference on ASSP,
vol. V, 1997, pp. 4033-4036.
[8] I. Barhumi, G. Leus, and M. Moonen, "Time-varying fir equalization
for doubly selective channels," IEEE Transactions on Wireless Communications,
vol. 4, no. 1, pp. 202-214, Jan. 2005.
[9] J. Mai and A. Sayed, "A feedback approach to the steady state
performance of fractionally spaced blind adaptive equalizers," IEEE
Transactions on Signal Processing, vol. 48, no. 1, pp. 80-91, Jan. 2000.
[10] I. Fijakow, C. Manlove, and C. R. Johnson, "Adaptive fractionally spaced
blind cma equalization: Excess mse," IEEE Transactions on Signal
Processing, vol. 46, no. 1, pp. 227-231, Jan. 1998.