A Cost Function for Joint Blind Equalization and Phase Recovery
In this paper a new cost function for blind equalization
is proposed. The proposed cost function, referred to as the modified
maximum normalized cumulant criterion (MMNC), is an extension
of the previously proposed maximum normalized cumulant criterion
(MNC). While the MNC requires a separate phase recovery system
after blind equalization, the MMNC performs joint blind equalization
and phase recovery. To achieve this, the proposed algorithm
maximizes a cost function that considers both amplitude and phase of
the equalizer output. The simulation results show that the proposed
algorithm has an improved channel equalization effect than the MNC
algorithm and simultaneously can correct the phase error that the
MNC algorithm is unable to do. The simulation results also show that
the MMNC algorithm has lower complexity than the MNC algorithm.
Moreover, the MMNC algorithm outperforms the MNC algorithm
particularly when the symbols block size is small.
[1] J.A. Cadzow, "Blind deconvolution via cumulant extrema," IEEE Signal
Processing Magazine, vol. 13, no. 3, pp. 24 - 42, May 1996.
[2] C.-Y. Chi and M.-C. Wu, "Inverse filter criteria for blind deconvolution
and equalization using two cumulants," Signal Processing, vol. 43, no.
1, pp. 55 - 63, Apr. 1995.
[3] C.-C. Feng and C.-Y. Chi, "Performance of cumulant based inverse
filters for blind deconvolution, " IEEE Trans. Signal Processing, vol. 47,
no. 7, July 1999.
[4] C.-C. Feng and C.-Y. Chi, C.-H. Chen, "Blind Equalization and System
Identification: Batch Processing Algorithms, Performance and
Applications, " Springer Publications , 2006.
[5] C.-Y. Chi, C.-Y. Chen, C.-H. Chen, C.-C. Feng, "Batch Processing
Algorithm for Blind Equalization Using Higher-Order Statistics," IEEE
Signal Processing Magazine, January 2003.
[6] M. Babaee, "Design, Simulation and Improvement of a Blind
Equalization Algorithm," Master-s thesis, Imam Hossein University,
Tehran, Iran, July 2007.
[7] 0. Shalvi and E. Weinstein, "Super-exponential methods for blind
deconvolution," IEEE Trans. Information Theory, vol. 39, no. 2, pp.
504-519, March 1993.
[1] J.A. Cadzow, "Blind deconvolution via cumulant extrema," IEEE Signal
Processing Magazine, vol. 13, no. 3, pp. 24 - 42, May 1996.
[2] C.-Y. Chi and M.-C. Wu, "Inverse filter criteria for blind deconvolution
and equalization using two cumulants," Signal Processing, vol. 43, no.
1, pp. 55 - 63, Apr. 1995.
[3] C.-C. Feng and C.-Y. Chi, "Performance of cumulant based inverse
filters for blind deconvolution, " IEEE Trans. Signal Processing, vol. 47,
no. 7, July 1999.
[4] C.-C. Feng and C.-Y. Chi, C.-H. Chen, "Blind Equalization and System
Identification: Batch Processing Algorithms, Performance and
Applications, " Springer Publications , 2006.
[5] C.-Y. Chi, C.-Y. Chen, C.-H. Chen, C.-C. Feng, "Batch Processing
Algorithm for Blind Equalization Using Higher-Order Statistics," IEEE
Signal Processing Magazine, January 2003.
[6] M. Babaee, "Design, Simulation and Improvement of a Blind
Equalization Algorithm," Master-s thesis, Imam Hossein University,
Tehran, Iran, July 2007.
[7] 0. Shalvi and E. Weinstein, "Super-exponential methods for blind
deconvolution," IEEE Trans. Information Theory, vol. 39, no. 2, pp.
504-519, March 1993.
@article{"International Journal of Electrical, Electronic and Communication Sciences:60177", author = "Reza Berangi and Morteza Babaee and Majid Soleimanipour", title = "A Cost Function for Joint Blind Equalization and Phase Recovery", abstract = "In this paper a new cost function for blind equalization
is proposed. The proposed cost function, referred to as the modified
maximum normalized cumulant criterion (MMNC), is an extension
of the previously proposed maximum normalized cumulant criterion
(MNC). While the MNC requires a separate phase recovery system
after blind equalization, the MMNC performs joint blind equalization
and phase recovery. To achieve this, the proposed algorithm
maximizes a cost function that considers both amplitude and phase of
the equalizer output. The simulation results show that the proposed
algorithm has an improved channel equalization effect than the MNC
algorithm and simultaneously can correct the phase error that the
MNC algorithm is unable to do. The simulation results also show that
the MMNC algorithm has lower complexity than the MNC algorithm.
Moreover, the MMNC algorithm outperforms the MNC algorithm
particularly when the symbols block size is small.", keywords = "Blind equalization, maximum normalized cumulant
criterion (MNC), intersymbol interference (ISI), modified MNC
criterion (MMNC), phase recovery.", volume = "1", number = "6", pages = "866-5", }