Blind channel estimation: comparison studies and a new algorithm

Blind estimation techniques have the potential of significantly increasing the transmission efficiency for time-varying channels. A performance analysis of some blind channel estimation algorithms is presented. It is shown that ill-conditioned channels and the local minima of optimization techniques are the main causes of estimation error. To overcome these obstacles, a new optimization criterion that integrates two key aspects of the channel characteristics is introduced. Furthermore, the statistical characteristics of the channel are utilized by exploiting the principal component structures. The proposed algorithm based on this new criterion includes several existing algorithms as special cases. The new algorithm offers significant saving in computation and improvement in performance.

[1]  Dirk T. M. Slock,et al.  Blind fractionally-spaced equalization, perfect-reconstruction filter banks and multichannel linear prediction , 1994, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing.

[2]  W. Gardner.  Exploitation of spectral redundancy in cyclostationary signals , 1991, IEEE Signal Processing Magazine.

[3]  Eric Moulines,et al.  Subspace methods for the blind identification of multichannel FIR filters , 1995, IEEE Trans. Signal Process..

[4]  Lang Tong,et al.  A deterministic approach to blind equalization , 1993, Proceedings of 27th Asilomar Conference on Signals, Systems and Computers.

[5]  Lang Tong,et al.  A new approach to blind identification and equalization of multipath channels , 1991, [1991] Conference Record of the Twenty-Fifth Asilomar Conference on Signals, Systems & Computers.

[6]  Lang Tong,et al.  Blind identification and equalization based on second-order statistics: a time domain approach , 1994, IEEE Trans. Inf. Theory.

[7]  Y. Li,et al.  Blind channel identification based on second order cyclostationary statistics , 1993, 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing.