Fractionally spaced decision feedback equalizer based on decorrelation algorithm

To speed up the convergence rate and reduce the residual mean square error, fractionally spaced decision feedback equalizer based on decorrelation algorithm is presented in this paper. The equalizer structure contains two finite impulse response (FIR) filter sections. The input to the feed-forward part is the channel output, and it is a single-input and more-output multi-channel system model which is deduced from over-sampled received signals system model. Its coefficients are estimated by channel output decorrelation characteristics and second-order statistics of the received signals. The input to the feedback section is the output of decision equipment. Its coefficients are estimated by constant modulus algorithm (CMA), which can equalize the channel spectral nulls. The proposed algorithm has the advantages of fast convergence rate and low residual mean square error. The efficiency of the method is proved by computer simulations in underwater acoustic channels.