Sequential sequence estimation of bandlimited signals over a frequency selective Rayleigh channel

The sequential sequence estimator is based on a set of whitening Kalman filters. It has considerably less computational complexity with an error performance almost identical to that of the maximum likelihood sequence estimation. Due to the presence of both fading and ISI, the error probability is still poor. To improve the error performance, diversity transmission is investigated. The error performance and computational complexity are simulated by computer for two different fading channel models. Comparisons with analytical predictions show good agreement.<<ETX>>

[1]  B. Anderson,et al.  Optimal Filtering , 1979, IEEE Transactions on Systems, Man, and Cybernetics.