Method and apparatus for reducing noise correlation in partial response channel

摘要

A method and apparatus are provided for reducing noise correlation through optimization of a look-ahead maximum likelihood (ML) detector. In the method of the present invention, the ML detector is optimized in terms of noise correlation generated in the partial response channel. Improved ML detectors offer comparable or superior performance to Viterbi detectors in the presence of colored noise. In the present invention, a set of finite impulse response (FIR) cross filters are used as the ML estimator for the predictive detector. The weighted sum outputs of the ML detectors are compared to a set of thresholds based on previously detected data to make a decision on the current detection. The present invention improves the performance and reduces the complexity of the ML detector by optimizing the coefficients of the FIR filter in the presence of correlated or colored noise. Based on the noise correlation of the channel at a given user density, the SNR (signal-to-performance ratio) of each FIR filter is determined over a range of coefficients, and the coefficient providing the best SNR is determined by the respective decision function ( decision function). The result is a less complex noise whitening ML detector that provides improved performance over prior art ML detectors.