Noise predictive belief propagation

We introduce iterative noise whitening for belief propagation (BP) based channel detectors over intersymbol interference (ISI) channels with correlated noise. Called noise predictive belief propagation (NPBP), the new scheme iteratively whitens the noise samples by modifying the edge probability computation of the BP algorithm. NPBP detectors based on finite impulse response (FIR) and infinite impulse response (IIR) prediction filters are introduced. In addition, we propose a novel prediction filter optimization method leading to a better noise whitening performance. Simulation results for both coded and uncoded systems and comparisons with maximum a posteriori (MAP) and BP detectors show that, significant improvements can be obtained.