Turbo decision aided equalization for magnetic recording channels

Applying turbo codes to intersymbol interference (ISI) channels requires a soft a posteriori probability (APP) detector for channel inputs. The complexity of conventional APP detectors grows exponentially with the channel memory length. In this paper we present a new APP module for binary ISI channels based on decision aided equalization (DAE), whose complexity grows linearly with the channel memory length. The reduction of channel detection complexity reaches 95% for a five-tap ISI channel. Simulations performed on the MEEPR4 channel show close to optimum performance for this turbo-DAE structure.