A novel adaptive receiver with enhanced channel tracking capability for TDMA-based mobile radio communications

The paper presents a novel fast-adaptive nonlinear receiver which exploits soft statistics for tracking the random fluctuations experienced by time division multiple access (TDMA) mobile radio links impaired by frequency-selective time-variant multipath phenomena. The detection task is accomplished by an Abend-Fritchman-like symbol-by-symbol maximum likelihood (SbS-ML) detector which delivers both hard decisions and soft statistics in form of a posteriori probabilities (APPs) of the states of the intersymbol interference (ISI) channel. In the proposed adaptive receiver, these APPs are employed in place of the conventional hard-detected data to feed an ad hoc developed nonlinear recursive Kalman-type channel estimator. Extensive computer simulations show that the exploitation of soft statistics enhances the tracking capability of the channel estimator so that the proposed receiver generally outperforms the usual ones based on adaptive maximum likelihood sequence estimators (MLSEs) for signal-to-noise ratio (SNR) values over 12-13 dB. Furthermore, the experienced performance gap with respect to more complex per-survivor processing (PSP)-based multi-estimator detectors appears generally small on slowly and moderately fast time-varying channels characterized by values of the product Doppler bandwidth × signaling period B_DT_S below 5×10^-3