Prefiltered Low Complexity Tree Detection for Frequency Selective Fading Channels

We target M-ary data sequence estimation over time-variant frequency selective fading channels. Recently, we proposed a low complexity breadth first tree detector, termed improved M-algorithm (IMA). Unlike the M-algorithm (MA), it takes the whole energy of the received symbol of the current step into account when choosing survivors. Unknown future symbols overlap with postcursors of the current symbol and are modelled as additive Gaussian noise in the metric. Thus, unlike the standard MA, IMA works well without an energy compacting front end prefilter (FEP) even in frequency-selective channels, whereas a FEP is crucial for ordinary reduced state sequence estimators such as the delayed decision feedback sequence estimator (DDFSE) and reduced state sequence estimator (RSSE). However, although not explicitly needed, we show that the complexity of IMA can be reduced even more applying a FEP, because the first channel tap has increased energy, and the effective channel would be even shorter, so that the influence of future symbols on the postcursors is smaller, enabling the tree detector to chose the paths through the tree more precisely. We propose to make use of the cepstrum to compute the FEP via a minimum phase target impulse response. We use the GSM/EDGE modulation format here, in order to enable other researchers to compare their results. Simulation results indicate that for the given application, IMA is superior to RSSE with similar complexity.