Efficient adaptive equalization of doubly dispersive channels in MIMO-FBMC/OQAM systems

The problem of adaptively equalizing doubly dispersive MIMO channels for FBMC/OQAM systems is studied in this paper. The challenges in this type of multicarrier systems include their intrinsic self-interference and the need to cope with time- and frequency-selective subchannels in realistic propagation conditions. An efficient and numerically stable algorithm is adopted, relying on a decision feedback structure that implements BLAST ordering for the input signals recovery. The ability of this algorithm to address the above challenges has been demonstrated. The focus of this paper is on reducing the needs of this equalizer in training information. A channel estimate-based (re-)initialization scheme is developed and shown to be quite effective in lowering the training overhead, at an affordable additional cost in complexity. For the sake of comparison, the MIMO-OFDM problem is also studied. Simulation results for practical scenarios demonstrate the effectiveness of the proposed approach.