Iterative tree search detection for MIMO wireless systems

This paper presents a reduced-complexity detection scheme, called iterative tree search (ITS) detection, with application in iterative receivers for multiple-input multiple-output (MIMO) wireless communication systems. In contrast to the optimum maximum a posteriori (MAP) detector, which performs an exhaustive search over the complete set of possible transmitted symbol vectors, the aim of the new scheme is to evaluate only the symbol vectors that contribute significantly to the soft output of the detector. To this end, a list of "good" candidate symbol vectors is generated prior to the actual computation of the detector output, with the aid of a sequential tree searching scheme based on the M-algorithm. For high-order QAM modulation formats, the complexity of the ITS detector can be further reduced with the aid of a special type of bit mapping called multi-level mapping. This results in a complexity per bit that is linear in the number of transmit antennas and roughly independent of the modulation order. Results from computer simulations are presented which demonstrate the good performance of the new scheme over a quasi-static Rayleigh fading channel, even for relatively small list sizes.