Hard and Soft Spherical-Bound Stack decoder for MIMO systems

Classical ML decoders of MIMO systems like the sphere decoder, the Schnorr-Euchner algorithm, the Fano and the stack decoders suffer of high complexity for high number of antennas and large constellation sizes. We propose in this paper a novel sequential algorithm which combines the stack algorithm search strategy and the sphere decoder search region. The proposed decoder that we call the Spherical-BoundStack decoder (SB-Stack) can then be used to resolve lattice and large size constellations decoding with a reduced complexity compared to the classical ML decoders. The SB-Stack decoder will be further extended to support soft-output detection over linear channels. It will be shown that the soft SB-Stack decoder outperforms other MIMO soft decoders in term of performance and complexity. INTRODUCTION In this work, we are interested in the decoding of multi-antenna systems using spatial multiplexing [1] and linear space time block codes (STBC). In [2], a lattice representation of MIMO systems is proposed. The decoding problem can then be seen as a closest lattice point search problem (CLPS) and the lattice decoders are then used. For MIMO systems, we find in the literature mainly two classes of decoding strategies. On one hand, there are the optimal decoders (ML) such as the sphere decoder and the Schnorr-Euchner algorithm, which have a complexity that increases drastically with the lattice dimension and the constellation size. In the other hand, there are the sub-optimal decoders like the ZF, ZF-DFE, Ghaya Rekaya Ben-Othman, Rym Ouertani and Abdellatif Salah are with TELECOM ParisTech, 46 rue Barrault, 75013 Paris FRANCE. Phone: +33 1 45 81 76 33 / + 33 1 45 81 78 40, Fax: +33 1 45 80 40 36, Emails: rekaya, ouertani, [email protected] November 6, 2008 DRAFT