Coding for multiple antennas with linear and nonlinear suboptimum interface

Summary form only given. Transmission and reception with multiple antennas is known to provide potentially enormous capacities. However, the receiver complexity may become impractical as the number of antennas increases above a certain limit; thus, it is interesting to derive suboptimum interfaces with limited complexity. Here we compare suboptimum linear and nonlinear interfaces to be used for decoding space-time codes transmitted over a multiple antenna Rayleigh fading channel, with perfect channel-state information available at the receiver. The performance of Vertical BLAST and that of linear receiver interfaces are examined with codes obtained by apportioning evenly among the transmit antennas the symbols of off-the-shelf convolutional codes. We first observe how the introduction of an interleaver can be beneficial here. Next, we introduce a new simple iterative linear interface, based on hard Viterbi decoding and offering a performance considerably improved with respect to noniterative receivers. Numerical results show how a performance close to optimum (maximum-likelihood) detection can be achieved with simpler receiver interfaces.