Design of low-complexity linear dispersion codes

We consider the design of space-time codes that offer low decoding complexity. It is well known that for codes based on orthogonal designs, maximum-likelihood (ML) decoding reduces to symbol-by-symbol decoding. However, this simplicity of decoding is achieved at the expense of symbol rate. It is natural to ask if the symbol rate can be increased if the code construction allows ML decoding to reduce to pairwise decoding of symbols. Our work attempts to answer this question by describing a method of designing pairwise decodable codes for any number of transmit antennas based on the linear dispersion (LD) code framework. We then show that for codes designed using this method, ML decoding can be reduced to pairwise decoding of symbols. For the case of M=3 and M=4 transmit antennas, we provide error-performance simulation results for these codes.