Iterative decoding of space-time error correcting codes
Résumé
Most of multiple-input multiple-output (MIMO) coding schemes can be sorted into two families. The first one includes the space-time block codes (STBCs) and in particular, the orthogonal STBCs (OSTBCs) as they can be optimally decoded with a low complexity. They improve the transmission quality, the communication range or the receiver robustness. The second family is based on the layered space-time (LST) architecture and aims at increasing the data rate without bandwidth nor power extra cost. To guarantee a given transmission quality, the use of a forward error correcting code (FEC) is mandatory like in the space-time bit interleaved coded modulation (ST-BICM) case. The space-time error correcting codes (STECCs) we introduced, which are based on any linear FEC, appear to be a good compromise between both families. They achieve better performance than the usual Alamouti code associated with the same FEC and propose higher data-rate with space-time puncturing techniques. However, the bottleneck of these codes is the detection stage which complexity grows exponentially with the number of FEC codewords involved in the creation of one STECC codeword. By applying the turbo equalization principle, we can significantly reduce this complexity, while achieving equivalent level of performance.