Different constellations mapping between layers in V-BLAST architecture

The BLAST architecture is a promising scheme to gain great throughput. In this paper, we propose a new scheme mapping different constellations between different layers of the V-BLAST schemes, which releases the demand to hardware compared to power allocation schemes and improves the performance greatly compared to the conventional mapping scheme. We also derive the optimal constellation mapping in theory. The simulation results show that the proposed scheme achieves both the reduction of computing of ordering and enhancement of throughput. For example that one scheme (the “6421” scheme) can have a performance gain 10dB at BER=10⁻⁴ compared to 16QAM modulation at expense of just 3 bit per symbols slot (bpss) throughput decreased. Compared to QPSK modulation, there is not only 1 dB gain at BER=10⁻⁴, but also 5 bpss throughput enhancement.