Performance study of space-time communications systems based on the VBLAST algorithm

Capacity in wireless channels has become the prime aspect in mobile communications. Significant improvements are possible with MIMO channels, with the use of multiple antennas at both the transmitter and receiver. Reception in this type of channel is achieved with space-time coding by jointly designing channel coding, modulation and equalization, allowing an important increase in the throughput of bandlimited wireless channels. One of the most important space-time coding systems is the Bell-Labs layered space-time (BLAST) coding technique. Different variants of this algorithm exist, focusing on computational complexity reduction. One of these variants is the vertical BLAST algorithm which has different realizations depending on the way the signal processing at the receiver is performed. In this paper, an analysis of these realizations is presented from a performance-complexity standpoint. These variants are the singular-value decomposition, the sorted-QR decomposition and the least-square algorithm. It is shown that the latter presents the best performance/complexity trade-off.