基于QR分解的V-BLAST迭代检测算法

QR分解检测算法的一种运算量较低的V-BLAST检测算法,但其性能较差。第一层判决输出的性能对整体性能的影响很大。除了初次迭代外,提出的算法用前次迭代中性能最好的层作为迭代开始,进而寻求性能更好的数据层,这样经过几次迭代循环,可以做出性能很高的第一次判决输出。由于第一次判决输出性能的提高,总体性能也得到很大提高。仿真结果显示,对于4发送4接收MIMO系统,在误符号率为0.1%时提出算法的增益较排续连续干扰抵消V-BLAST检测算法高6dB.而其算法复杂度是排序连续干扰抵消的30%。

Iterative Detection Algorithm for V-BLAST Based on QR Decomposition

QR decomposition detection for V-BLAST is a low computational cost algorithm and the performance is not ideal. The first detection step is crucial for the overall performance. Except for the initialization iteration ,the proposed iterative detection begins at the substream that performance is best in the previous iterative detection .Using this method a few times of loop the first decision is made whose performance is good enough. Since the performance of the first decision is excellent, the overall performance is improved greatly. For the multiple-input multiple-output (MIMO)system with 4 transmitted antennas and 4 received antennas ,simulation result shows that there is more than 6 d B gain by comparing with the traditional V-BLAST optimal order detection algorithm (OPT)at the bit error rate(BER) of 0.1% .but the proposed algorithm is 3 times less complex than the V-BLAST OPT.