现有的天线选择算法大多都是假设信道是独立同分布的,而实际的信道是相关的。针对这一问题,本文在研究盖尔圆算法的基础上,提出了一种新的用于相关信道的低复杂度天线选择算法。算法利用奇异值估计方法给出最小奇异值的下界,通过递增的方法,选择使信道矩阵最小奇异值的下界最大的列从而提高最小奇异值,使得容量最大化并得到较好的误码性能。理论分析与仿真结果表明,该算法在射频链路较少时具有很低的复杂度,同时容量和误码性能优于随机选择算法,接近最优选择算法。
The MIMO channels are correlated in real propagation environment,they are assumed to be independent and identically distributed in the existing schemes.Focusing on this problem,a new antenna selection algorithm with low complexity is presented.The proposed algorithm uses the lower bound of the minimum singular value given by the singular value estimation method,and chooses the column whose lower bound of the minimum singular value is maximum to increase the minimum singular value,so the capacity is maximized and the performance of BER is better.The theoretical analysis and simulation results indicate that both capacity and BER performance of the proposed algorithm are better than that of the random selection algorithm and close to the optimal algorithm.The complexity is very low when the selected antennas are small.