针对多输入多输出空间相关信道环境,提出了一种基于格规约理论的最小均方误差检测算法.该算法考虑了空间相关性的影响,利用复数Lenstra-Lenstra-Lovász格规约算法克服了传统的线性最小均方误差检测器分集阶数随相关系数增加而减小的缺点,使检测到的信号在高信噪比下可以达到比传统算法低个数量级的误码率.理论分析与仿真结果表明,在信道为Kronecker相关平坦衰落模型下,接收端采用最小均方误差信道估计,发射端采用QPSK调制未使用信道编码,当比特信噪比大于18dB时,基于格规约理论的最小均方误差检测器的误码率低于传统最小均方误差检测器的误码率 发射端采用(2,1,3)卷积码、接收端采用硬判决Viterbi译码,当信噪比大于16dB时,基于格规约理论的最小均方误差检测器的误码率低于传统最小均方误差检测器的误码率,且基于格规约理论的最小均方误差检测器的分集阶数不受相关系数的影响,等于接收天线数N.-Toovercomethedrawbackthatthediversityorderofthetraditionallinear MMSEdetection

algorithmdecreaseswiththeincreaseofthecorrelationcoefficient,alattice-reduction-aidedMMSE(LRAMMSE)

detectionalgorithmisproposedforspatialcorrelatedMIMOchannels.TheLRA-MMSEdetection

algorithmconsiderstheeffectofspatialcorrelationinthealgorithm designandtakesadvantageofthe

complexLenstra-Lenstra-Lovász(LLL)algorithmsothatitsBER(biterrorrate)isseveralordersof

magnitudesmallerthanthatofthetraditionallinearMMSEdetectionatahighSNR(signaltonoiseratio).

Theoreticalanalysisandsimulationresultsshowthatwith QPSK modulationandthe MMSEchannel

estimatoratthereceiver,theuncodedBERoftheLRA-MMSEdetectorislowerthanthatofthetraditional

linearMMSEdetectorifthebitSNR(Eb/N0)isgreaterthan18dB,whilethecodedBERoftheLRAMMSEdetectorislowerthanthatofthetraditionallinearMMSEdetectorifEb/

N0isgreaterthan16dB

when(2,1,3)convolutionalchannelcodingandViterbidecodingareadoptedundertheKroneckercorrelated

flatfadingchannel.Furthermo