Massive MIMO Signal Detection Using Cascaded OQRD-PMD Method

The minimum mean square error (MMSE) is adopted widely as a signal detection technique for massive MIMO (mMIMO) systems; however, it needs matrix inversion with complexity of cubic or-der. Therefore, several signal detection techniques were adopted, including successive over relaxation, Richardson, Jacobi, and Gauss Seidel (GS) to avoid matrix inversion. Although these techniques offer lower complexity, their performance remains inferior to MMSE. Therefore, in this paper, we propose four signal detection techniques namely QR decomposition (QRD), QRD-parallel multi-stage detection (QRD-PMD), ordered QRD (OQRD), and OQRD-PMD. In this work, at first, the QR decom-position is applied to the channel matrix followed by backward substitution method to obtain the transmitted signal. After that, the ordered QRD (OQRD) method is applied to avoid the error propagation problem arises in QRD method. Furthermore, the parallel multistage detection (PMD) method is cascaded with QRD and OQRD to estimate more accurate transmitted signal at the receiver side. The performances of conventional and proposed signal detection techniques are evaluated based on bit error rate (BER) and complexity. The simulation results show that the proposed techniques significant achieve superior performance as compared to MMSE while maintaining comparable complexity.