Design and Execution of an MMSE MIMO Detector for High-Speed Wireless Communication Using a Low-Complexity Neumann-Series
DOI:
https://doi.org/10.62643/Abstract
In order to increase transmission reliability and achieve high data rates, current wireless communication systems frequently employ Multiple-Input Multiple-Output Orthogonal Frequency Division Multiplexing, or MIMO-OFDM. However, high performance MIMO detection hardware implementation is computationally demanding, especially when it comes to matrix inversion operations. A traditional 4x4 MIMOOFDM receiver may use a pipelined matrix-inversion design and a Minimum Mean Square Error (MMSE)- based detection method. Despite the potential for minimal processing delay, matrix inversion necessitates substantial mathematical resources. A low-complexity Neumannseries-based MMSE MIMO detector is presented in this study. The suggested approach uses a small number of Neumann-series terms to approximate the inverse of the MMSE matrix, substituting iterative matrix multiplication and addition for direct matrix inversion. To meet the approximation's convergence condition, a matrixnormalization step is included. The architecture includes symbol estimation, output decoding, iterative matrix computation, MMSE matrix generation, matrix normalisation, and Neumann-series inverse approximation. The design employs pipelined processing to provide effective hardware operation while maintaining a 4x4 MIMO arrangement. The architecture is written in Verilog HDL and is meant to be used with Xilinx Vivado for functional simulation and synthesis evaluation. Performance is evaluated in terms of processing latency, power consumption, hardware utilisation, and symboldetection accuracy. With respect to MMSE detection performance, the method seeks to minimise matrixinversion complexity.
Downloads
Published
Issue
Section
License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.













