A Low-Complexity Detection Framework for Signed Quadrature Spatial Modulation Based on Approximated MMSE Sparse Detectors

The design of low-complexity data detection techniques for massive multiple-input multiple-output (mMIMO) systems continues to attract considerable industry and research attention due to the critical need to achieve the right tradeoff between complexity and performance, especially with the signed quadrature spatial modulation (SQSM) scheme. However, the SQSM scheme attains a high spectral efficiency and good performance but suffers from a high computational complexity with mMIMO systems. In this article, we propose an efficient low-complexity detection framework for the SQSM scheme. Sparsity detection is amalgamated in this article with minimum mean-square error (MMSE) detector by decoupling the detection of the real and imaginary vector streams. Unfortunately, the MMSE-based detector has a matrix inversion which incurs a high computational complexity. Therefore, we employed several iterative methods; i.e., conjugate gradient and Gauss–Seidel, to avoid the exact matrix inversion, and hence, the computational complexity is significantly reduced. Moreover, the proposed framework can host other iterative methods such as the JA, successive over relaxation, accelerated over relaxation, Neumann series, Newton iteration, two-parameter over relaxation, and Richardson methods. The proposed detection framework attains a significant complexity reduction with a small or insignificant deterioration in the performance.