Deep Unfolding of Chebyshev Accelerated Iterative method for Massive MIMO Detection

The zero-forcing (ZF) and minimum mean square error (MMSE) based detectors can approach optimal performance in the uplink of massive multiple-input multiple-output (MIMO) systems. However, they require inverting a matrix whose complexity is cubic relation to dimension. This lead high computational effort, especially MIMO To mitigate this, several iterative methods have been proposed literature. In this paper, we consider accelerated Chebyshev SOR (AC-SOR) AOR (AC-AOR) algorithms, which improve detection conventional Successive Over-Relaxation (SOR) Accelerated (AOR) methods, respectively. Additionally, propose using deep unfolding network (DUN) optimize parameters AC-SOR AC-AOR leading AC-AORNet AC-SORNet DUN-based method leads significant improvements compared for various channels. results demonstrate that are effective, outperforming other state-of-the-art algorithms. Furthermore, highly particularly high-order modulations such as 256-QAM (Quadrature Amplitude Modulation). Moreover, almost same number computations respectively, since use has negligible impact on system’s complexity. DUN features fast stable training scheme due its smaller trainable parameters.