Low-complex Bayesian estimator for imperfect channels in massive muti-input multi-output system

<span lang="EN-US">Motivated by the fact that complexity of computations is one main challenges in large multiple input output systems, known as massive multiple-input multiple-output (MIMO) this article proposes a low-complex minimum mean squared error (MMSE) Bayesian channel estimator for uplink channels such systems. First, we have discussed necessity covariance information MMSE and how their imperfection knowledge can affect its accuracy. Then, two reduction phases dimension floating-point operations been suggested to reduce complexity: phase 1, eigenstructure matrices implemented based on some truncation rules, while 2, arithmetic matrix multiplications equation followed. The proposed procedure has significantly reduced first order O(M), which less than required conventional with O(M<sup>3</sup>) terms dimension. It shown estimated using our are asymptotically aligned serve same quality full-rank channels. Our results validated averaging normalized (NMSE) over length 500 sample realizations through Monte Carlo simulation MATLAB R2020a.</span>