A Low Computational Complexity for 2D MUSIC Algorithm in Massive MIMO Systems

Abstract

A massive Multiple-Input Multiple-Output (mMIMO) systems are one of the primary techniques in 5G. It utilizes hundreds and even thousands of antennas collected in one panel. This increase in the number of antennas leads to an increase the computational complexity of the direction-of-arrival (DOA) algorithms. In this paper, two steps propose to reduce the computational complexity of two-dimensional multiple signal classification (2D MUSIC) in mMIMO systems. The first step is reducing the dimensional of the data covariance matrix by determining the optimum matrix compression factor. The second step is searching for the optimum number of noise eigenvectors used to obtain the 2D MUSIC spectrum, a uniform circular array (UCA) used as the antenna array. The simulation results indicate that the covariance matrix can be compressed two consecutive times without affecting the performance accuracy and resolution of the 2D MUSIC algorithm. Moreover, the optimum number of noise eigenvectors used in the 2D MUSIC algorithm is close to the number of signal sources.