AN EFFICIENT PARADIGM FOR EVALUATING THE CHANNEL CAPACITY OF CLOSED-LOOP MASSIVE MIMO SYSTEMS

A particular challenge encountered in designing massive MIMO systems is how to handle the enormous computational demands and complexity which necessitates developing a new highly efficient and accurate approach. Considering the large antenna array employed in the Base-Station (BS), in this work, we present a new paradigm to significantly reduce the simulation runtime and improve the computational efficiency of the combined rigorous simulations of the antenna array, 3-D channel model, and radiation patterns of the User Equipment (UE). We present an approach for evaluating a closed-loop massive MIMO channel capacity using 3-D beamforming to take advantage of spatial resources. The approach subdivides an M × N array at the BS into columns, rows, rectangular, or square subarrays, each consisting of a sub-group of antenna elements. The coupling is rigorously taken into account within each subarray; however, it is ignored among the subarrays. Results are demonstrated for a dual-polarized microstrip array with 128 ports. We consider simulation runtimes with respect to two different propagation environments and two different Signal-to-Noise-Ratios (SNRs). It is shown that the maximum difference in the closed-loop capacity evaluated using rigorous electromagnetic simulations and our proposed approach is 2.4% using the 2×(8×4) approach for both the 3-D Channel Model in the 3rd Generation Partnership Project (3GPP/3D) and the 3-D model in the independent and identically distributed (i.i.d/3D) model with a 46% reductional in computational resources compared with the full-wave antenna array modeling approach.