Impact of different finite MIMO array geometries on system throughput with considering mutual coupling and edge effect between array elements

In this paper, the antenna array appropriate topologies and configurations are investigated to reduce the impact of large gain variation between array elements that is produced due to the mutual coupling. Moreover, the most appropriate antenna array topology that can realize a maximum system throughput and achieve fairness between users in terms of their spectral efficiency is determined. Our analysis focuses on the comparison between various array geometries (uniform planar array (UPA), circular array (CA), conformal array (CfA), uniform planar circular array (UPCA), centric planar circular array (CPCA) and, finally, conical array (CoA)) with half-wavelength dipole elements. The simulation results show that UPA, CA, and CfA have total efficiency around 96% at 2.6GHz. Moreover, all the proposed geometries have achieved Envelope Correlation Coefficient (ECC) less than 0.5, and high Diversity Gain (DG) that makes it suitable for MIMO. The CfA geometry has been elected as the most suitable topology, which can provide minimum gain variation, maximum system throughput, and achieve fairness between users’ spectral efficiency. Finally, curl antenna is considered in the designed antenna array to study the effect of antenna gain pattern. Two array geometries have been designed using the curl antenna, which are the UPA and the CfA. The CfA curl has insertion loss -30dB between the edge elements and achieve fairness index between the users up to 0.98733. Moreover, the results show that curl antenna geometries can achieve higher system throughput compared to half-wavelength dipole based array geometries.