A Generalized Methodology for Obtaining Antenna Array Surface Current Distributions With Optimum Cross-Correlation Performance for MIMO and Spatial Diversity Applications

Based on a new formulation of far field cross-correlation involving the currents on the radiating sources, we propose a general methodology employing a Genetic Algorithm (GA) to find optimum distributions of current amplitudes and phases on MIMO antennas such that the resulting system has good cross-correlation (high diversity gain.) The obtained currents can help guide the design and fabrication process of final MIMO antennas by providing valuable information about which current distributions can achieve the best “complementarity” of the individual far fields such that the total diversity gain is maximized. Moreover, this approach will help to explicate what is meant by a `MIMO antenna' from the electromagnetic viewpoint by defining a MIMO antenna as an antenna supporting the optimum currents such as those obtained by the proposed method itself. The method is quite general and can be applied to arbitrary antenna types and array topologies. Verifications for examples comprised of small arrays of half-wavelength dipoles are provided and the practical significance of the results is discussed.