On the Covariance Matrix and Capacity Evaluation of Reconfigurable Antenna Array Systems

In this paper, we derive analytical expressions to compute the covariance matrix of the signals impinging on a reconfigurable antenna array. We consider a receiver equipped with a linear antenna array where each antenna element can be independently configured to create a directive radiation pattern toward a selected direction. In our derivation, we assume a multimodal truncated Laplacian distribution to model the power angular spectrum (PAS) of the signals arriving at the array. Using the derived expressions, we are able to evaluate the correlation between the received signals as a function of the signal spatial distribution, the antenna array topology, and the radiation pattern characteristics of each element in the array. We also study the capacity of a reconfigurable multiple-input multiple-output (RE-MIMO) system and illustrate its relation to radiation pattern configurations and characteristics. We demonstrate how the derived expressions can be used to efficiently choose the configuration for each reconfigurable antenna element in the array that leads to an optimal compromise between low receive correlation and high receive power in order to maximize the link capacity. Simulation results are provided to validate our analytical expressions.