Electromagnetic-based signal processing for arbitrary arrays.

We recently proposed an electromagnetic manifold that characterizes arrays through the antenna current distributions and can be leveraged to model arbitrary antennas in MIMO systems. The manifold represents arrays with many Hertzian dipoles, yielding an analytically tractable representation that accounts for nonideal effects such as mutual coupling, polarization, near-field effects, and antenna patterns. We use this electromagnetic manifold to model a transceiver using arbitrary arrays. We highlight how array properties, such as coupling and polarization, affect signal processing at both the transmitter and receiver. We also demonstrate how the proposed model can be applied to diverse arrays with varied elements. Our simulation results in a realistic wireless environment show how the proposed transmission methods achieve higher spectral efficiency than approaches that assume ideal conditions.