Requirements on radiating structures are constantly increasing, demand for faster speed, smaller size, and higher reliability drives today's technical development. However, for electrically small structures, less than half-a-wavelength in size, performance is fundamentally limited by physical size. This thesis explores how to construct and calculate physical bounds for advanced antennas and complex environments that are in use in modern communication today. Previously, physical bounds have mainly been formulated for single feed, single resonance antennas in free space. However, in modern communication settings antennas are much more advanced. In all cellular networks after the third generation of mobile networks (3G) multiple input multiple output (MIMO) systems are being utilized, where antennas have multiple feeds. Formulating physical bounds for these antennas is not trivial due to classically limited performance metrics, such as the Q-factor, being difficult to define or calculate. It is not only the antennas themselves that are more advanced, antennas are also used in implants, medical devices, meta materials, and in plasmonics. Calculating physical bounds in these scenarios require new methods that reliably predict accurate results for all different types of materials.
In this thesis a method for constructing physical bounds for general MIMO antennas is presented. By idealizing the channel and representing the antenna by the equivalent currents excited across it, a bound can be calculated with convex current optimization. It is shown that that bound is effectively reached by exciting different sets of modes depending on what constraints are put on the optimization. Different shapes and sub-regions are analyzed using the strength of these modes.
A new method for calculating stored energy and Q-factor in the presence of complex media is presented and investigated. By viewing the antenna as a dynamic system the method of moments (MoM) impedance equation can be formulated as a state space model. The energy stored within in such a model is identified as the stored energy. This method can be generalized to dispersive and inhomogeneous media.
Place: Lecture Hall E:1406, E-Building, Ole Römers väg 3, Lund University, Faculty of Engineering LTH
Name: Migliore, Marco Donald
Affiliation: University of Cassino and Southern Lazio, Italy
- Annan elektroteknik och elektronik