Unconventional Architectures and Methodologies for Antenna Array Synthesis

Phased array (PA) antennas represent a key enabling technology for current and next-generation wireless systems, yet their large-scale deployment is often discouraged by the cost and complexity associated with fully populated architectures. Within this context, this thesis investigates unconventional PA architectures to explicitly quantify and control the trade-off between achievable radiation performance and the reduction of hardware complexity imposed by practical feeding constraints. On this basis, a set of optimization-driven synthesis methodologies is developed to address the design of unconventional arrays across different formulation domains. The proposed approaches are conceived to remain computationally viable under realistic operational requirements, while explicitly accounting for the interaction between array architectural constraints, and the resulting radiation characteristics. Within the presented framework both classical power pattern-oriented objectives, and criteria arising from application-driven requirements are considered. Numerical results obtained with realistic element models and scenarios demonstrate the effectiveness of the proposed approaches in controlling sidelobes, supporting beam steering, preserving pattern stability under scanning, or providing quality-of-service, while significantly reducing the number of active modules and the overall PA architectural complex- ity. The outcomes of this work indicate that unconventional PAs, when supported by a computationally tractable and constraint-aware synthesis methodology, are a concrete and technically sound solution for scalable PA architectures under practical design constraints.